Does Anybody Use Point Buy?

[KOLE]

Sort of an informal survey, I’ve just started playing with a new group which is exciting as most of 2018 I was stuck as the “Forever DM”. Of this group, 5 (including myself) DM for other groups outside of the current campaign. When starting, we rolled for stats, which I certainly don’t mind. We were talking about different styles and I found I was the only one in the group who institutes point buy for campaigns.

One of my campaigns last year, which never got past session 0, I had a LOT of push back for using point buy. Having lurked these forums for a while, it seems the grand majority of people roll for stats.

Now, I don’t have anything AGAINST rolling for stats, but I personally feel point buy is more interesting. You have to make hard choices to build your character. Almost every character ive seen has rolled a good array of stats, to the point that nothing is lackluster. This seems a bit dull to me. Am I the only one in 5e that actually implements point but at their table?

[Jama7301]

In 5e, I prefer point buy myself.

[Coffee_Dragon]

Always, religiously. Also non-random HP.

[Sigreid]

My table rolls, though if a player wanted to do point buy or standard array I dont think anyone would mind.

[Contrast]

I've only once not used point buy (DM instruction) and would typically discourage rolling.

[MaxWilson]

Nope. I allow it as a fallback option if you don't like your rolls, but do not mandate it.

[Rukelnikov]

I've never used point buy or standard array for PCs in any edition.

I'm tempted to do it in my next campaign though.

[DeTess]

I've done both, but I generally prefer point-buy. With point-buy, you're guaranteed a competent character, and the same doesn't work out with rolling. It can also create some annoying power disparities.

[Slybluedemon]

I love point buy, it's my go-to unless a DM requires rolling or array.

[GlenSmash!]

Always, religiously. Also non-random HP.

Same here for every 5e PC I've played.

I've stated up some with rolls for giggles, but never played in a campaign that allowed them.

[RSP]

Im fine with point buy, standard array or rolling, depending on the campaign and how powerful you want the PCs (in my experience rolling allows for a better main stat than SA or PB, so the PC usually has room for an extra Feat or stronger secondary stat).

I tend to prefer rolling, but that’s more because I like to work with random numbers in abilities, including RPing low stats.

[Man_Over_Game]

Always, religiously. Also non-random HP.

Agreed. There's no distinct power difference drawn between players. There's little risk of difficulty fluctuating between them. It rewards things like choosing an ASI over a feat, or racial benefits even more than they normally do, and it keeps players from trying to min-max.

It just solves so many problems. Rolling can provide some interesting team dynamics, where weaker members might rely more on powerful ones, but that's generally not a mechanic that most people enjoy.

[JNAProductions]

Usually point buy.

Sometimes generous rolling, if I want a higher powered game.

But usually point buy.

[DMThac0]

I love the random nature of rolling, and when I DM that's what I give my players at first.

However, if players want, I give them a pool of 72 points to distribute as they see fit. This is 12s across the board before racial mods, but it really makes a player think when they push numbers higher/lower.

[dejarnjc]

Have done rolling twice and point buy a dozen or more times. Point buy all the way.

[Rerem115]

I've left it up to the player; from a numbers standpoint (explained here (https://www.reddit.com/r/Pathfinder_RPG/comments/5owftt/closest_equivalent_point_buy_value_for_several/)), 4d6 drop 1 six times has an average value of about 20 points, so they have the choice between choosing the higher variance method, or the higher floor (but lower ceiling) offered by point buy.

I guess it says something that everybody except for one guy, a first-time player, rolled.

[Yora]

I always used point buy in 3rd edition. But with 4d6 drop lowest, arrange as desired, +3 for race, and +2 roughly every 4 levels, I think rolling is so generous that some randomness in starting stats is perfectly workable.

[Man_Over_Game]

Thinking about it, rolling 6d4, losing 2, will have some pretty balanced rolls compared to 4d6 lose 1.

[bc56]

Here's a bit of analysis.

Rolling:
Allows more variety by making nonstandard combinations easier to accomplish (with high rolls)
Higher ceiling but lower floor, leading to more varied stats.

Point buy:
Everyone has the same power, one player doesn't have a 14 as their best stat while another has both an 18 and a 17 (this actually happened to my players)


I generally use rolling, but allow players to choose any of the arrays rolled, so they can copy their neighbor's rolls if they do poorly.

[Unoriginal]

As a DM, I tell my players to use either point buy or standard array (which is the same, but some of my players don't like calculating the cost of stats ).

[Vorpalchicken]

What I most often have these days is roll 4d6, drop 1 but allow point build as a safety net. This allows for the possibility of exceptional characters but won't give anyone a terrible array. (Unless they want it.)

[Laserlight]

The campaign we just started is the first I have played or DM'd that used point buy. All the rest have been rolled, usually with a starting feat as well, sometimes "everyone rolls, you can use your own or someone else's".

One thing that irritates me about point buy is when I have a point left over. Sure, I could do CHA 9 or WIS 11, but there's no reason to and it feels like I'm getting shortchanged.

And of course if you roll well, you might feel more inclined to try an unusual race / class combination.

[thoroughlyS]

I use point buy at my table.

My friend uses rolls at theirs.

[MaxWilson]

Agreed. There's no distinct power difference drawn between players. There's little risk of difficulty fluctuating between them. It rewards things like choosing an ASI over a feat, or racial benefits even more than they normally do, and it keeps players from trying to min-max.

I don't see how this can be true. Point buy means that min-maxers will always be able to do things like dump Int and Cha in favor of combat stats (and then ignore their low Int/Cha while roleplaying) will always be more powerful than players who want to play smart and likable characters without breaking character. If you roll stats or use standard array, you can't steal points from one stat to spend on another, so there's a limit on how much min-maxing can impact play.

One of the things I most dislike about point buy, personally, is the tension it sets up between my powergaming instincts and my roleplaying instincts. Rolling stats doesn't present the same level of temptation, and I doubt I'm the only one who feels that way.

[VonKaiserstein]

Point buy is so... predictable.

I much prefer rolling for stats, because it helps define your character from the beginning.

The idea of knowing you have the ideal stat build feels crunchier and more gamey than anything from the previous editions ever was.

But then I feel the same way with fixed monster damage and hit points. I love the improbable moments in DND.

[ad_hoc]

yeah, point buy and I will never go back.

I find 5e doesn't work well with rolling.

It's not just the number disparity. It's that there is nowhere to go. Starting with a 20 makes ASIs underwhelming. If you use feats then they both become much more powerful and less interesting as there is no tradeoff. So everyone gets feats instead of a good number of people taking ASIs.

[Tvtyrant]

I just let my players pick their stats. Most players will pick lower stats in order to play a character they like, stats rolls and point buy turn character creation into a sub-game where more is better.

[J-H]

In 5th edition, I have used nothing but point buy. Thanks to the way 5e is balanced, every class is viable this way.

For 3.5, when I'm DMing, it's roll 4d6b3, and if you don't like that array, pick a point buy (28 or 32 point, depending on the game). MAD classes need the help in 3.5.

[Rukelnikov]

I don't see how this can be true. Point buy means that min-maxers will always be able to do things like dump Int and Cha in favor of combat stats (and then ignore their low Int/Cha while roleplaying) will always be more powerful than players who want to play smart and likable characters without breaking character. If you roll stats or use standard array, you can't steal points from one stat to spend on another, so there's a limit on how much min-maxing can impact play.

One of the things I most dislike about point buy, personally, is the tension it sets up between my powergaming instincts and my roleplaying instincts. Rolling stats doesn't present the same level of temptation, and I doubt I'm the only one who feels that way.

Yeah, I feel the same way, my biggest gripe with point buy is the tendency to create bots instead of characters.

[MaxWilson]

It's not just the number disparity. It's that there is nowhere to go. Starting with a 20 makes ASIs underwhelming. If you use feats then they both become much more powerful and less interesting as there is no tradeoff. So everyone gets feats instead of a good number of people taking ASIs.

That's not really connected to stat rolling. Even if you roll low or use standard array, feats are generally a better investment than ASIs anyway, especially early on.

[Angelalex242]

In Adventure League, there's no choice. Point buy or bust.

[Danielqueue1]

I've run it a few ways, when going for a more survivalist style game point-buy keeps it nice and clean. Especially at low levels. When I run a more high fantasy or higher level type campaign I find rolling more fun. But then again I've been mostly playing with people who enjoy playing out their characters flaws for fun so a 6 in something won't stop the fun.

[JoeJ]

Always rolling, but you can use the rolled array of anybody at the table, as long as you take the entire set (no mixing and matching).

[The Big Bear]

I much prefer point-buy all the way as I feel that it helps keep parity among players more than rolling.

There are really only two reasons that I see for wanting to roll stats:
1) It is more "realistic" that everyone has varying levels of overall ability.
2) It is gambling, and people like gambling.

I like point 1, but honestly, I would prefer my players to start with equal resources when creating a character and then they can choose what they want to do.

Point 2 is my real sticking point for rolling stats. People like gambling as much as they hate losing, which is why half of the tables that roll for stats have a fallback like point-buy, copying another player's rolls, or enforcing minimums.

[Jophiel]

Campaign I'm starting is point buy/array. I want to be able to start Session 0 with people's characters mostly done (they can hash out who plays what via online) and start talking about backgrounds, relationships, etc instead of having to watch people roll dice and decide what race/class they want. Or someone coming in who is just soooo lucky and rolled 18, 18, 18, 17, 15, 15

Point Buy just makes it easier and let's me know that we can start quickly.

[Chronos]

My group officially uses point buy (and average HP). The DMs might allow it if someone wanted to roll for either, but nobody's asked.

I actually prefer arrays over point buy, because of the min-maxed 8,8,8 to unimportant stats, but since the standard array is a valid standard point buy, I just go with that (I don't think anyone else has three dumpstats, either).

[Bartmanhomer]

I used point buy and dices as well.

[Lyracian]

I prefer point buy but some players prefer to roll.

My players can use either but have to stick with what they roll.

The party has a Wizard with Int 18 and Chr 4; Cleric with Wiz 16 and Int 5. Thief with Dx 17 and Str 7.
In a party of five only one was significantly better than using points buy. Most would be worse once you get to level 8 with 2 ASI to make your prime ability 20. When they levelled up 3 characters took fixed HP and two rolled (both got below average).

That said my daughter just rolled up a new character for a different game and got 16,16,14,14,12,10!

[Sigreid]

My group officially uses point buy (and average HP). The DMs might allow it if someone wanted to roll for either, but nobody's asked.

I actually prefer arrays over point buy, because of the min-maxed 8,8,8 to unimportant stats, but since the standard array is a valid standard point buy, I just go with that (I don't think anyone else has three dumpstats, either).

Your first paragraph is probably key. The table will have a culture to it that biases one of the methods and most people dont buck their culture.

[youtellatale]

I have used both point buy/array and rolling for stats. I much prefer point buy because you don't have one character outshining another. I used to be pro-rolling for stats but that has significantly changed since I started DMing more. Just my 2 cents, it creates balance. If tables share the best roll then why not just pick stats and go with that? I mean if tables are just going to roll for stats and then share the stat array then it seems easier to just let the players pick the stats that they use. Anyway, whatever works at your table and is fun should be just perfect, everyone else's opinions are just that - opinions. Do what works for you.

[Astofel]

I usually go with rolling for stats, but I'll usually allow rerolls if the player rolls below a certain threshold. I feel like point buy restricts options somewhat, if I want to play my charismatic charlatan monk then I need to sacrifice something to get that charisma. If I want to play a high elf strength paladin I can do that, but my starting stats would be lackluster compared to if I went something more conventional, like human or dragonborn. Plus, having high stats means more room to take feats that are cool, but might have been difficult to squeeze into the build with point buy.

I also like rolling for HP even though I know it's not the best idea, because I'm just crazy like that. I don't force my players to roll for it though.

[Monster Manuel]

Always rolling, but you can use the rolled array of anybody at the table, as long as you take the entire set (no mixing and matching).

There's an interesting idea I've never seen used before. You roll, but only once, and everyone at the table uses the same array of rolls, arranged however they want.

You get the randomnes of rolling (the potential for exceptionally good or bad characters), but you don't have a huge disparity between any given players.

Doesn't work for Adventurer's League play, but it's an interesting thing to try at a home game.

[Lyracian]

Out of curiosity I compared the stats of the characters in my game to the standard array.

Level 8: Human with two ASI's
20 / 14 / 14 / 12 / 10 / 8 - Standard Array
20 / 14 / 14 / 12 / 10 / 7
20 / 15 / 13/ 10 / 9 / 5
20 / 18 / 10 / 10 / 9 / 4
20 / 18 / 16 / 10 / 10 / 10

First two are worse off than standard. Third player is slightly better with there two main stats but overall less power. Last character is better.

[Willie the Duck]

I have no strong preference, as I feel other decisions-on-what-to-allow by the DM will have a greater effect on gameplay. However, I will make an anecdotal observation. Everyone I know who plays Adventure League, and many people who spend a lot of time analyzing the game online (such as people like us forum-dwellers)… particularly if they have strong opinions on balance and the like... prefer point buy. People who do none of those things, and just want to play the game (particularly if they started gaming pre-WotC) all seem to do 4D6best3, and don't seem to know why they are supposed to want to change.

[KyleG]

I played a game where we all rolled 4d6 dropped lowest and rerolled 1s in order. Then we pooled each of the ability scores and worked out what each player wanted from each pool. So someone wanting to be a dex character chose from the dex pool first then the person who wanted high str chose etc etc.
I might do something similar but do 4d4 reroll ones and give everyone a feat.

[Ganryu]

I use it religiously.

[ChampionWiggles]

I've had 2 primary DMs over the couple years that I've played 5e. One has exclusively used rolling and the other always uses standard array. As a DM, I prefer array/point buy. I allow my players to roll for HP or take average, but they can't take average after seeing the roll. Most are taking the average.

The thing that I dislike about rolling for stats, which has been said before, is that it creatures a power gap and unequal distribution of stats. I don't personally like that.

With standard array, everyone has the same scores to distribute as they please, but I also find this...confining in a sense. To the point that when I make a character, I kind of go on auto-pilot of what scores go where ("15 and 14 in my class prime stats. 8 in what I've chosen as my dump stat. 10, 12, and 13 are where it gets "varied")

With point buy, everyone has the same amount of resources, but how they're distributed is up to them. I just prefer that method

[Luccan]

I bounce back and forth on my preferred method. I like rolling conceptually, but I think if you're going to roll, you can't have too much stake in what you'll be playing. You could end up with stats that meet or exceed what you need, but you could also fall short of your goal (especially if playing with multiclassing rules). So, mostly I like it because it can help me determine a new character, rather than help me build one I've wanted for awhile.

Point-buy is reliable, lets you decide what you want to be (mostly), and creates equitable stats for all PCs. I sometimes say roll or Point-buy or roll and use Point-buy if you're really unhappy with the results, just so people who don't want to gamble on a concept can build what they want. Some people have rotten luck, so I find it helps.

I think in the games I've applied to for 5e on this forum I've seen a fairly even spread, but maybe that's just because I'm applying to the right games.

Now here's the question, does anyone do Standard Array?

[DeTess]

I don't see how this can be true. Point buy means that min-maxers will always be able to do things like dump Int and Cha in favor of combat stats (and then ignore their low Int/Cha while roleplaying) will always be more powerful than players who want to play smart and likable characters without breaking character. If you roll stats or use standard array, you can't steal points from one stat to spend on another, so there's a limit on how much min-maxing can impact play.


Yeah, but at least then the power disparity is the result of human decisions, rather than the dice randomly deciding that Pete was getting a really good array, and Maria was getting all 12s. Rolling is also no guarantee that the min-maxer won't minmax, and can make the difference even worse. I once played in a 3.5 game where I was the only martial, and rolled worse than the elite array (where everyone else had multiple 16+ rolls).

I think that if you want to do point-buy fairly, you need to do some variant on 'everyone gets to pick any array to use', so no one gets left behind by the dice.

[MaxWilson]

I prefer point buy but some players prefer to roll.

My players can use either but have to stick with what they roll.

The party has a Wizard with Int 18 and Chr 4; Cleric with Wiz 16 and Int 5. Thief with Dx 17 and Str 7.
In a party of five only one was significantly better than using points buy. Most would be worse once you get to level 8 with 2 ASI to make your prime ability 20. When they levelled up 3 characters took fixed HP and two rolled (both got below average).

That said my daughter just rolled up a new character for a different game and got 16,16,14,14,12,10!

It's worth noting that unless she goes for a MAD class like a monk, in practice there's not much combat-relevant difference between this array and a more typical rolled array such as 12, 14, 16, 9, 7, 10. The biggest difference is on the roleplaying side: she can be e.g. a smart strong and sturdy paladin instead of just a strong and sturdy paladin.

It would be boring if all paladins were smart as well as strong and sturdy, and that is one of the other things I like about rolling: the variety of characters that result from it. Anecdotally, I find that there's a negative correlation between how much you like rolling and how long-lived you expect characters to be. If you expect to see four or five PCs from each player over the course of the next twelve months, you and they will probably like rolling more than someone who expects to see one PC from each player every three or four years.

[Captain Panda]

I tend to require rolling for stats when I DM, and I've played in campaigns with both point buy and rolling both.

I prefer there being some randomness to a character, and find point buy characters to all be very similar. There really is a 'right' and 'wrong' way to build, using point buy. With rolls you might have to dump something super hard and have a big weakness that comes out in roleplay, and you might get strong stats. The variability and randomness, in my experience, is fun. Even having a character who has poor stats isn't the end of the world, I've seen players who got crap stats make it work and got to level 20 with them, and they loved their characters.

[Blood of Gaea]

Stats rolling is fine, but if you do so everyone should use the same set of rolls. Have everyone at the table roll once each, then have the DM (or a player) roll up to a total of 6 stats. Everyone at the table makes their PC up with that.

[MaxWilson]

Does anyone on this thread do 4d6k3 in order or 3d6 in order? (I've done it occasionally for NPCs or one-shot PCs and it's fun, but because it's not the default for 5E I usually don't.)

Just for fun, what would you do with the following stat rolls? Feats and multiclassing allowed but no stat swapping.

#1: Str 17, Dex 16, Con 9, Int 14, Wis 14, Cha 4

#2: Str 10, Dex 11, Con 13, Int 16, Wis 16, Cha 16

#3: Str 11, Dex 10, Con 14, Int 12, Wis 17, Cha 9

#4: Str 12, Dex 11, Con 7, Int 18, Wis 15, Cha 13

I find all of these arrays quite interesting, especially the ones that result in characters you would never ever see under point buy. E.g. high Str/Dex and low Con on #1 would never happen in point buy.

[solidork]

I strongly prefer point buy.

[EggKookoo]

I never use point buy myself as a player. Too much work for too little gain.

I don't allow it at my table because, well, no one's asked to use it. But if someone did I'd ask why the standard array isn't good enough. I guess if they really wanted to go through the process I wouldn't stop them.

[Cynthaer]

I prefer point buy. The approach some have described, where you get to pick from anybody's roll, also works great because it eliminates massive power disparities.

For anyone interested, here's one of my older posts on why I think individual rolling for stats is bad for 5e (really anything post-3e), while it's a blast in earlier D&D editions:


I can't speak for WereRabbitz, but I wouldn't say players have changed. (I also wouldn't say it's a matter of "maturity".)

I think there are two factors:

First, I don't think having drastic (uncontrollable) stat disparities between characters has ever made a good game experience for games with long-term characters. Some people really enjoy it, of course, but most of the time I'd say it is a net negative for a gaming table chosen at random.

Second, the longevity of individual characters has massively increased across different editions. Everyone always talks about how back in the 80's they rolled up to the table with a binder full of character sheets because characters died permanently so easily and so frequently.

In that context, I think using pure dice rolls is vastly more fun than point buy. If I'm going through five characters in a single session, it's incredibly boring for them all to have nearly identical stats in slightly different configurations. Rolling stats makes Jonathan Fightingman and Jonathan Fightingman, Jr. different from each other.

With so many characters, it also means that each player is going to have a handful of high-stat characters at some point, and it means they're unlikely to have to play a low-stat character very long before they die anyway and the player can move on to a better character. And if they don't die, that's also exciting—now you have the story of how Medric the Mediocre survived a dungeon that killed two dozen more competent explorers!

You don't really have any of these effects in the modern D&D paradigm, where death is unlikely and reversible. Even at level 1, we expect our characters to be characters, rather than board game pieces, and we expect to play that character through to the end of the campaign unless something extraordinary happens.

In this context, there's no intrinsic emotional reward when you roll low on your one character. You don't need the "lows" to experience the "highs", because for you there is never going to be a "high"—this is the only character you're going to be playing. The only satisfaction to be had is if you take pride in standing by your word and using your low rolls, but for most people that doesn't make up for just having worse stats than their party members.

TL;DR: Power disparities are fun when they are temporary and players go through multiple characters over time. Power disparities are not fun when they are permanent and players only have one character for the entire campaign.

TL;DR: Power disparities are fun when they are temporary and players go through multiple characters over time. Power disparities are not fun when they are permanent and players only have one character for the entire campaign.

[JAL_1138]

In Adventure League, there's no choice. Point buy or bust.

Came here to say this, and also to add that I don’t roll in home games either.

Rolled stats worked okay in the TSR era. For example, 2e had pretty wide ranges where your stats kinda didn’t do much other than affect skill rolls (or the max level spells you could learn, for Int or Wis, if you were a wizard, cleric, bard, or druid). You didn’t start getting bonuses or penalties until much higher or lower stats than you do now. E.g., in 2e, bonus HP from Con didn’t start until 14 con, and were capped at a max of +2/level unless you were a warrior. Attack bonuses didn’t start until like 16 Str, and penalties didn’t kick in until you got as low as 7. Dex likewise—7 through 14 no effect on combat; penalties started at 6 and bonuses started at 15. Saving throws were by class and level, not by stats. Thief skills were off a separate percentage chance you could buy up as you gained rogue-group (thief or bard) levels, rather than your Dex. You also didn’t raise your stats as you leveled, or have to trade stat boosts for feats (because there were no feats). This meant that most characters didn’t have much in the way of stat-based bonuses or penalties one way or the other—for example, a character with stats of 12 12 10 10 12 11 wasn’t much different from 16 14 14 8 15 15. The second character qualifies for more classes/races and has a few more small bonuses, but not a huge amount.


In 5e, the game math is a lot tighter. Bonuses start at 12, penalties start at 9, your main stats need to increase with level to keep up, and everything keys off your stats—all skills/tools, all your character saving throws, etc. Rolling is much more likely to produce unbalanced characters, either stronger than usual or much weaker than usual (given the bell curve of 4d6b3, more likely to be underpowered/lackluster). The disparity between someone who rolled reasonably well and somebody who rolled middlin’ is much more noticeable and is likely to stick somebody with a bad set of stats. Point-buy allows people to play what they want to play, by and large, and largely eliminates the disparity. It is much, much, much better for 5e than rolling.

[Lance Tankmen]

I only do rolling and generally feel like majority use point buy/simple array. I do 4d6 drop 1 just like the PHB, and after 6 (two on going) campaigns Ive kept track of stats for fun, Simple array added together nets you 73, without racial. all my players past and present average around 72. lowest Ive seen was 60, highest 89. That said the 60 was a normal human fighter , he did alright.








Does anyone on this thread do 4d6k3 in order or 3d6 in order? (I've done it occasionally for NPCs or one-shot PCs and it's fun, but because it's not the default for 5E I usually don't.)

Just for fun, what would you do with the following stat rolls? Feats and multiclassing allowed but no stat swapping.

#1: Str 17, Dex 16, Con 9, Int 14, Wis 14, Cha 4

#2: Str 10, Dex 11, Con 13, Int 16, Wis 16, Cha 16

#3: Str 11, Dex 10, Con 14, Int 12, Wis 17, Cha 9

#4: Str 12, Dex 11, Con 7, Int 18, Wis 15, Cha 13

I find all of these arrays quite interesting, especially the ones that result in characters you would never ever see under point buy. E.g. high Str/Dex and low Con on #1 would never happen in point buy.
#1 id roll a human fighter, level four id go 20 str, level 6 id get 18dex 12 con. n just pump stats.
#2 drow bard, college of lore, id max cha then go from there
#3 hill dwarf Cleric of the forge, id go melee
#4 wizard, Vhuman one in dex one in con, tough for first feat lol

[Jophiel]

Point 2 is my real sticking point for rolling stats. People like gambling as much as they hate losing, which is why half of the tables that roll for stats have a fallback like point-buy, copying another player's rolls, or enforcing minimums.
Yeah. i mean, people should run their tables how they want but rolling feels cheaper to me than point buy because it's always gamed so you can't actually get a bad set which means you'll either be average or above average. For that, you might as well have everyone point buy with 36 points and allow purchases up to 18.

If someone out there actually does a straight "4d6, that's what you get" then that's a bit different. I'm not sadistic enough to promote 3d6 in order :smalltongue:

[Citan]

Sort of an informal survey, I’ve just started playing with a new group which is exciting as most of 2018 I was stuck as the “Forever DM”. Of this group, 5 (including myself) DM for other groups outside of the current campaign. When starting, we rolled for stats, which I certainly don’t mind. We were talking about different styles and I found I was the only one in the group who institutes point buy for campaigns.

One of my campaigns last year, which never got past session 0, I had a LOT of push back for using point buy. Having lurked these forums for a while, it seems the grand majority of people roll for stats.

Now, I don’t have anything AGAINST rolling for stats, but I personally feel point buy is more interesting. You have to make hard choices to build your character. Almost every character ive seen has rolled a good array of stats, to the point that nothing is lackluster. This seems a bit dull to me. Am I the only one in 5e that actually implements point but at their table?
Hi!

I'm not sure where you get the feeling the majority uses rolls here, I was on the opposite impression. Yet again when theorycrafting everyone usually stands on point-buy so we all use the same grounds so my impression may come from that.

Honestly I prefer (slightly bumped) point buy myself, but I get the reasons why people prefer roll stats and I appreciate using it sometimes, mainly when I go with reverse logic (I don't want to create a concept with rolls, I want to create it *from* rolls ^^).

I wouldn't say one is more "interesting" than the other really. IMO whether a character is interesting, in concept, RP or otherwise, all comes down to player choices.

[opaopajr]

Did A LOT of point buy, especially for AL. Got tired of it and 4d6 (a little too easy to get high stats for my 3d6 tastes). Now I do Random Point Buy: d8-1 stat increase, in any stat order, until your point buy pool of 27 points is spent. :smallsmile:

It bides the time until I get my ideal table 3d6 fix.

[Citan]

I don't see how this can be true. Point buy means that min-maxers will always be able to do things like dump Int and Cha in favor of combat stats (and then ignore their low Int/Cha while roleplaying) will always be more powerful than players who want to play smart and likable characters without breaking character. If you roll stats or use standard array, you can't steal points from one stat to spend on another, so there's a limit on how much min-maxing can impact play.

One of the things I most dislike about point buy, personally, is the tension it sets up between my powergaming instincts and my roleplaying instincts. Rolling stats doesn't present the same level of temptation, and I doubt I'm the only one who feels that way.
I dont see how the behaviour you describe is in any way related to pointbuy.
Unless you impose that rolling stat is made in respective order whatever happens. But that's usually not the case from what I've seen, players can choose how to relate dice results to attribute.

So a min-maxer will always set up its highest result in highest stat and ignore the dumped stats.
And in fact, if DM forced one to roll in order, he'd just decide what race and class to play after the fact.

The only situation where your assessment could hold truth is if a DM imposed player to *first* choose a class (and possibly a race) *then roll in order*.

[MaxWilson]

I dont see how the behaviour you describe is in any way related to pointbuy.
Unless you impose that rolling stat is made in respective order whatever happens. But that's usually not the case from what I've seen, players can choose how to relate dice results to attribute.

You would be correct to disagree if you were replying to someone who had written "min maxing cannot happen with rolled stats," but you were replying to me, and I happen to know I didn't write that. I wrote "there's a limit to how much min-maxing can impact play," not "the limit is zero."

You're refuting an argument that no one is making.

[CorporateSlave]

Agreed. There's no distinct power difference drawn between players. There's little risk of difficulty fluctuating between them. It rewards things like choosing an ASI over a feat, or racial benefits even more than they normally do, and it keeps players from trying to min-max.


I dunno, I loathe point buy, although I've certainly been in plenty of campaigns that have mandated it.

"There's no power difference drawn between players?" Well, with point buy, there certainly can be (and always has been in any of the point buy campaigns I've been in), but it tends to reflect along just how well optimized a player makes their PC. The ones who know the rules and tricky combos in and out do most everything better, and the other players wonder why their guy with the same stats to choose from hasn't managed to shine.

"It keeps players from trying to min-max?" I feel like when you say "min-max" you mean something different that how I understand it...I find just the opposite, it fully encourages min-maxing to the max-max! You get "forced" in to bumping your primary stats, along with dumping "the other" stats. Of course, that half gets fixed in game when players whose PC's have dumped to 8 INT and 8 CHA can still minimized those "weaknesses" by RPing their character as if they had 18 INT and 18 CHA. Sure, they still get the modifier on saves or whatnot...but hey just play a Paladin or Monk or somebody else who gets plusses in all the right spots to nullify the negatives! In fact, it forces min-maxing since if you don't optimize and dump, you better hope nobody else does, because you'll end up with just that "power difference" you were trying to avoid.

I find that rolling characters makes for generally more rounded PC's, although certainly not as optimized in most cases. Every once and a while someone ends up with a long and short end of the stick, but in the campaigns we have used rolling, usually there is some "contingency" for a player that rolls really craptacular arrays (along the lines of a full re-roll if all stats added together don't equal a certain number like 70 or so).

Of course, either way it will come down to how you decide to play your character.

[druid91]

I prefer point buy. The approach some have described, where you get to pick from anybody's roll, also works great because it eliminates massive power disparities.

For anyone interested, here's one of my older posts on why I think individual rolling for stats is bad for 5e (really anything post-3e), while it's a blast in earlier D&D editions:

TL;DR: Power disparities are fun when they are temporary and players go through multiple characters over time. Power disparities are not fun when they are permanent and players only have one character for the entire campaign.

The issue with that is that that hasn't changed. Certainly, official modules shifted towards not killing players. But that's been the case as far back as AD&D 2nd Edition. I played a game of 5e where only the Wizard managed to survive to the end of the campaign. With everyone else going through several characters, and the only reason the Wizard survived was a combat doctrine of 'Run away if it looks like you might get attacked.' and 'Zap it from VERY FAR AWAY.'

[DeTess]

You would be correct to disagree if you were replying to someone who had written "min maxing cannot happen with rolled stats," but you were replying to me, and I happen to know I didn't write that.

You're refuting an argument that no one is making.

It seemed that you where arguing that point-buy facilitates min-maxing, which, lacking a point stating that rolling does nothing to deter min-maxing either makes it look like you're arguing that rolling is superior in this regard.

Now, clearly this isn't what you meant to communicate, but it was what your post seemed to imply.

[Pex]

I don't like 5E's implementation of it, but I use it. I have a personal favorite rolling method, but I chose to go for straight simplicity. My campaign is light hearted. You don't need to be too concerned. Optimize as you want; just be competent.

[NaughtyTiger]

You would be correct to disagree if you were replying to someone who had written "min maxing cannot happen with rolled stats," but you were replying to me, and I happen to know I didn't write that. I wrote "there's a limit to how much min-maxing can impact play," not "the limit is zero."

You're refuting an argument that no one is making.

You do imply that there is less min-maxing when you roll...

you can still ignore roleplay effects of low rolled Int/Cha
you can still ignore roleplay effects of standard array Int 8/Cha 10

as Citan pointed out (and you did not refute) you can still assign the low 8 to Int and high 17 to Str.


I don't see how this can be true. Point buy means that min-maxers will always be able to do things like dump Int and Cha in favor of combat stats (and then ignore their low Int/Cha while roleplaying) will always be more powerful than players who want to play smart and likable characters without breaking character. If you roll stats or use standard array, you can't steal points from one stat to spend on another, so there's a limit on how much min-maxing can impact play.

Nothing stops you from ignoring low Int and Cha in rolled stats either... except, a competent DM.

[MaxWilson]

It seemed that you where arguing that point-buy facilitates min-maxing, which, lacking a point stating that rolling does nothing to deter min-maxing either makes it look like you're arguing that rolling is superior in this regard.

Now, clearly this isn't what you meant to communicate, but it was what your post seemed to imply.

Point buy provides new, reliable, additional ways to min max. Under point buy, min maxing can now be used on *every* character, not just the ones where you happen to roll three high stats and three low stats, which is rare.

It exacerbates the issue instead of alleviating it.



as Citan pointed out (and you did not refute) you can still assign the low 8 to Int and high 17 to Str.

Nothing stops you from ignoring low Int and Cha in rolled stats either... except, a competent DM.

But only in point buy can you turn Str 13 Int 12 into Str 15 Int 8 and then ignore the Int 8.

[KyleG]

I have a half elf who although growing up was trained in the sword 10/12/12/12/10/12 pursued an academic career (int now 14). When his wife was killed he joined an army for 5yrs (dex & con now 14, wis 12) thru circumstances he takes a pact and along with a new sense of purpose came the willpower (cha now 14) to hunt down those responsible. Wee bit boring stat wise but point buy allows me to tell his story to this point, although im considering taking some intelligence as part of his pact as some of his memories, namely whom he is hunting has been taken by his patron.

[DeTess]

Point buy provides new, reliable, additional ways to min max. Under point buy, min maxing can now be used on *every* character, not just the ones where you happen to roll three high stats and three low stats, which is rare.

It exacerbates the issue instead of alleviating it.

As opposed to situations where the min-maxer doesn't have to minmax because they get 6 decent rolls, or absolutely has to pull out all the optimization tricks because they got 6 average-poor rolls?

Maybe it's just a different way of looking at it, so let me do a quick breakdown, in which I'll use very broad generalizations. Let's take Player A, who is a min-maxer, and player 1, who prefers to have a well-rounded character.

With point-buy, Player A will make a min-maxed character, and player 1 will make a well-rounded character. Both will be happy with their character, though player A's character will be better in his area of expertise than player 1's character, which might cause some friction, but at the end of the day it'd be clear to both players that the initial playing field was level. Player 1 chose to sacrifice some power for making a well-rounded character, and a good DM will make sure the specialist has to deal with their weaknesses once in a while.

With rolling (assuming every player rolls their own stats and can't use anything else), player A will either get a really good array, removing the need to minmax, a really poor array, which someone enjoying power would either dislike or see as a permission slip to go deeper in TO, or an array that roughly lines up with what they'd have done anyway (some highs, some lows, to match the dump-stats and primary and secondary stats). Player 1, on the other hand, might end up in a situation where they're forced to pick dump-stats and sacrifice their well-rounded concept, just because they rolled 3 lows and 3 highs, or all lows and 1 high. In this case, player 1 is more likely to end up with a character that they're less happy about. The power disparity also remains, and is potentially worse, unless player 1 got a good roll and player A a poor one, which could once again lead to some friction, and this time it's harder to argue that the initial conditions for both character are equal.

So based on this I'd say point-buy is probably better because it gives both players what they want, and if the power difference bothers player 1 so much, there's nothing preventing them from evening the playing field with the next character. The one situation where rolling might be better is if you're switching character pretty often, like once per month at least. This'd result in statistical extremes averaging out, and will make building a lot of characters in a short time less formulaic.

[Helldin87]

imo rolling doesn't add anything. Point buy forces players to make choices and inherently balances them against other players. Its not fun when one member of your party just happens to have multiple 18's and you are stuck with 14's. Even systems that allow you to roll and skew results in your favor just create a party of superheros that the DM now has to balance against.

Point buy at all my tables!

[NaughtyTiger]

Point buy provides new, reliable, additional ways to min max. Under point buy, min maxing can now be used on *every* character, not just the ones where you happen to roll three high stats and three low stats, which is rare.

It exacerbates the issue instead of alleviating it.

But only in point buy can you turn Str 13 Int 12 into Str 15 Int 8 and then ignore the Int 8.

again, you imply that there is more min-maxing with point buy. and imply that point buy min maxing is more of a power imbalance than rolling really well compared to a poorly rolled character.

i agree rolling 15, 16, 18, 4, 5, 6 is rare.
but me rolling everything above 13, and you rolling 4 stats below 10 is common.

[ad_hoc]

I think if you are going to roll you really ought to roll in order.

If your thing is about having random stats so that the sort of character you are playing is determined randomly, then rolling in order is the way to do it.

Rolling and picking where the stats go is just point buy but with higher numbers.

[Luccan]

I think if you are going to roll you really ought to roll in order.

If your thing is about having random stats so that the sort of character you are playing is determined randomly, then rolling in order is the way to do it.

Rolling and picking where the stats go is just point buy but with higher numbers.

So, I know rolling can generate higher numbers and, depending on how you do it, does so reliably. However, it doesn't always. Even 4d6b3 7 times can produce much lower numbers than point-buy. That's the risk (assuming you stick to the numbers rolled and don't give a back up). Maybe you get 2 18s, a 16, 14, 12, and a 10. You might also get what I got once: 6,16, 12, 9, 12, 7.

Edit: Point being, the order didn't exactly matter.

[Agent-KI7KO]

Standard Array or Pointbuy only.

But if my players want more power to start i give either:
31 pointbuy (or 16,14,13,12,10,9)
27 pointbuy (or standard array) with extra ASI/FEAT AT 4/12.

[Mitsu]

I absolutely hate stats rolling and I never roll stats in any system. Not only it can lead to huge gaps between players (one will roll 11s and 9s while other one will roll 15s and 14s) but it either leads to underpowered character or totally overpowered one who starts with 20 in one ability.

Point Buy is fair for all players and it requires to have some stats lower than others, thus giving some weakness to characters. High STR, CON character? Suffer low DEX saves. High DEX and WIS character? Watch those athlethic checks and CON saves etc.

Also Point Buy in my opinion balances races nicely. With rolls some races with +2 to stats can really start totally OP.

[KyleG]

Standard Array or Pointbuy only.

But if my players want more power to start i give either:
31 pointbuy (or 16,14,13,12,10,9)
27 pointbuy (or standard array) with extra ASI/FEAT AT 4/12.

Is that the feats at 4 and 12? Why not to start with?

[Agent-KI7KO]

Is that the feats at 4 and 12? Why not to start with?

I thought about it, but I decided to let Vumans have their extra feats for 3 levels.

[RogueJK]

Just for fun, what would you do with the following stat rolls? Feats and multiclassing allowed but no stat swapping.

#1: Str 17, Dex 16, Con 9, Int 14, Wis 14, Cha 4

#2: Str 10, Dex 11, Con 13, Int 16, Wis 16, Cha 16

#3: Str 11, Dex 10, Con 14, Int 12, Wis 17, Cha 9

#4: Str 12, Dex 11, Con 7, Int 18, Wis 15, Cha 13


#1 Shadar Kai Battlemaster Fighter with 2-3 level dip into Rogue, focusing on ranged Longbow attacks from hiding, with Expertise in Perception and Stealth. +2 DEX or Sharpshooter at 4th. Sharpshooter or +2 DEX at 6th. Skulker at 8th. Tough at 12th.
#2 Half Elf Divine Soul Sorcerer with 2-3 levels of Hexblade Warlock, wearing Medium Armor and Shield, and using CHA for both casting and weapon attacks. +1s in DEX and Con. +2 DEX at 4th level ASI. +2 CHA at 8th. +2 CON or Warcaster at 12th.
#3 Ghostwise Halfling Moon Druid with 1 level Monk dip. Spend most combats Wildshaped. +2 WIS at 4th. +2 DEX at 8th. +2 DEX at 12th.
#4 Hill Dwarf Abjuration Wizard with 1 level dip in Life Cleric for Heavy Armor and Shield, plus some backup healing spells and additional Abjuration spells. Resilient CON at 4th. +2 INT at 8th. +2 CON at 12th.

(I actually really like the concept of that 4th one... A fairly weak/sickly Dwarf who turned to healing magic and arcane power to compensate for it.)

[Lunali]

The biggest problem with rolling is differences in power level within the party, if the party as a whole is too strong or weak, the DM can adjust encounters. That's why if I do roll for stats I let/argue for having everyone roll and anyone can use any of the rolled sets.

[Draconi Redfir]

tables i go to almost exclusively use point-buy. which is a shame since i like the more varied ability-score diversity that rolling provides.

nothing worse then trying to roll up a character and finding that the best you can get is one 16-ish, one 8, and then like, four 11's.

[JakOfAllTirades]

Only one 5E table (my current one) that I've played it since the game came out uses randomly rolled stats. The rest all used point buy, except for one which preferred the standard array.

[TyGuy]

All personal opinion but I don't like how point buy is kind of sterile and I would rather discover my character's stats instead of design them. But I like how point buy/standard array doesn't have power disparities.

ENTER THE CARD METHOD
You get randomization, you get the excitement of discovering your array, you get a set score sum.
1 give your players 12 sorted cards
2 have them pick from the stack and place pairs for each stat six times
3 have them arrange their pairs, or don't

Boom, done. It's random but standardized at the same time.

Example of a 72 point deck with a chance at the coveted starting 18 or a dreaded 6:
10, 8, 8, 7, 7, 6, 6, 5, 5, 4, 3, 3

As the most polar draw one could get 18/15/13/11/9/6 or in the most neutralized a 13/12/12/12/12/11. But realistically it'll be somewhere between.

[ad_hoc]

All personal opinion but I don't like how point buy is kind of sterile and I would rather discover my character's stats instead of design them. But I like how point buy/standard array doesn't have power disparities.

ENTER THE CARD METHOD
You get randomization, you get the excitement of discovering your array, you get a set score sum.
1 give your players 12 sorted cards
2 have them pick from the stack and place pairs for each stat six times
3 have them arrange their pairs, or don't

Boom, done. It's random but standardized at the same time.

Example of a 72 point deck with a chance at the coveted starting 18 or a dreaded 6:
10, 8, 8, 7, 7, 6, 6, 5, 5, 4, 3, 3

As the most polar draw one could get 18/15/13/11/9/6 or in the most neutralized a 13/12/12/12/12/11. But realistically it'll be somewhere between.

I'm not sure I am understanding this.

The card method I have used is:

Take 2s-6s from a deck of cards. Toss out two 4s.

Shuffle. Deal out 6 piles of 3 cards.

Result is 6 stats from 6-18. Sum is the same for all players (you could leave in the two 4s for a variable sum) and odds of a low or high score are quite low.

[Citan]

You would be correct to disagree if you were replying to someone who had written "min maxing cannot happen with rolled stats," but you were replying to me, and I happen to know I didn't write that. I wrote "there's a limit to how much min-maxing can impact play," not "the limit is zero."

You're refuting an argument that no one is making.
Great way to avoid actual argumentation.
Quoting yourself : "Point buy means that min-maxers will always be able to do things like dump Int and Cha in favor of combat stats (and then ignore their low Int/Cha while roleplaying) will always be more powerful than players who want to play smart and likable characters without breaking character."

You can do EXACTLY the same with rolled-stats. As long as you choose which result goes to which attribute.
It's just potentially worse since you could get a 18 and a 6 so the min-maxing has more variance.

Try again, less agressivity, more objectivity plz.
I dunno, I loathe point buy, although I've certainly been in plenty of campaigns that have mandated it.

"There's no power difference drawn between players?" Well, with point buy, there certainly can be (and always has been in any of the point buy campaigns I've been in), but it tends to reflect along just how well optimized a player makes their PC. The ones who know the rules and tricky combos in and out do most everything better, and the other players wonder why their guy with the same stats to choose from hasn't managed to shine.

"It keeps players from trying to min-max?" I feel like when you say "min-max" you mean something different that how I understand it...I find just the opposite, it fully encourages min-maxing to the max-max! You get "forced" in to bumping your primary stats, along with dumping "the other" stats. Of course, that half gets fixed in game when players whose PC's have dumped to 8 INT and 8 CHA can still minimized those "weaknesses" by RPing their character as if they had 18 INT and 18 CHA. Sure, they still get the modifier on saves or whatnot...but hey just play a Paladin or Monk or somebody else who gets plusses in all the right spots to nullify the negatives! In fact, it forces min-maxing since if you don't optimize and dump, you better hope nobody else does, because you'll end up with just that "power difference" you were trying to avoid.

Of course, either way it will come down to how you decide to play your character.
Nothing is ever "forcing you" to bump main stats and dump the others besides the latent minmaxer inside you.
And you see to get the feeling rolling nets you overall better stats, but the statisticical average total can hide a wide disparity of actual rolls.
If you get a 7 and an 16 on roll, where would you put the 16, hmm? :)
If you get all rolls between 11 and 13, what's the big difference with point-buy, hmm?

Nobody forces anyone to play point-buy characters with dumped stats.

In the end, as you imply at the end of your post, it's all about how the *player* thinks, builds and roleplay.
Method of getting attributes is overall irrelevant, the only differendecce between point-buy and roll is that because the total points are set, whether someone min-maxes or balances, you tend to see the same half a dozen combinations (like 14 and 12 across the board, starting 17/16 and several 8, etc).

Point buy is cool when you have something in mind, or when you want precisely to avoid dump stats (of course, it means you won't have that high as primary stat). Roll is cool when you don't mind...
- Getting one or several trash stats.
- Or deciding what to play "from" the stats.
Or you simply feel lucky. :)

As a reminder, that is precisely the reason why designers went for bounded accuracy (and otherwise went to great length to ensure even the most improbable multiclass character can still work somewhat). So people can play whatever concept they would like without feeling useless.
Whether one embraces that or keep your min-max shadow pulling the strings, that's up to oneself. Putting that on attribute determination method is just blinding self. ^^

[Aquillion]

I don't see how this can be true. Point buy means that min-maxers will always be able to do things like dump Int and Cha in favor of combat stats (and then ignore their low Int/Cha while roleplaying) will always be more powerful than players who want to play smart and likable characters without breaking character. But you can't lower a stat to below 8 or above 15 via point-buy. Rolling allows for much bigger disparities, which will stand out more when optimized around and will make ignoring them much more stark.

[Lyracian]

Just for fun, what would you do with the following stat rolls? Feats and multiclassing allowed but no stat swapping.
That does sound fun; lets see...



#1: Str 17, Dex 16, Con 9, Int 14, Wis 14, Cha 4

Race - Half-Elf +Str, Con; +2 Chr.
Class - Maybe Eldrich Knight or a Rogue/Fighter with a Wizard dip to make use of the Int.
Final stats Str 18, Dex 16, Con 10, Int 14, Wis 14, Cha 6



#2: Str 10, Dex 11, Con 13, Int 16, Wis 16, Cha 16

Bard, Sorcerer, Wizard or Cleric are all good here.
Lack of Dex pushes me towards Half-Elf Bard with a level of Cleric for Medium Armour. Can use an ASI to pump dex to 14. Second choice would be Hill Dwarf Cleric for Heavy Armour.



#3: Str 11, Dex 10, Con 14, Int 12, Wis 17, Cha 9

Wiz 17 Screams Cleric. This will be Hill Dwarf Cleric for Heavy Armour.
Racial Mods make the stats Str 11, Dex 10, Con 16, Int 12, Wis 18, Cha 9



#4: Str 12, Dex 11, Con 7, Int 18, Wis 15, Cha 13

Human with Res(Con) as feat then first ASI being +Con to make it 10. Alternate is Dwarf for starting with Con 9 then taking Res(Con) at level 4. Wizard with first level being Cleric for Armour.

Human has a choice to pump dex to 12 and wear Scale Mail+Shield = AC 17. Level 12 ASI to get 14 Dex. Otherwise start with 13 Str for Chain Mail + Shield giving AC 18 with no speed penalty. Is is one more point of AC but means you end up with even more odd numbered stats. Dwarf just takes Heavy Armour since there speed is not reduced.

I think my final choice would he Human with Str 12, Dex 12, Con 8 Res(Con), Int 18, Wis 16, Cha 13

[TyGuy]

I'm not sure I am understanding this.

The card method I have used is:

Take 2s-6s from a deck of cards. Toss out two 4s.

Shuffle. Deal out 6 piles of 3 cards.

Result is 6 stats from 6-18. Sum is the same for all players (you could leave in the two 4s for a variable sum) and odds of a low or high score are quite low.

It's pretty much the same thing. One method uses 12 cards with pairs, the other uses 18 cards with triplets.

[Malifice]

Only point buy here.

[Particle_Man]

I implemented the array when I DMed. The players were fine with that.

[ad_hoc]

It's pretty much the same thing. One method uses 12 cards with pairs, the other uses 18 cards with triplets.

I see, I like my version better. Gives a much better curve.

Mine is also 72pts but much less likely to give 16-18 scores (or even multiple ones).

[Cybren]

Rolling is high variance: no matter how skilled you are, an 18 is an 18 and a 3 is a 3. Rolling in order is obviously the highest variance, as you have no decision points with your ability score array at all, while rolling and assigning the scores where you like introduces six decision points while still retaining variance.

Point buy is low variance and has the most decision points. In terms of “min maxing” point buy encourages players to be more utilitarian with how they assign ability scores and select races, because the system encourages it.

Optimizers and power gamers tend to dislike variance and want more decision points where their greater game knowledge and skill can shine. But this also means that point buy isn’t an “even playing field” unless all the players are engaging with it on the same level.

Also I composed this tweet on mobile while distracted at work so I have no idea how many responses I missed

[Willie the Duck]

Does anyone on this thread do 4d6k3 in order or 3d6 in order? (I've done it occasionally for NPCs or one-shot PCs and it's fun, but because it's not the default for 5E I usually don't.)

We've experimented somewhat with this. I think we tried 3d6 in order, but then used the B/X or BECMI stat bonuses, plus one (so +2 for 13-15, +3 for 16-17, +4 for 18) and then ASIs could only be spent on feats (including half feats, so there was some stat changing, but minimal). It worked fine as an experiment, but didn't achieve any particular goal, so we didn't continue.


Just for fun, what would you do with the following stat rolls? Feats and multiclassing allowed but no stat swapping.

#1: Str 17, Dex 16, Con 9, Int 14, Wis 14, Cha 4
Vuman (evening out Str and Con) fighter with Toughness or MAM as feat. MAM if you want to go the criminal background route (replacing deception with acrobatics or the like), or Toughness (and later SS and GWM) to play a traditional 2e AD&D fighter who wears plate mail, and wields both longbow and greatsword. StRogue would also work.


#2: Str 10, Dex 11, Con 13, Int 16, Wis 16, Cha 16
Start out vuman to even of Dex and Con and grab war caster. Cleric other than Arcana cleric, then sorcerer, warlock, or wizard. Then make that Spirit Guardians 'penalize you for staying close, but use Booming Blade and Warcaster to really penalize you for trying to escape' build people built right when SCAG came out, just without using Arcana Cleric to pick up the SCAG cantrips.


#3: Str 11, Dex 10, Con 14, Int 12, Wis 17, Cha 9 Wood elf Knowledge cleric Investigator. Get Dex to 14, Wisdom to 20 (eventually), and then thematically appropriate noncombat feats.


#4: Str 12, Dex 11, Con 7, Int 18, Wis 15, Cha 13 With 18 Int it is hard not to go wizard (perhaps also an investigator, what with good observation and social skills). In a social campaign, class almost wouldn't matter (perhaps rogue for extra skills and expertise). If survival is ever challenged, a hill dwarf Cleric/Wizard would be hard to pass up.

[CorporateSlave]

Nobody forces anyone to play point-buy characters with dumped stats.


Well, nobody specific, just the concept that point buy creates some sort of "equality" between the players. Because unless ALL players either optimize or ALL players don't and go for well rounded, you end up with just the same problem of power imbalance, which seems to be a key rallying cry for point buy.

My point is, point buy DOESN'T prevent imbalance. It just creates predictability. If thats what you want, then go for it. But don't tell me it is some panacea of equity between players, because unless the DM forces them to spend the points a certain and similar way, or they all decide to do so as a group, it simply doesn't.

Like this guys says:


Optimizers and power gamers tend to dislike variance and want more decision points where their greater game knowledge and skill can shine. But this also means that point buy isn’t an “even playing field” unless all the players are engaging with it on the same level.


BTW, I'm fine with point buy when it is implemented at a table, I don't whine and threaten to take my dice and go home. What the DM wants is what the DM wants, and I'll work with it. But that isn't going to mean I like point buy, or that I (point) buy a lot of the arguments people make in favor of it.

[Rhedyn]

If you aren't rolling 3d6 down (as in you roll 3d6 for a stat with no switching around), then I see rolling as just a way to get higher stats than point buy.

One system I like is you roll 4 to 5 "arrays" of 3d6 down and then you pick a character that "survives" to level one and play that one. I feel like if you are rolling stats you just shouldn't have that much control, but I also think one bad set of rolls shouldn't ruin you for a whole campaign.

[Ganymede]

I always use point buy for my characters, and most of them end up being in line with the standard array.

Rolling down the line can be a fun way to make a new character based on the stats you roll. That's how I ended up with one of my most memorable characters: a dexterity six dwarven cleric with a penchant for drinking mineral spirits.

[Gtdead]

I generally like point buy, because I want to have more control over the character I build. For example, if I want to play a gish, and I only get one super high stat, and all the others average, it will make some of the more MAD builds obsolete. I like to decide on class and background first, and build/numbers later.

So in a campaign where the background will be explored and may affect the story, I'd rather go point buy.

If the background is irrelevant to the story and the introduction to the party is forced, I may as well go full random on everything, stats, class, alignment etc. This may be a scenario of an all out war, where the party is a bunch of conscripts in a unit, or a crew that gets marooned on an island.

[Contrast]

Well, nobody specific, just the concept that point buy creates some sort of "equality" between the players. Because unless ALL players either optimize or ALL players don't and go for well rounded, you end up with just the same problem of power imbalance, which seems to be a key rallying cry for point buy.

My point is, point buy DOESN'T prevent imbalance. It just creates predictability. If thats what you want, then go for it. But don't tell me it is some panacea of equity between players, because unless the DM forces them to spend the points a certain and similar way, or they all decide to do so as a group, it simply doesn't.

Except you really have to try to mess up point buy badly that the difference is more than +1, whereas its pretty easy to end up with +2-3 differences with rolling. If someone turned up having used point buy to build a fighter with 10 Str, Dex and Con I would have a word with them to ask if they realise what they're doing. If someone rolls terribly I...let them reroll? In which case why are we rolling?

I don't mind rolling but I prefer when the DM introduces rules to mitigate the variance - a common one is that everyone rolls and people can chose whichever set they like. Except now we're just back to the issue whereby everyone has the same stats and everyone is a cookie cutter of each other. In that situation it feels to me as if you may as well have just used the standard array or point buy and let people play what they want.

[Chalkarts]

I like point buy.

If a player rolls stats and ends up with 5 10s and a 15 I could understand a level of annoyance and occasional complaints about their crappy stats.

However, with point buy the players can't complain.
If a player complains about having an 8 in wis and int, just remind them, YOU DID THIS TO YOU.


My favorite method of all, is 4d6, drop the lowest, roll one set, no rerolls. Your first 6 rolls are your stats. Fate is your master.

[Solusek]

I like rolling stats a lot. I think it's fun. Though the default 4d6 drop lowest isn't the system I would use. It creates characters that vary too much in ability. Some type of group roll where everyone ends up with the same choices would be my preference. That way no one feels screwed that their rolls sucked compared to the rest of the parties.

HOWEVER, the only D&D game I have regularly played in over the last 20 years exclusively uses point-buy so that's what I'm stuck with playing.

[SociopathFriend]

Bear in mind Stat Rolling typically has additional caveats like "take the highest 3" or "reroll 1s" and the like.
If you instead just rolled 3 die and made those your score, you've got some pretty decent odds of having abysmal scores.

Rerolling 1s tends to give you a generally favorable spread but doesn't guarantee you'll have all great stats. I've rolled and started with I believe 3 16s and 3 10s after racial adjustment. Good, but not THAT good.

We did use point-buy the latest campaign because the store-owner wants more Adventure League so we at least tried to go with those rules.

The result is most people having 10s or lower in non-main abilities and we got to see firsthand why that's dangerous when an Intellect Devourer pair jumped one guy and each hit him to drain his Intelligence- dude dropped in one go. Didn't matter about his HP at all. Not your every day battle mind you but still.

I like rolling because it feels more natural to me. You're not a Strong because you're a Fighter, you naturally would've leaned towards a Fighter-style adventurer career because you were Strong growing up due to the (random) factors of your birth.

[Sigreid]

So, to sum up the answer to the OP:

A good number of the people like to roll because they like the variance between characters and/or the chance that they might get to be a little above at the risk of being a little behind.

A good number of the people like standard array because they know what they are getting and/or they prefer that no player feels that their character is gimped by luck compared to another.

A good number of people like point buy because it gives them the ability to make exactly the character they want to play.

Very few people will let a particular campaign's method determine whether they play the game or not.

That about sum it all up?

[Calimehter]

That about sum it all up?

Pretty much!

I have my 5e group use one of two arrays: 15,14,13,12,11,10 . . . or 17,13,12,11,10,9. I've played with lots of other character generation systems though, and would do so again. :)

I admit to being a bit biased against random rolls by my history of bad dice during character generation. I don't usually have a problem with bad dice during game play, but for some reason the little buggers just seem to "know" when I'm rolling them for stats that will impact the entire campaign.

[KyleG]

Point buy or standard array also means the characters which are already exceptional aren't Gods yet. With a max 15 and racials maxing 17 they are already well above normal. Maybe point buy and standard array needs to be lower but feats are offered. What impact would maxing out level 1 abilities at 14 but giving 2 feats have? If a campaign reached level 20 (by all accounts rare) how maxed do ability scores need to be?

[KorvinStarmast]

People who do none of those things, and just want to play the game (particularly if they started gaming pre-WotC) all seem to do 4D6best3, and don't seem to know why they are supposed to want to change.Yeah. I didn't start doing point buy until I was considering an AL character


I never use point buy myself as a player. Too much work for too little gain.
Not much work to add and subtract, but I guess YMMV.
But if someone did I'd ask why the standard array isn't good enough. It sucks due to being a low ball of the average of the default roll, and no flexibility.

It also shows us that the average roll is roughly (https://anydice.com/articles/4d6-drop-lowest/) 16, 14, 13, 12, 10, 9.
I prefer the "roll and see" excitement or "build it the way you want it"

There was a method on the AD&D 1e DMG that was one used by a lot of DMs I played with. Roll 3d6 12 times and pick the six best. Move the scored where you want them.
I had a DM to also did the "roll 18d6 and place the numbers where you will." Which kind of led to some min and max behaviors, but it let you build a paladin if you wanted one. Rangers, paladins, Illusionists: you had to qualify for those classes with various minimum scores. this method let you do that.

[McSkrag]

We always use point buy.

It's fair to all players and balanced for the game.

[Calen]

I have always used point-buy in all my games.

I once told myself I would roll some NPC's randomly just to see what would happen. When the first one had 3 18+ stats I decided to never do that again :P

Though as an aside. If you like the idea of random why not roll a d6 to determine which stat is the players highest? (To start) this allows those that like the idea of random to create a character with limitations but still be balanced statistically with their allies.

[Coffee_Dragon]

A good number of the people like to roll because they like the variance between characters and/or the chance that they might get to be a little above at the risk of being a little behind.

That about sum it all up?

I'm not really seeing much of the "risk of being a little behind" bit. There are a very few celebrating that time the dice told them they were going to play a flat 11 character. The majority are using safety valves, rigging rolls to ensure above-elite arrays, or simply ignoring the prospect of getting average or sub-average results. "I prefer rolling because unlike point buy that lets me put a high score in a non-synergistic ability for the role-playingz [after I already put equally high or better in my primary stat, secondary stat and CON, because I'm not giving up minmaxing, you crazy?]."

[MaxWilson]

I'm not really seeing much of the "risk of being a little behind" bit. There are a very few celebrating that time the dice told them they were going to play a flat 11 character.

Low stats are part of the fun. Low stats are often more fun than mediocre stats (like your flat 11 character--very mediocre, someone who is slightly above average at everything). But even the mediocre stats can be fun, especially because getting all mediocre rolls with nothing good (14+) and nothing bad (8 or less) doesn't happen too often.


Though as an aside. If you like the idea of random why not roll a d6 to determine which stat is the players highest? (To start) this allows those that like the idea of random to create a character with limitations but still be balanced statistically with their allies.

You still wind up with a boring, artificially-constrained stat range that looks too much like every other character's stat range (e.g. no stats below 8 or above 15, every single time). That's why.

Mechanically there's not much difference between an Int 6 Dex 8 Paladin and an Int 12 Dex 4 Paladin, and they'll be about equally "powerful" in play, but they're pretty different from a RP angle, and that is interesting to some of us.

[Sigreid]

I'm not really seeing much of the "risk of being a little behind" bit. There are a very few celebrating that time the dice told them they were going to play a flat 11 character. The majority are using safety valves, rigging rolls to ensure above-elite arrays, or simply ignoring the prospect of getting average or sub-average results. "I prefer rolling because unlike point buy that lets me put a high score in a non-synergistic ability for the role-playingz [after I already put equally high or better in my primary stat, secondary stat and CON, because I'm not giving up minmaxing, you crazy?]."

A lot do seem to do that, but there are those of us out there that do enjoy a sub standard character. One of the most fun I've ever played had a high stat of 12.

[Tawmis]

Sort of an informal survey, I’ve just started playing with a new group which is exciting as most of 2018 I was stuck as the “Forever DM”. Of this group, 5 (including myself) DM for other groups outside of the current campaign. When starting, we rolled for stats, which I certainly don’t mind. We were talking about different styles and I found I was the only one in the group who institutes point buy for campaigns.

One of my campaigns last year, which never got past session 0, I had a LOT of push back for using point buy. Having lurked these forums for a while, it seems the grand majority of people roll for stats.

Now, I don’t have anything AGAINST rolling for stats, but I personally feel point buy is more interesting. You have to make hard choices to build your character. Almost every character ive seen has rolled a good array of stats, to the point that nothing is lackluster. This seems a bit dull to me. Am I the only one in 5e that actually implements point but at their table?

I prefer Point Buy also. This gives everyone the same amount of points to "pool" from. So you're not running with a Cleric who has 18 Wisdom, 17 Strength, 16 Constitution, 14 Dexterity, 16 Intelligence, and 12 Charisma. While rolling such stats seems fair, you can then also have a character (such as said cleric) who can be overpowering and overshadow other players.

[Coffee_Dragon]

You still wind up with a boring, artificially-constrained stat range that looks too much like every other character's stat range (e.g. no stats below 8 or above 15, every single time). That's why.

Mechanically there's not much difference between an Int 6 Dex 8 Paladin and an Int 12 Dex 4 Paladin, and they'll be about equally "powerful" in play, but they're pretty different from a RP angle, and that is interesting to some of us.

It wouldn't have killed them to put something in the Ability Options section of the DMG along the lines of, "Ordinarily the game assumes each character is a promising adventurer with no glaring deficiencies, but for experienced players who want characters with such flaws, just relax the restrictions in the point buy system thus and so." It's a little odd that the default method allows for low scores but the more involved method doesn't.

[Luccan]

It just comes down to people not wanting to feel like they aren't contributing. 5e might not rely on numbers quite as much as some editions (up for debate, I know), but it certainly relies on them more than the earliest editions. Point-buy at least gives the option that if you're mediocre at everything, it was because you built the character that way, not because you rolled a bunch of 10s, 11s, and 12s.

Also, to echo a sentiment from earlier, having bad stats and surviving just isn't as impressive this edition. It's hard enough to die in the first place, healing is sometimes only a short rest away, and full recovery doesn't take weeks or months anymore.

[Merudo]

I never use point buy myself as a player. Too much work for too little gain.


There a few online point buy calculators, I typically use this one (http://chicken-dinner.com/5e/5e-point-buy.html).

It makes it extremely easy to build a character using point buy.

[EggKookoo]

There a few online point buy calculators, I typically use this one (http://chicken-dinner.com/5e/5e-point-buy.html).

It makes it extremely easy to build a character using point buy.

I'm probably doing something wrong but that calculator lets me use up to 54 points. Even if I limit myself to spending 27 (which isn't something I'd want to police as a DM), I don't seem to be able to make a set that varies significantly from the standard array. I'm not sure that after adding racial adjustments, a point buy PC and a standard array PC would play differently enough for me to notice.

I have my players roll 4d6 and drop the lowest (although I'm intrigued by 3d6 with rerolling 1s), or have them pick the standard array. I'm not even sure I'd bring up point buy but if a player did I suppose I wouldn't stop them.

[Maelynn]

I very much prefer point-buy, as it allows me to customise my character to what I want them to be - a good reflection of their personality and background. I'm not a min-maxer, I'm not an optimiser, I just really enjoy making a 'person' and then get the stats that fit them.

Standard array also works, but is a little restrictive for what I enjoy about character building. Still, the stats are decent and there's no outliers within the party. I wouldn't be against a DM using this.

I hate rolling. I had to do it in my first game (3.5) and ended up with a character that had 15 as his highest stat (and the only one above 12). Other characters in my party had stats as high as 18 and would perform better. I already was a tad insecure being a new player, but when I was forced to be so mediocre compared to half the party, who often excelled at what their class was meant to do... it just felt unbalanced and unfair. It left a bit of a bitter taste for rolling stats.

However, I must say that the suggestion mentioned here by a few - to create a pool of stats together and have all character use the same set.. I could see that work. It would prevent the possibility of having large differences between characters in the same party. And if you still want a bit of diversity, you could roll together for 8-9 stats and have everybody pick 6.

[Jophiel]

Point buy doesn't vary much from Standard Array because the array is based off of the point buy system. However, point buy allows you to fine control your stats a little more from 13 13 13 12 12 12 to 15 15 15 8 8 8.

[Chalkarts]

Point buy doesn't vary much from Standard Array because the array is based off of the point buy system. However, point buy allows you to fine control your stats a little more from 13 13 13 12 12 12 to 15 15 15 8 8 8.

1 like 13,13,13,13,13,14

just a little bit better than average at everything.

I made a "Noboby" for a one shot.
CHA 14, all 13s.
No defining features, average at everything, indescribably average.
No one even knew what class it was.

[ChildofLuthic]

I advise my players heavily to use point buy and I let them take max HP because I dont want anyone gimped because they rolled bad.

[LaserFace]

I think point buy is good if you want to craft a very specific setting in regards to power-level (I think lower is maybe better for something with a horror subtheme for example), and if you want to ensure your players are all mostly operating on the same level. But in my latest campaign, the players suggested a way to achieve the latter while rolling, instead suggesting that everybody roll their own stats, while also feeling free to just use somebody else's results if yours came out unsatisfactory. Essentially it just set us at a higher point buy, but you get the excitement of rolling, and maybe you get that moment of temptation to take that set that includes two 17s but also a 6.

I have no problems with point buy but I love getting weird outliers, so rolling is just my go-to.

[Merudo]

I'm probably doing something wrong but that calculator lets me use up to 54 points. Even if I limit myself to spending 27 (which isn't something I'd want to police as a DM), I don't seem to be able to make a set that varies significantly from the standard array. I'm not sure that after adding racial adjustments, a point buy PC and a standard array PC would play differently enough for me to notice.


With point-buy you can get these desirable sets:

Half-Elf & Triton with 16/16/16
Half-Elf with 17/16/16 (for Elvish Accuracy)
Variant Human with 16/16/14 or 16/16/16 with half feat.
Mountain Dwarf with 17/17 in Str/Con, to get 18/18 at level 4
Any +2/+1 race with 16/16/14

Nearly every race can benefit from point buy compared to standard array.

I do agree that standard array is the superior case if you find it too complicated to spend only 27 points.

[Pex]

With point-buy you can get these desirable sets:

Half-Elf & Triton with 16/16/16
Variant Human with 16/16/14 or 16/16/16 with half feat.
Mountain Dwarf with 17/17 in Str/Con, to get 18/18 at level 4
Any +2/+1 race with 16/16/14

Nearly every race can benefit from point buy compared to standard array.

I do agree that standard array is the superior case if you find it too complicated to spend only 27 points.

But at what cost to your other stats? One 8 isn't terrible, but two? three? The human gets 16/16/16 but he also has 8/8/8 as Variant, 9/9/9/ as regular. Those -1 modifiers are a big deal because every ability score is a saving throw. You get worse at saving throws as you level since DCs increase while your non-proficient saves do not. It's not as bad with ability (skill) checks, but you're very limited in what you're proficient in you will be making non-proficient checks often. A -1 to about half your skills is bad.

Point Buy as a concept has its own problems. It hurts MAD classes. To get everything the class promotes means you cripple elsewhere. You need to sacrifice a bit of your class power to prevent that. 5E's implementation of Point Buy makes it worse because Bounded Accuracy makes every plus or minus matter. Do what you're supposed to do and be The Suck elsewhere or gimp what you're supposed to do to be less The Suck elsewhere. Point Buy worked in 4E because every class was dual ability dependent, making it easy to choose between specializing in one and be good at both. Meanwhile your saves were a choice between two scores, so you weren't gimped there even if one was low which is fine. Point Buy works in Pathfinder because you start at base 10 for all scores, and the mount of build points you have and costs are reasonable combined with racial ability scores. However, I think the low powered 15 point buy hurts paladins and monks. They're MAD and 15 points is too low to get decent, but they do fine with 20 and 25 points. The point is reinforcing the idea Point Buy as a character creation process is fine for the philosophical reasons of its benefits, but 5E's way of doing it is terrible.

[Jophiel]

Point Buy as a concept has its own problems. It hurts MAD classes.
Being MAD hurts MAD classes. You're statistically unlikely to roll a set of "good" stats for a MAD class unless you start gaming the rolls*, at which point you may as well stop pretending and just give everyone an extra 10 points of Point Buy or start them with 15 15 15 12 12 12 or something.

*The average 4d6 array is 16, 14, 13, 12, 10, 9, almost identical to the standard array. The chance of getting three 15+ rolls with 4d6 is 14.13%, the chance of three 14+ rolls is 36.29%

[MaxWilson]

Being MAD hurts MAD classes. You're statistically unlikely to roll a set of "good" stats for a MAD class unless you start gaming the rolls*, at which point you may as well stop pretending and just give everyone an extra 10 points of Point Buy or start them with 15 15 15 12 12 12 or something.

Or just let them make up their own stats, based on their character concept. I've heard that method suggested before and it's an interesting idea. Has the upsides of point buy (all players on the same footing, can build to concept, works in a low-trust environment like a game with strangers) with none of the downsides of point buy (sameyness, limited range of expressable character concepts, encourages min maxing).

[Jakinbandw]

Or just let them make up their own stats, based on their character concept. I've heard that method suggested before and it's an interesting idea. Has the upsides of point buy (all players on the same footing, can build to concept, works in a low-trust environment like a game with strangers) with none of the downsides of point buy (sameyness, limited range of expressable character concepts, encourages min maxing).

I mean, yeah, I think honestly that this would make more sense than rolling dice.

Anyone who wants to roll dice at my tables can enjoy rolling 1d10+5 for stats. You want variance? You got it without hampering the players that actually want to play the game without being outshone all the time.

[igor140]

Being MAD hurts MAD classes. You're statistically unlikely to roll a set of "good" stats for a MAD class unless you start gaming the rolls*, at which point you may as well stop pretending and just give everyone an extra 10 points of Point Buy or start them with 15 15 15 12 12 12 or something.

*The average 4d6 array is 16, 14, 13, 12, 10, 9, almost identical to the standard array. The chance of getting three 15+ rolls with 4d6 is 14.13%, the chance of three 14+ rolls is 36.29%

I agree with what you're saying in general, but I do have a few "super powerful" characters that I only build when I get those REALLY amazing rolls. Hexblade/ Mystic is INCREDIBLY powerful, but requires high CHA, INT, and CON, and at least decent DEX. Eldritch Knight, Monk/ Arcane Trickster, and Cleric/ Sorcerer are other great combinations that are all MAD. My point is that I often wait to see my rolls before settling on a build; if I get mediocre/ average rolls, I'll build a straight Druid or Rogue. So there is some value in theory-crafting MAD classes, even if you roll for stats. The prime example of this was the 3.5 Paladin, which required... almost everything.

As to the original question, my table does different things, depending on who's DMing. I personally like to roll for stats. Four times out of five, this yields stats that are only slightly higher than standard array/ point buy. Another guy at our table insists on either standard array or point buy (yielding the same overall stats) when he's DM. I get his point about keeping everything even and no one being overpowered, and it's his prerogative as DM, but I think a lot of people overlook the fact that because 5e is so much simpler than 3.5, the attributes are actually fairly unimportant as compared to what they used to be.

Specifically, in 5e, there are (to my knowledge) NO calls to the actual attributes; only the modifiers. That may seem a pedantic thing, but that means that only even attribute values have any actual value. Fair enough. But think of those modifiers as compared to rolling the d20 in the first place. Mathematically, because a d20 has a 5% chance to land on any given side, the hit/ evasion system increases and decreases by increments of 5% (this is slightly incorrect when considering extremely high or low values, and crit successes/ failures, but the basic idea remains true). So if your weapon has a +5 to hit, and the enemy has an AC of 15, you will have the same 55% chance to hit as a +0 attack bonus against an AC of 10; that is, 55% of the possibilities on a d20 will meet the requisite value after adding the attack bonus. In practice this means that if you increase your Strength (for example) by two every fourth level, you've increased your chance to hit by 5%. Now, that's not insignificant (literally; it means meets most measures of statistical significance), but because this is typically your ONLY source of increasing the attack bonus, you spend a lot more effort to increase your chance-to-hit by 5%.

Technically the math on this has always been true since the beginning of DnD, but there are two major differences in 5e versus 3.5 (I know nothing about 4e). The first is the lack of differentiation of BAB (specifically: the lack of BAB at all). This makes the game MUCH easier to play and less tedious to calculate, but it does mean that the AC and relative CR of the enemies scales evenly for all classes. So this means that a wizard is-- in terms of chance-to-hit percentage scaling-- as proficient with a greatsword as a fighter; the only appreciable difference is that the fighter typically has stats that give it a head start on that chance-to-hit relative scaling percentage. We have a guy who build his wizard with a STR of 18 and an INT of 14. He took a feat to be proficient in greatswords, and he outperforms the Paladin in melee DPS.

The second major difference between 5e and 3.5 is the severely reduced number of attacks. Sure, many classes get an extra attack, and a handful (fighter, rogue, monk) have clever ways of stacking that into more than two attacks, but this still pales in comparison to the 20+ attacks per turn from 3.5. Now, I for one am glad that we no longer have to count or add or roll all of those for damage... but what this does mathematically is make each use of the Attack action far more valuable. If three of your twenty attacks (in 3.5) miss, that's no big deal because the other 17 hit and destroyed the target. If three of your three attacks (in 5e) miss, that's a huge deal, because (depending on your class) you've probably had to use a class resource (spell slot, sorcery points, ki points, etc) to attack three times at all.

All this adds up to the fact that a single roll of a d20 has far more of an impact on the final result than the actual modifiers. Due to the fact that average enemy AC scales at approximately the same rate (slightly higher) as PC proficiency bonus, the relative difference between the two-- and therefore the relative 5%-incremented gains and losses to-hit-- make up less than half the variance of actual attack success rates.

tl; dr
Because math, attack success in 5e is based far more on luck than it was in previous versions of DnD. Additionally, there are fewer total attacks per character per turn, meaning that each miss is a bigger deal.

Point in case, I rolled a Paladin (that I intended to multi-class into Sorcerer, though he didn't live that long) that-- after racial bonuses and one feat (variant human)-- had 18/ 14/ 18/ 12/ 14/ 18 as his starting attributes. Those are obviously INCREDIBLY high starting stats... and I was very fortunate... and then he died at level 3. In previous editions (3.5), that character would have been almost invincible. But because an individual d20 roll in 5e counts for a larger percentage of the variance in attack success, I got a few (ok, a LOT) of really bad rolls, and poor Udal got overwhelmed. He had a +6 to attack, but in a 12 round battle, he only once rolled a double-digit value on the d20, so his +6 was useless; and instead of chopping down enemies, he was whittled down himself.

The moral of the story is that the most impressive, seemingly "overpowered" stats don't help as much in 5e as they did in previous versions of the game; the luck of the d20 is MUCH more of a factor. This is why the best builds don't rely on ability scores; they rely on attacking with Advantage or stacking nova damage on successful attacks (this is why, imo, Paladin/ Sorcerer and Hexblade/ Mystic are the two most powerful classes; I believe Rogue/ Monk can ultimately do more damage... but not until at least lvl 17). That being the case, I have less of a problem with people rolling-- or even fudging their rolls-- in 5e.

[Aquillion]

The first is the lack of differentiation of BAB (specifically: the lack of BAB at all). This makes the game MUCH easier to play and less tedious to calculate, but it does mean that the AC and relative CR of the enemies scales evenly for all classes. So this means that a wizard is-- in terms of chance-to-hit percentage scaling-- as proficient with a greatsword as a fighter; the only appreciable difference is that the fighter typically has stats that give it a head start on that chance-to-hit relative scaling percentage. We have a guy who build his wizard with a STR of 18 and an INT of 14. He took a feat to be proficient in greatswords, and he outperforms the Paladin in melee DPS.

I know this wasn't your main point, but I would argue that that's a good thing. It makes more builds viable, and it increases the chance that if someone gets their heart set on something weird or builds their character strangely, it's still likely to be viable (like, the wizard who maxes out strength in your example would have been totally useless in prior editions. In 5e it is probably still not an optimal way to build a wizard, but you can still contribute meaningfully and have fun with it.)

5e seems generally structured to have as few "trap" options as possible and to try and ensure that most builds achieve at least some bare minimum of playability. It's possible to totally screw things up, sure, but you usually have to actually try, whereas in earlier editions there were totally-reasonably-seeming things that were actually traps and would result in a nearly unplayable build (ie. putting half your levels in a casting class and half your levels in something that doesn't advance your casting in order to build a gish.)

[Coffee_Dragon]

I agree with what you're saying in general, but I do have a few "super powerful" characters that I only build when I get those REALLY amazing rolls. Hexblade/ Mystic is INCREDIBLY powerful, but requires high CHA, INT, and CON, and at least decent DEX.

And with the standard array's 16, 14, 14, 12 to those four stats it just falls apart? Seems like there should be a gap between trash and "super powerful" where the system is working as intended.

[Luccan]

And with the standard array's 16, 14, 14, 12 to those four stats it just falls apart? Seems like there should be a gap between trash and "super powerful" where the system is working as intended.

I've heard it suggested the real average baseline in D&D would be 8-12; we all know someone who is a little better or worse at some things than the average person, but not so much they're astounding/can't function.

Under that perception, a 12 in dex isn't necessarily in the "decent range". And I'd certainly argue regardless that a 14 in a stat isn't particularly high. You're beating out a person with a 10 by quite a bit, yes, but operating under the assumption a MAD build essentially has multiple stats that must all be treated as a main stat (as opposed to say, a Wizard who could just boost Int), I'd say you need at least a 16 in most of them. You don't get enough ASIs otherwise.

[Foxhound438]

Being MAD hurts MAD classes. You're statistically unlikely to roll a set of "good" stats for a MAD class unless you start gaming the rolls*, at which point you may as well stop pretending and just give everyone an extra 10 points of Point Buy or start them with 15 15 15 12 12 12 or something.

*The average 4d6 array is 16, 14, 13, 12, 10, 9, almost identical to the standard array. The chance of getting three 15+ rolls with 4d6 is 14.13%, the chance of three 14+ rolls is 36.29%

I agree with you here. I love playing paladins and monks, and usually start thinking about what I want to play in my next game before session zero even happens. Often times, I end up landing on wanting to play one of those two. But when I roll stats and get like one 14 and the rest somewhere between 7 and 12, I really can't play either of those and be effective at anything. And at that point, I would really have to consider whether I even want to show up to a 3 hour session of not being able to do anything. That's just not fun to me. On the other hand, point buy allows me to play those classes to a very reasonable degree of effectiveness without having to drop a wizard off of a 20 foot ladder at level one until the stat rolls work out.

[Stygofthedump]

I think the 4d6 drop low is overpowered for the system, we are now using dirty old 3d6 rolled stats and man it’s great!

[Astofel]

I've never found rolling stats to be "overpowered", even when my players do roll well. Sure, it can certainly be better than point buy, but it's not game-breaking. If I really need to as a DM it's not difficult to just up encounter difficulty a touch. I don't worry about players with better rolls overshadowing players with weaker rolls either. A fighter with 20 Str might have better stats than a rogue with 16 Dex, but they play differently enough and fulfill different roles so no matter how strong the fighter is he doesn't step on the rogue's toes at all.

[Glorthindel]

I consider point buy the worst of the options. Now don't get me wrong, it is probably the fairest option, but its problem is it places the blame for stat distribution on to the player, and opens up the possibility of friction between players.

In the case of both rolling and standard array, you have your available stats, and it is just down to the player to assign them, and attempt to optimise them (or not) in a limited way using racials. But with point buy, you custom build every stat value, so you are to "blame" if you don't dump your dump stats as low as possible. It creates potential places of friction between optimisers, and those less inclined that way, because the player could be seen as "wasting points". This doesn't happen with the other methods, since the numbers you have to use are fixed - you can't be berated for putting a 10 in a dump stat if that is the lowest value you have available, but in point buy, you could easily be berated for wasting points that are better spent on your primary or secondary stat.

[EggKookoo]

I do agree that standard array is the superior case if you find it too complicated to spend only 27 points.

It's not about difficulty. It's about getting value out of the process.


I think the 4d6 drop low is overpowered for the system, we are now using dirty old 3d6 rolled stats and man it’s great!

I've been considering trying that. 4d6 drop the lowest is a holdover from back before the days of ASIs and all races providing bonuses, instead of some plus and some minus.

[Lyracian]

Anyone who wants to roll dice at my tables can enjoy rolling 1d10+5 for stats. You want variance? You got it without hampering the players that actually want to play the game without being outshone all the time.
That is pretty harsh. I only need to roll better than 10,9,8,7,5,2 to get above average!
https://www.random.org/integers/?num=6&min=1&max=10&col=6&base=10&format=html&rnd=new
First attempt = 13 15 12 12 8 13


We have been playing some retro First Edition DnD which has a rule of "Hopeless Characters" on page 13 allowing players to roll a new character if the DM thinks they are. New player rolled up a character with a pair of 9's has his highest scores and got told it was not hopeless enough!

Point Buy means everyone is build from the same base line. Some characters thought, like Dwarven Mages, really benefit from Rolling unless you are happy starting with a 15 Int. I think it become a bit more difficult when you are mixing players some with point buy and others with rolled stats.

[EggKookoo]

Another solution to poor rolling is to have everyone roll up their scores, and then if your combined modifiers don't amount to at least +1 you can increment up your lowest score until they do.

But really I just let my players toss out a set they don't like and roll up a new one. I don't put limits on how often they can do this and it never goes beyond 2-3 times. It works out fine, everyone ends up with something that makes them happy, and with ASIs and racials they feel like they can self-buff.

[Chunkosaurus]

I like My PCs to be good at stuff, so I prefer rolling. I actually allow each PC to roll three stat sets of 4d6 drop lowest and then pick the stat set they want for their character and then get a feat at level 1. This makes everyones stats pretty high and lets them take more feats rather than I add two to my primary every 4 levels. I would rather my PCs be strong than weak. In return I hammer them with brutal encounters, but they can take it because they are strong. They have come close to TPKing a few times but no one has ever managed to die because they made the one or two crucial rolls.

That being said next time I'm going to use a slightly different method that eliminates interparty variance. Everyone will roll 1 set of 4d6 drop the lowest and then we will vote on what stat set to use as a group, so everyone has the same stats distributed differently. This allows people to play something MAD and take more feats than ASIs without ever feeling like they arent contributing.

[MThurston]

Some types of stat building

1. Roll 3d6 and hope for the best.

2. Roll xd6 and they go in the order of stats. Your stats build your class.

3. Roll 4d6 take the highest 3. Total stats 64+ points or you can roll again.

4. Roll 4d6 take the highest three. No rerolls of stats.

5. Standard array. People may find boring while DMs find none optimized.

6. Point buy. Same build number as Standard array but you may find min/max.

An Example of that is three stats at 8 and three stats at 15.

My order 3, 6, 5, 4, 1

As for 2, I don't mind.

[Jophiel]

I agree with you here. I love playing paladins and monks, and usually start thinking about what I want to play in my next game before session zero even happens.
This isn't a new problem. The chance to roll (https://muleabides.wordpress.com/2012/02/11/you-must-be-this-lucky-to-play/) a 1e Paladin using 3d6 is 0.10% -- granted the DMG offered alternate methods of generation (including the 4d6 method) but most people were stuck on 3d6 in order as "official". I think we would let you roll your six stats to place where you wanted, except that you had to use your highest numbers to meet the requirements; if you rolled a 15, you had to make that your 17 in Charisma.


But with point buy, you custom build every stat value, so you are to "blame" if you don't dump your dump stats as low as possible. It creates potential places of friction between optimisers, and those less inclined that way, because the player could be seen as "wasting points". This doesn't happen with the other methods, since the numbers you have to use are fixed - you can't be berated for putting a 10 in a dump stat if that is the lowest value you have available, but in point buy, you could easily be berated for wasting points that are better spent on your primary or secondary stat.
This honestly sounds more like a terrible table problem than a poor stat generation problem. If someone at our tables started berating people about their stats and "doing it wrong", they'd be finding a new place to play pretty quickly.

[Willie the Duck]

I've never found rolling stats to be "overpowered", even when my players do roll well. Sure, it can certainly be better than point buy, but it's not game-breaking. If I really need to as a DM it's not difficult to just up encounter difficulty a touch. I don't worry about players with better rolls overshadowing players with weaker rolls either. A fighter with 20 Str might have better stats than a rogue with 16 Dex, but they play differently enough and fulfill different roles so no matter how strong the fighter is he doesn't step on the rogue's toes at all.

I will certainly note that, in a game with power disparities such as the difference between playing with or without feats (particularly for classes like fighters), with or without multiclassing (particularly for dips, or things like sorlocks and socadins), and one person playing a PHB single class Dex-bladelock and the other person playing a hexblade, whether one does point buy or 4D6b3 is certainly not the highest point on my imbalance-meter.

[Chunkosaurus]

I will certainly note that, in a game with power disparities such as the difference between playing with or without feats (particularly for classes like fighters), with or without multiclassing (particularly for dips, or things like sorlocks and socadins), and one person playing a PHB single class Dex-bladelock and the other person playing a hexblade, whether one does point buy or 4D6b3 is certainly not the highest point on my imbalance-meter.

I agree and my players aren't super powergamey.

[igor140]

I know this wasn't your main point, but I would argue that that's a good thing. It makes more builds viable, and it increases the chance that if someone gets their heart set on something weird or builds their character strangely, it's still likely to be viable (like, the wizard who maxes out strength in your example would have been totally useless in prior editions. In 5e it is probably still not an optimal way to build a wizard, but you can still contribute meaningfully and have fun with it.)

5e seems generally structured to have as few "trap" options as possible and to try and ensure that most builds achieve at least some bare minimum of playability. It's possible to totally screw things up, sure, but you usually have to actually try, whereas in earlier editions there were totally-reasonably-seeming things that were actually traps and would result in a nearly unplayable build (ie. putting half your levels in a casting class and half your levels in something that doesn't advance your casting in order to build a gish.)

Oh, I agree with you that it's a good thing; my only point with this is that the actual ability scores (and therefore whether rolling for stats yields 30 points to spend as compared to point-buy's 27) is far less important than it was in previous versions.

The biggest place I see this is in the social encounters. I REALLY like both Warlocks and Paladins, so I often run high CHA characters, so I frequently have lvl 5 characters with +9 to Persuasion or Intimidation... and they frequently fail DC checks. Conversely, the half-wit Barbarian with the table manners of a rabid squirrel passes the Persuasion check because he got lucky on the d20.

I don't see this as a problem at all at early levels; who lives and who dies IS often left up to chance.

The potential problem I see is that when you have a lvl 19 demi-god who wakes up every morning to a breakfast of fresh dragon eggs, your success in battle (or social situations) is still more dependent on the luck of the d20 than it is on the character you've built. Again, I generally think that this is better, but I'm looking at homebrewing a system in which your Ability Score Modifiers become more potent at higher levels. If I my character survives literal (real life) years worth of campaigning, fighting, escaping, finagling, party dischord, dissolved tables, and all the other things that happen to DnD groups in three years... I expect him to connect with that damn sword the vast majority of the time; even against difficult enemies.

[MaxWilson]

I will certainly note that, in a game with power disparities such as the difference between playing with or without feats (particularly for classes like fighters), with or without multiclassing (particularly for dips, or things like sorlocks and socadins), and one person playing a PHB single class Dex-bladelock and the other person playing a hexblade, whether one does point buy or 4D6b3 is certainly not the highest point on my imbalance-meter.

I think it's more about feelings than mechanics. Some people pay a lot of attention to the numbers written on the character sheet, and will feel second-rate just having a 14 where someone else has an 18, even if the guy with an 18 is a TWF ranger and the guy with a 14 is a Sharpshooter Battlemaster who winds up doing more damage than the TWF guy on a regular basis. I think it has something to do with the knowledge that the 18 guy theoretically could have been a better Sharpshooter than you if he had made different choices.

Many people aren't very good at perceiving actual balance in play, as you can see from the occasional complaints of weakness from someone who took PAM/GWM while someone else boosted their Str to 20. The PAM/GWM guy complains that he is "weak" because he "always misses" whereas the Str 20 guy is hitting reliably, but we know from the actual math that the extra attacks and damage from PAM/GWM will more than make up for it unless the DM is doing something really strange like using exclusively AC 19+ enemies. (Str 20 guy will hit on a roll that PAM/GWM guy would have missed on 10% of all d20 rolls, but PAM/GWM guy is getting one or two extra attacks every single round, depending.)

Likewise the people who say that point buy eliminates imbalance cannot possibly mean that it makes everyone equally effective during play, so they must mean it eliminates the unpleasant feeling for them.

[Morty]

We did point buy in the one 5E game I've played and I see no reason not to do it. I don't find attributes exciting or interesting in any way, so I just want to get them on the sheet with minimum fuss. They're an illusion of choice either way.

I also agree that simply assigning the attributes you want, with the GM making sure you're not going overboard, accomplishes everything rolling does but better. Or you can just put an 18 in your class's key attribute and then roll.

[MaxWilson]

I also agree that simply assigning the attributes you want, with the GM making sure you're not going overboard, accomplishes everything rolling does but better. Or you can just put an 18 in your class's key attribute and then roll.

It accomplishes everything point buy does but better. It lacks some of the variance of rolling though--though of course nothing is stopping you from rolling anyway and assigning those numbers.

[Willie the Duck]

We did point buy in the one 5E game I've played and I see no reason not to do it. I don't find attributes exciting or interesting in any way, so I just want to get them on the sheet with minimum fuss. They're an illusion of choice either way.

I also agree that simply assigning the attributes you want, with the GM making sure you're not going overboard, accomplishes everything rolling does but better. Or you can just put an 18 in your class's key attribute and then roll.

If you don't find attributes interesting, assignment with GM check works fine. There is a level of play (often advocated by osr crowds) of enjoying the challenge of making something fun/useful out of what the dice give you that only some level of randomness will provide. Having also enjoyed the heck out of Traveller character creation, I certainly see the value, although it is something of a niche appeal at this point.

[EggKookoo]

We did point buy in the one 5E game I've played and I see no reason not to do it. I don't find attributes exciting or interesting in any way, so I just want to get them on the sheet with minimum fuss. They're an illusion of choice either way.

This is why I prefer the standard array to point buy. It literally is the most level you can get that playing field.

[Tanarii]

IMX point buy is used pretty extensively in 5e. But that's because my experience outside of my own campaign is AL, and hearsay on the forums. And these forums definitely don't represent a typical slice of 5e players.

I suspect that the game defaults (choice of rolling or standard array) are far more common than the optional rule (point buy)when it comes to home games. Mostly because my experience across many RPGs is that given the choice, players typically want to roll dice for ability scores. But also because they are presented as the defaults.

[Cynthaer]

The issue with that is that that hasn't changed. Certainly, official modules shifted towards not killing players. But that's been the case as far back as AD&D 2nd Edition. I played a game of 5e where only the Wizard managed to survive to the end of the campaign. With everyone else going through several characters, and the only reason the Wizard survived was a combat doctrine of 'Run away if it looks like you might get attacked.' and 'Zap it from VERY FAR AWAY.'

At the risk of reopening and old debate, I believe that between AD&D and 5e, both player and designer expectations have shifted much further toward "build a full backstory and expect character death to be uncommon", and away from "just slap a name on that sheet and develop a personality if they manage to survive". Individual tables, as always, will vary drastically.

THAT SAID, I think my core analysis still holds up either way:


Power disparities are fun when they are temporary and players go through multiple characters over time. Power disparities are not fun when they are permanent and players only have one character for the entire campaign.

Whether we're playing AD&D or 5e, I'd say rolling individual stats is generally more fun if you expect a bunch of character deaths, and point buy (or some other method that minimizes power disparities) is generally more fun if you expect to play just the one character for the entire campaign.

If you don't know what the expected level of lethality is for your campaign? Well, then I don't know which approach would generally be more fun.

And of course some individuals prefer one approach so strongly that it doesn't matter what kind of campaign they're in. Nobody's personal preferences are wrong. I'm just talking about broad game design.

[MaxWilson]

IMX point buy is used pretty extensively in 5e. But that's because my experience outside of my own campaign is AL, and hearsay on the forums. And these forums definitely don't represent a typical slice of 5e players.

I suspect that the game defaults (choice of rolling or standard array) are far more common than the optional rule (point buy)when it comes to home games. Mostly because my experience across many RPGs is that given the choice, players typically want to roll dice for ability scores. But also because they are presented as the defaults.

And even more because point buy is presented as a complicated variant. Newbies will steer clear just because of all the addition and subtraction and table lookups needed.


Whether we're playing AD&D or 5e, I'd say rolling individual stats is generally more fun if you expect a bunch of character deaths, and point buy (or some other method that minimizes power disparities) is generally more fun if you expect to play just the one character for the entire campaign.

Yes, that seems to be true, anecdotally. People who play one PC for years on end (at one extreme) seem to care a lot about getting to tailor their character to their concept instead of tailoring their concept to their random stat rolls; and sameyness of course isn't an issue if you are only ever going to play this one PC.

[Max_Killjoy]

If you don't find attributes interesting, assignment with GM check works fine. There is a level of play (often advocated by osr crowds) of enjoying the challenge of making something fun/useful out of what the dice give you that only some level of randomness will provide. Having also enjoyed the heck out of Traveller character creation, I certainly see the value, although it is something of a niche appeal at this point.


On the flip side, I've had the most fun in systems ranging from HERO, to oWoD, to WEG d6 Star Wars, and the thing they have in common (in the context of this discussion) is that they're "point-buy" of some sort. That is, they have fixed character creation resources that are used by the player to come as close to the character in their head, within the confines of the "power limit" for the starting PCs, as they can.

So as I'm trying to figure out 5e, and build some characters, it's a bit frustrating that there's nothing I can "trade off" to get better Ability Scores, or more Skills, or proficiency in more Saves, or...

[MaxWilson]

On the flip side, I've had the most fun in systems ranging from HERO, to oWoD, to WEG d6 Star Wars, and the thing they have in common (in the context of this discussion) is that they're "point-buy" of some sort. That is, they have fixed character creation resources that are used by the player to come as close to the character in their head, within the confines of the "power limit" for the starting PCs, as they can.

This is a bit of a tangent, but speaking of point buy systems...

While I really enjoy GURPS the game, I hate the GURPS chargen system because while in theory equally-strong abilities have equal costs, in practice the costs are so arbitrary and easy to game that the whole system relies entirely on the DM gating overpowered combinations ad hoc behind Unusual Background costs. Otherwise you wind up with characters who are able to destroy anything on a 50-point budget via shotgun multiattacks or untraceable assassination specials, etc.

One of the things that brought me back eventually to the D&D family is realizing that I prefer to have some structure on how you can spend your "points," to be "forced" to invest in diversification instead of hyperspecialization... and that is what a class system is. GURPS: Dungeon Fantasy basically has re-invented classes via templates--the only thing lacking is a defined progression order that says e.g. "you can't take raise your Broadsword skill above +4 until you have at least 100 points total, and then it can go up to (points/25)."

[Max_Killjoy]

This is a bit of a tangent, but speaking of point buy systems...

While I really enjoy GURPS the game, I hate the GURPS chargen system because while in theory equally-strong abilities have equal costs, in practice the costs are so arbitrary and easy to game that the whole system relies entirely on the DM gating overpowered combinations ad hoc behind Unusual Background costs. Otherwise you wind up with characters who are able to destroy anything on a 50-point budget via shotgun multiattacks or untraceable assassination specials, etc.

One of the things that brought me back eventually to the D&D family is realizing that I prefer to have some structure on how you can spend your "points," to be "forced" to invest in diversification instead of hyperspecialization... and that is what a class system is. GURPS: Dungeon Fantasy basically has re-invented classes via templates--the only thing lacking is a defined progression order that says e.g. "you can't take raise your Broadsword skill above +4 until you have at least 100 points total, and then it can go up to (points/25)."

That's fair, the more open a point-buy system is, the more player (GM and otherwise) discretion it requires. Freedom has a cost, so to speak.

I would point out that not all point-buy systems are created equal. GURPs is rather notorious for what you describe there, but I can't say how fair it is. HERO gets a bad rap for it both because too many groups seem to skip the character creation "soft rules" parts of the book, or just ignore them... and then blame the system for their own failure to heed the advice that's right there in the book.

There's also a perspective issue... because I tend to make very well-rounded, multi-competent, jack-of-many-trades characters, I get the opposite experience of what you're describing. That is, 5e feels MORE specialized and narrow-focused to me, and there are a lot of character concepts that I'm struggling to build without having to twist and distort them to fit 5e.

[Dragons_Ire]

Here's my experience:

My first character, in an (ongoing) Adventures in Middle Earth campaign was rolled using 4d6b3, and I got a really bad set - 13/13/13/11/8/7, I think. My friends told me to reroll, but I stubbornly insisted on playing what I rolled, lol. A couple sessions later I asked to switch to standard array and was allowed to respect my character (8 dex Dwarven Scholar sounds fitting but was mechanically horrible).

My second character, in a standard (ongoing) D&D 5e campaign was rolled the same way. I got 17/15/12/11/11/8, and am very happy with it.

At my table I use 4d6b3, assign as desired, if you don't like your set you may make an NPC using the set (for my potential future use) and then roll again, no limit. It works pretty well.

[MaxWilson]

At my table I use 4d6b3, assign as desired, if you don't like your set you may make an NPC using the set (for my potential future use) and then roll again, no limit. It works pretty well.

This part here is important IMO. I do this too*, and I think it helps prevent stat inflation.

* With a twist--you also have to roll up five NPCs on 3d6 and donate them to me too. It's not a ton of work but it helps remind you what the "average" person is like.

[Coffee_Dragon]

When I was a kid the reroll tax was one gnoll outpost, three Sigil districts and one dwarven culture.

[Willie the Duck]

When I was a kid the reroll tax was one gnoll outpost, three Sigil districts and one dwarven culture.

Now that was funny, although it made me feel so old.

[LordCdrMilitant]

I use Points Buy nearly exclusively.

It occurred to me one day that rolling is pretty much no different from choosing the stats you want sans limitation. Between the ever-increasing amount of fixing for the die rolls [At one time, it was 3d6, when I started it was 4d6B3, then it was 4d6b3 re-roll 1's, I've seen generate 7 with an aforementioned methods and take the 6 best, and I've seen people who generate 2+ complete statblocks and chose the one they want] and then the petitions for adjustments, it occurred to me that rolling the die was basically the same as selecting your stats without the cap of 15.

I also performed a limited experiment, in which I asked my players to generate balanced character stats however they wanted; and [after the players got over the inherent difficulty of actually being free to chose what stats they wanted] I noticed it looked almost exactly like the stats that were produced by rolling, [even more so when they weren't rolled in my presence]. More importantly, I noticed all such characters started with at least one 16+ stat [before racial modifiers].

Thus, it occurred to me, that the real "purpose" of rolling up stats [for 5e] was to generate characters who were able to get a stat to 20 by level 4 or earlier to reduce the number of ASI's, maximize early power, and increase the number of feats taken.



As such, I switched to points buy and have been using it in most games since. I've had some serious pushback from players not in my games [and a heated discussion with another person about build optimization where they considered statblocks to start with an 18 while I consider them to start at 15], but overall my players don't mind it.

There has been one observation, though, from Dark Heresy: A player observed that, when we used Points Buy for their stats, permanent stat damage was far more punishing [because getting any points of permanent damage nearly doubles the XP cost to increase the modifier, and permanently drops your maximum modifier by a point], and more importantly, everybody ended up dumping the same stats to 25/20, and so nobody can pass any sort of INT or FEL based checks. This is less relevant for 5e, though, because IIRC stat damage isn't permanent [or common], and there are classes that require social/intelligence skills for their offensive power.

There is an argument for 5e, that there's merit to starting with a 16+ in a stat, which was brought up by a different player. Notably, with a 15 max score, Humans are very good; because they get a 16, 14, 14, 12, 10, 8 or some such statline, plus a feat and a skill. 2 ASI's to your 20, and you're basically ahead of everybody else, who still need 2 ASI's to a 20 [and either waste half their ASI or get a weaker feat] and don't get a starting skill and feat. However, with a 16 [or 18] starting score, nonhumans outclass humans, being able to start with a +4 [or even +5] modifier, and being able to take a feat right away while a Human will still need 2 ASI's to go from 17 to 20 [or 1 to go from 19 to 20]. It's not going to make me go back to rolling [in part because nobody but me seems to want to play humans anyway], but it's an idea if you want less humans in your games.

[Pex]

Being MAD hurts MAD classes. You're statistically unlikely to roll a set of "good" stats for a MAD class unless you start gaming the rolls*, at which point you may as well stop pretending and just give everyone an extra 10 points of Point Buy or start them with 15 15 15 12 12 12 or something.

*The average 4d6 array is 16, 14, 13, 12, 10, 9, almost identical to the standard array. The chance of getting three 15+ rolls with 4d6 is 14.13%, the chance of three 14+ rolls is 36.29%

Rolling for stats isn't perfect either and people have done more than just 4d6 drop lowest, but I think it undermines your point as you admit even the standard array is poorer than the average 4d6 array. Note that Pathfinder has you start with all 10s, so yes, it is possible and not unheard of for Point Buy to help get you a decent array for MAD classes. I already said as such. The problem is not Point Buy. The problem is how 5E implements it.

[Jophiel]

Rolling for stats isn't perfect either and people have done more than just 4d6 drop lowest, but I think it undermines your point as you admit even the standard array is poorer than the average 4d6 array.
Well, my argument is that the issue with MAD isn't point-buy, it's MAD. Having two extra points doesn't make up for the fact that you're still going to be lacking in some of your necessary stats and your other stats are going to be average or suboptimal. Especially considering that only the 15 (Std Array) to 16 (4d6 Array) matters and the 8 to 9 is meaningless in almost all cases since you're unlikely to waste an ASI/Half-Feat on pumping that to 10 on a MAD class. Point being, they're not much different so it's hard to pin the issues with MAD on them. Rolling isn't going to do you much better as an alternative unless you're faking the dice rolls by adding a bunch of gimmes. In theory, Rolling would do you worse around 50% of the time but I doubt anyone would let those rolls stand.

[Pex]

Well, my argument is that the issue with MAD isn't point-buy, it's MAD. Having two extra points doesn't make up for the fact that you're still going to be lacking in some of your necessary stats and your other stats are going to be average or suboptimal. Especially considering that only the 15 (Std Array) to 16 (4d6 Array) matters and the 8 to 9 is meaningless in almost all cases since you're unlikely to waste an ASI/Half-Feat on pumping that to 10 on a MAD class. Point being, they're not much different so it's hard to pin the issues with MAD on them. Rolling isn't going to do you much better as an alternative unless you're faking the dice rolls by adding a bunch of gimmes. In theory, Rolling would do you worse around 50% of the time but I doubt anyone would let those rolls stand.

Yes and no. Dice rolling doesn't guarantee an array suitable for a MAD class, but it easier to do so than through 5E's implementation. You're locked in either playing a Human or only the race that's suitable for it. Yes, you'll find the player who's a halfling barbarian or dragonborn monk, but it's so rare you can safely say it doesn't happen and be generally correct. When you are fortunate enough with a good rolled array such combinations become interesting choices.

In any case the MAD classes exist. Maybe in hypothetical 6E they'll fix it, but with what we have now 5E's implementation of Point Buy does not do the MAD classes any favors.

[Tanarii]

Yes, you'll find the player who's a halfling barbarian or dragonborn monk, but it's so rare you can safely say it doesn't happen and be generally correct.
God knows why there aren't more Halfing Barbarian Sailors out there. Who doesn't want to play a pint-sized Viking? :smallbiggrin:

[Max_Killjoy]

Yes and no. Dice rolling doesn't guarantee an array suitable for a MAD class, but it easier to do so than through 5E's implementation. You're locked in either playing a Human or only the race that's suitable for it. Yes, you'll find the player who's a halfling barbarian or dragonborn monk, but it's so rare you can safely say it doesn't happen and be generally correct. When you are fortunate enough with a good rolled array such combinations become interesting choices.


If I ever end up in a 5e campaign, I want to play a shadow-dragonborn shadow monk. :smallbiggrin:

[Maelynn]

God knows why there aren't more Halfing Barbarian Sailors out there. Who doesn't want to play a pint-sized Viking? :smallbiggrin:

I'm just politely informing you that your idea is hereby stolen.

[Theodoric]

In my group everyone uses the standard array. Dice rolls feel a bit too random (I'd do it, but I'm the DM) and point-buy is a bit too power-game-y. Having one or two characters with min-maxed stats while the rest has more statistically normal stats would be a bit awkward at the table.

Characters in 5e don't even need to have all feature-related stats to be high to be competent, anyway. One of my players has a Ranger who started with a Wis of 13 and an Int of 16. Bit weird, but he's contributing very well so far.

[Skylivedk]

Point buy, but we've gone with a variant in one group (the most houseruled).

Normal point buy +1 stat point to whichever score you'd like and a made up feat resembling actor.

We've also tried a variation of rolled stats where each player rolled 4d6b3 and then players could allocate the stats afterwards. I care mostly about balance in between players since I either create campaigns from scratch or modify published adventures to have more fun encounters anyway. Also, only 1 out of 3 groups is RAWish. The two others run with a completely revamped feat and magic item system.

[MThurston]

I don't think DMs want the start of the campaign to begin with optimizing character creation.

The easiest way is standard array. People still optimize with it however.

If you wanted to really do some fun character creation, you should have people pick their race first.

Then roll or randomly pick a standard array number to fill in their stats.

Say I pick Mountain Dwarf, I want to play a Barbarian.

I put scrabble letters in a bag and pick them out one by one.

A-8, B-10, C-12, D-13, E-14, F-15

I pick C, F, A, D, E, B.

I end up with Str 14, Dex 15, Con 10, Int 13, Wis 14, Cha 10

Now you pick your class and people really have to think about what class to play.

You could also pick the order first and then let them pick race.

But the above could make for some great Role-playing and not just murder hobo.

[Tanarii]

I'm just politely informing you that your idea is hereby stolen.
Fair is fair. I stole it from Dungeonomics. :smallamused:

[Aett_Thorn]

So I know I'm a little late to the party here, but I have been trying to follow this thread as it goes along. But please forgive me if I post something that's already been covered.

I like point buy in that it makes people make conscious choices for their characters, while also making sure that characters are roughly even power-wise. However, Having played a couple of games with 4d6b3, and a couple point buy, I do notice that there were more interesting race/class choices in the 4d6b3 campaigns than there are in the point buy ones. 4d6b3, I feel, allows you to get a good score in a stat even if it's not your races "thing". I know that the difference between starting at a 15 vs a 16 or 17 is only +1, but it can feel like you're starting off at a disadvantage, which can impact your decisionmaking.

For instance, in one game with 4d6b3, we had:

Bugbear Bard
Dragonborn Wizard
Tiefling Fighter
Half-Elf Sorc
Dwarven Cleric
Tabaxi Rogue

First three were definitely suboptimal, bottom three were pretty typical choices. But nobody felt over/underpowered compared to the rest. In the new campaign, we have:

Goliath Barbarian
Halfling Dex Barbarian
Hobgoblin Rogue
Human Cleric
Half-Elf Cleric
Half-Elf Warlock
Dwarf Paladin

Only the Hobgoblin Rogue is really against type here. Sure, if you have a table full or powergamers, rolling can lead to wild imbalances between characters. But if you're mainly there for the RP, rolling can allow you to go against race types a bit easier while still being competitive, too, depending on your rolls.

[Zanthy1]

When I DM I exclusively do point buy. This is largely due to not being near any of my players (most of our games take place online, or on a weekend when I visit). This way I know that everything is kosher and that all characters have a similar base. The issue I had when I allowed rolling for stats was that I would have a variety of characters in the game. Some would have crazy highs and others would have essentially the normal spreads, every now and then something bad. I also had some players who would fudge rolls, and point buy counters that.

[Jophiel]

I will say that I used to roll my eyes at the "What class should I play, I rolled a 18 17 16 16 14 12 at my table" type posts but, after hearing how many tables rig the system towards higher results when rolling, I should give them the benefit of the doubt.

[MaxWilson]

I will say that I used to roll my eyes at the "What class should I play, I rolled a 18 17 16 16 14 12 at my table" type posts but, after hearing how many tables rig the system towards higher results when rolling, I should give them the benefit of the doubt.

There's also selection bias to consider. 15 15 15 8 8 8 is a mechanically-powerful array, but no one gets excited and makes a forum post about how not to waste it because arrays like that are also *ubiquitous* under point buy. If you mess up, you can always just make another PC with the same stats and try again.

But 18 17 16 16 14 12 is maybe your one-and-only chance to play a dragonborn warbearian. Str 20 Dex 16 Con 16 Int 14 Wis 12 Cha 18. You'd have to roll another hundred characters to see another array like it. It might be YEARS or never.

There's a forum thread right now on "how do I spend a hundred and fifty million gp reward I'm expecting" and there are zero threads on "how do I spend a hundred gp." It's not because millions of gp is a common reward; it's because it's rare and exciting to that player.

[MThurston]

I will say that I used to roll my eyes at the "What class should I play, I rolled a 18 17 16 16 14 12 at my table" type posts but, after hearing how many tables rig the system towards higher results when rolling, I should give them the benefit of the doubt.

What do you mean by rigging the system?

3d6 is an average of 10.

4d6 pick 3 makes that average 11.

I am basing this this on the following.

4, 3, 3

4, 3, 3, 4

Not on a real formula.

As for your post, I have the same thinking as you but when you are asked to roll 3d6. You have a 6, and a 9 on your stat line and the DMs best friend has two 18s and a 17 with nothing under 10.

[Temperjoke]

Admittedly, I don't have a lot of experience when it comes to point buy, but I think it takes a lot more mental work to create a character that way, at least for me. But from a numbers stand point I don't think it actually makes much of a difference, unless you're comparing the extreme possibilities you can get from rolling.

[MThurston]

Admittedly, I don't have a lot of experience when it comes to point buy, but I think it takes a lot more mental work to create a character that way, at least for me. But from a numbers stand point I don't think it actually makes much of a difference, unless you're comparing the extreme possibilities you can get from rolling.
The big difference as I see it between Standard and point buy is this.

Both can be optimized but with point buy you will see many optimizers as so: Str 15, Dex 15, Con 15, Int 8, Wis 8, Cha 8 for a Warrior Class and Str 8, Dex 15, Con 15, primary spell stat 15 with the other two at 8 for a magic user.

Paladins end up with Str 15, Dex 10, Con 14, Int 8, Wis 8, Cha 15.

I do think rolling is better for randomness. Making them take it in order can get people to play things they haven't before.

I know some people that only play one or two classes.

[Jophiel]

What do you mean by rigging the system?
There's a bunch of posts talking about certain tables allow you to roll multiple arrays and pick one, re-roll any "1"s on your dice, use any array rolled at the table, re-roll if your stats don't total X many points, keep rolling until you get one you like, etc. Seemingly, very few just roll 4d6 and let the results be what they are. I'm not passing judgment on that -- I don't play at your table so what do I care? -- just saying that some of the improbable stats I see in threads are more probable when you're using additional rules to help favor high results.

There's also selection bias to consider. 15 15 15 8 8 8 is a mechanically-powerful array, but no one gets excited and makes a forum post about how not to waste it because arrays like that are also *ubiquitous* under point buy.
You could be right. My own perspective would be "The answer to 'What can I do with these great scores' is 'Anything, duh'" so I wouldn't consider asking a question unless I was just trying to brag. But if I had mediocre scores, I might ask for advice on how to make them work. To reverse the gold example; I wouldn't ask "How can I gear my character for 10,000gp?" but I might ask "What's the most effective way to spend 15gp on gear?". But, again, that's just me and you make a valid point.

[CantigThimble]

I do think rolling is better for randomness. Making them take it in order can get people to play things they haven't before.

I know some people that only play one or two classes.

Why is this a problem? Why not just let people play what they like to play?

I do enjoy playing very random characters and making odd arrays work, but plenty of my friends don't and have more fun when they have more control over who their character is. I'm not going to take that away from them just because I enjoy doing it differently.

[Skylivedk]

Point buy hasn't lead to optimization in our campaigns. Latest point buy campaign includes a Wizard and a Fighter both with 12 or more in charisma. Our other campaign has people dumping wisdom before int for RP reasons etc.

Optimization is not a result of your stat allocation choice, but of your player choice (unless you play with stats being rolled from the top down, rather than rolled and then allocated).

[MaxWilson]

You could be right. My own perspective would be "The answer to 'What can I do with these great scores' is 'Anything, duh'" so I wouldn't consider asking a question unless I was just trying to brag. But if I had mediocre scores, I might ask for advice on how to make them work. To reverse the gold example; I wouldn't ask "How can I gear my character for 10,000gp?" but I might ask "What's the most effective way to spend 15gp on gear?". But, again, that's just me and you make a valid point.

I can relate to that. I'd find your hypothetical 15 gp thread more interesting than the millions-of-gp thread. More likely to be relevant to my actual game.

[LordCdrMilitant]

Point buy hasn't lead to optimization in our campaigns. Latest point buy campaign includes a Wizard and a Fighter both with 12 or more in charisma. Our other campaign has people dumping wisdom before int for RP reasons etc.

Optimization is not a result of your stat allocation choice, but of your player choice (unless you play with stats being rolled from the top down, rather than rolled and then allocated).

This. I have yet to see a rolled statblock result in a sub-optimal character; and my experience has been that it's mostly being used to produce more optimal characters. There's a major difference between starting with a 15 and starting with a 16, 17, or 18. As already mentioned, it effectively gains you an extra feat [two if you roll an 18], and more early power.

And, in addition to the ability to start with a 16+ stat, all schemes of rolling I've encountered have been engineered to result in an overall stronger-than-point-buy statblock. You will never have less than the stats you want; it's a question of "how much better than baseline are you".

Theoretically, there's supposed to be a risk vs. reward approach to rolling vs. points buy; but rolling is, at this point, all reward with minimal risk.

[MThurston]

Why is this a problem? Why not just let people play what they like to play?

I do enjoy playing very random characters and making odd arrays work, but plenty of my friends don't and have more fun when they have more control over who their character is. I'm not going to take that away from them just because I enjoy doing it differently.
Ah, this is a good question.

Do you want people to grow as roleplayers or just play themselves in every game.

I have a buddy that asks you what motivates your character to get up in the morning.

If you can't answer that thrn he tells you to make a new character that does know.

I can roleplay the hell out of a rogue character. It's very easy for me. I however struggle at playing pure spell casters. It is very hard for me.

I found a very cool DM and we made characters for Harn Master 3. While talking he said that he had a spot for his 5e game. I agreed and we started talking character generation.

His way was roll 4d6b3 and you could Re roll if the total wasn't 72 total points. That is an average of 12 per stat. If you had 72 points, you could not reroll.

You also had to keep the order you rolled them in. My rolls were 11, 13, 18, 11, 14, 18. Yes a very lucky roll and I took straight human.

I also took Warlock as my class. A class I did not like and didn't know what to do with. I like getting into sword fights and hand to hand combat.

Looking at Warlock subclasses I stumbled on Hexblade.

So I went from dreading playing a spell caster to really liking the idea of it.

We just reaches level 5 and we play tonight. Can't tell you how much I like this character.

And that is why sometimes playing a class you don't know can be fun.

[CantigThimble]

Do you want people to grow as roleplayers or just play themselves in every game.

*Snip*

And that is why sometimes playing a class you don't know can be fun.

I agree that it is fun and has helped me be a better player, but I also think that it should be each players' choice whether or not to do this otherwise they're just going to be ticked off.

[MaxWilson]

I agree that it is fun and has helped me be a better player, but I also think that it should be each players' choice whether or not to do this otherwise they're just going to be ticked off.

So, do we all agree then that the PHB method (your choice of rolling stats or standard array) is a pretty reasonable default, so each player can have their choice?

[Coffee_Dragon]

Maybe some people who say "I like rolling because [if and when the dice give me a god array] it lets me play nonstandard build X" could try playing build X with a standard array and also find out it's something they didn't think would work but by some measure does?

[MThurston]

So, do we all agree then that the PHB method (your choice of rolling stats or standard array) is a pretty reasonable default, so each player can have their choice?

Depends on what your DM wants in their game. All methods can work.

[LordCdrMilitant]

Ah, this is a good question.
Do you want people to grow as roleplayers or just play themselves in every game.
I have a buddy that asks you what motivates your character to get up in the morning.
...

None of that has to do with points-buy or rolling.

And I'd argue that you definitely don't get better value to that effect out of rolling.

Because I believe, and strongly, that all that about "playing with odd arrays" or "using the hand you've got" is a load of crap. You can sort your stats however you want; and there's so much dice fixing you'll never have a bad block [and, as a GM, I wouldn't want one of my players to have a weaker block than the rest]. Rolling stats, at this point, just drives stat inflation and allows more powerful characters overall.

[Lance Tankmen]

I think ultimately it comes down to there is no wrong way to play. Each opinion is fine, my opinion is I prefer rolling 4d6 drop lowest.

I make people pick race class then roll. I don't re-roll ones, or have multiple groupings, no one is forced to play D&D at any table, and session 0 I lay out the rules, the time to argue character creation and/or leave is then. Ive never had a player quit my table in my 3 years as DM with rolled stats, HP(though I admit I allow a reroll keeping the second as CR/encounters sometimes hit harder plus gamble) And starting gold(keeping background gear but none of the preset stuff).

Though I did have one player quit it was due to him wanting to be a super hero and me failing to explain and catch certain things, he made a half elf rogue but then rerolled around level 3 into a human monk named Daniel rand, I never got into Iron fist so i had no Idea that's what he was doing, sadly Daniel rand died to an orc ambush, he didnt expect to be able to die, and when building a new character I foolishly said oath of vengeance paladins have a batman mindset sort of justice regardless of the law. He only heard like batman. He got to level 5 and quit after growing bored as he wasn't a super hero. I found it out later after talking with some other co-workers about it that knew him better, basically had we been playing a super hero game he'd have had a blast as it was, fantasy isnt his cup of tea.

[Tanarii]

So, do we all agree then that the PHB method (your choice of rolling stats or standard array) is a pretty reasonable default, so each player can have their choice?
As someone that runs a no feats campaign, it would have been nicer if the standard array had more odd numbers, especially the low value stats. Because Humans need a need a minor boost.

15 / 14 / 13 / 12 / 11 / 9 for example. Players aren't likely to go back and ASI lower abilities, and +1 mod to your two lowest scores would make standard array minorly more attractive as humans without coming anywhere close to overpowering them. Or any other races for that matter.

I've noticed Humans are a little more attractive if you allow players to roll first then choose race, and end up with a bunch of odd numbers (4 or so). So it'd be a good tweak to the standard array IMO.

[MaxWilson]

As someone that runs a no feats campaign, it would have been nicer if the standard array had more odd numbers, especially the low value stats. Because Humans need a need a minor boost.

I think that's a problem with the base ruleset actually, not with any particular array--odd stats are almost pointless in 5E except for encumbrance and against Intellect Devourers.

My fix for this is to make odd stats grant an extra +1 on ability checks (not attack rolls or spell DCs or anything else), so there is a measurable distinction between e.g. Str 18 and Str 19. Applies to monsters too.

Odd Con is still useless even with my fix, but for all the other abilities, no +1 is ever "wasted".

[Tanarii]

Odd Con is still useless even with my fix, but for all the other abilities, no +1 is ever "wasted".
Con checks, and if you use variant skills Con (Athletics) checks, are a thing. So it shouldn't be useless, unless you just never call for them.

One that I house rule is Dash checks (Chase rules) are Con (Athletics) checks, not con saves. That makes them somewhat common.

Anyway, I like your mod. Only downside is it may cause some confusion, but if the players can handle it cool.

[Pex]

I think that's a problem with the base ruleset actually, not with any particular array--odd stats are almost pointless in 5E except for encumbrance and against Intellect Devourers.

My fix for this is to make odd stats grant an extra +1 on ability checks (not attack rolls or spell DCs or anything else), so there is a measurable distinction between e.g. Str 18 and Str 19. Applies to monsters too.

Odd Con is still useless even with my fix, but for all the other abilities, no +1 is ever "wasted".

Something we finally agree on. :smallbiggrin:

I like this.

[Merudo]

I'm not passing judgment on that -- I don't play at your table so what do I care? -- just saying that some of the improbable stats I see in threads are more probable when you're using additional rules to help favor high results.


I also strongly suspect a considerable number of players outright cheat and give themselves better stats than they actually roll.

Frankly, why wouldn't they? It's such a powerful advantage, and there is so little risk. Especially when people roll at home or when everyone roll at the same time - unless someone observes all your six rolls it's nearly impossible to prove you didn't actually get that 18.

Last time I played at a table that rolls for stats, 4/6 of the players "rolled" an 18. The probability of getting that or something better is about .09%, lol!

[Bundin]

If someone wants to roll stats: fine! But I'll have to be present when you roll your 4d6 (drop lowest). If the total result is really really really below par, then the player can switch to an array, but if it's just meh, then that's what it'll be. Always allowing a switch to an array means it's a free shot at awesome without any risk.

Most players I know stick to point buy, which suits me fine. Less drama from the one that gambled and lost during the campaign.

[Cybren]

I also strongly suspect a considerable number of players outright cheat and give themselves better stats than they actually roll.

Frankly, why wouldn't they? It's such a powerful advantage, and there is so little risk. Especially when people roll at home or when everyone roll at the same time - unless someone observes all your six rolls it's nearly impossible to prove you didn't actually get that 18.

Last time I played at a table that rolls for stats, 4/6 of the players "rolled" an 18. The probability of getting that or something better is about .09%, lol!

If you can't expect all the players to engage in the game in good faith the rules you're using don't matter.

(also, obviously, your anecdote is meaningless, as there's a sufficient number of groups rolling for that to come up multiple times. That's akin to being astounded that today is someones birthday)

[MThurston]

3d6 chance of 18 = 1/216

4d6b3 chance of 18 = 1/61

Even better if you reroll 1's.

So if 10 people roll up stats, only one of them should have an 18.

I however did roll up my Warlock with two 18s.

[Jophiel]

If you can't expect all the players to engage in the game in good faith the rules you're using don't matter.
Eh, stat generation is major enough and has long-lasting effects on the game that I'm fine with treating it a bit more carefully than just assuming everyone will cheat always. Most rolls are done openly at the table and with less chaos than five people all stat rolling at once. And maybe someone will fudge an Initiative roll or something, though I'd rather they didn't, but that's still less detrimental to the game than someone showing up with their sheet full of 16-18s that they totally rolled for reals in their kitchen. When you're working with players new to you that are friends of friends or people from Meetup or guys from the local game store who heard you're looking, I'd rather not have to worry about it.

[Temperjoke]

If you can't expect all the players to engage in the game in good faith the rules you're using don't matter.

(also, obviously, your anecdote is meaningless, as there's a sufficient number of groups rolling for that to come up multiple times. That's akin to being astounded that today is someones birthday)

Somewhere in the world, there is one person who has never rolled above a 1 every time they play, not realizing that they are the one chosen by the dice gods to bear the bulk of the universe's bad dice luck.

EDIT: Yeah, I will say when you're in a situation that you can't observe the rolls, and you don't trust the individuals involved to not fudge things (like strangers in public game settings) then having a system like point buy or standard array where you can verify the numbers relatively easy to ensure no cheating is fine.

[MaxWilson]

Eh, stat generation is major enough and has long-lasting effects on the game that I'm fine with treating it a bit more carefully than just assuming everyone will cheat always. Most rolls are done openly at the table and with less chaos than five people all stat rolling at once. And maybe someone will fudge an Initiative roll or something, though I'd rather they didn't, but that's still less detrimental to the game than someone showing up with their sheet full of 16-18s that they totally rolled for reals in their kitchen. When you're working with players new to you that are friends of friends or people from Meetup or guys from the local game store who heard you're looking, I'd rather not have to worry about it.

Tangent: a cheater doesn't have to cheat on die rolls. They could, for example, "forget" to deduct HP when damaged, "forget" to deduct spell slots when spellcasting, change on the fly which spells they have prepared/memorized, write extra equipment down on their equipment list, rewrite their backstory on the fly (Old Man Henderson), etc. Who wants to deal with that?

That being said, the simplest way to deal with stat rolls if you want to avoid even the appearance of anyone cheating, but still want to create characters in advance, is to have the DM roll up the stats and email/IM them to everyone. (If you can't trust your DM to be fair and impartial your campaign is doomed already no matter what the stats are.)


Last time I played at a table that rolls for stats, 4/6 of the players "rolled" an 18. The probability of getting that or something better is about .09%, lol!

I played at at table like that when I first started 5E. (Wound up leaving because the DM was too railroadey, but that's another story.) There were two newbies (ages eleven and sixteen), an old grognard, and me, and I think but am not 100% sure that everyone else had at least one 18. At the time I chalked it up to teenagers not understanding that you're not supposed to reroll, but maybe they were deliberately cheating, who knows. The funny thing is though that the willingness to reroll/cheat and actual skill at the game seem to negatively correlate with each other IME, so those PCs were actually less effective than my PC simply by virtue of me doing things like not standing in Fireball formation with everyone else--not making a saving throw is always better than making a saving throw with a +2ish bonus. I certainly never felt overshadowed despite rolling "only" a couple of 15s IIRC.

But I do find it unpleasant when the PC has you roll up a high-level PC and you get mediocre stats. It hurts my suspension of disbelief: "Why didn't this guy just take the money and retire at level 7?"/"Is this really the PC I would have chosen to play from levels 1 to 14?" IMO the solution here is not to change how you roll stats, it's just to play everyone up from low level. If you're actually emotionally invested in the mediocre-stats guy's destiny and want to keep playing him for weeks or months on end, that's different from the DM just telling you that suddenly this guy is 14th level (or whatever).

I enjoy playing low-stats guys just fine at low/mid levels, but I feel like in a reasonable world there should be some correlation between inborn talent and success at adventuring, so low-stats guys should be less common at high levels than at low levels. Breaking that implied correlation bugs me.

[Knaight]

What do you mean by rigging the system?

3d6 is an average of 10.

4d6 pick 3 makes that average 11.

The average of 3d6 is 10.5, the average of 4d6b3 is 12.24. If you're going to give numbers at least give the right numbers.

[Jophiel]

Tangent: a cheater doesn't have to cheat on die rolls. They could, for example, "forget" to deduct HP when damaged, "forget" to deduct spell slots when spellcasting, change on the fly which spells they have prepared/memorized, write extra equipment down on their equipment list, rewrite their backstory on the fly (Old Man Henderson), etc. Who wants to deal with that?
All true. But sometimes you don't know until you know. And some of that stuff is more obvious in the early game ("Wait, how did you cast three spells at level 1?") so hopefully you've weeded out the habitual cheaters by then. I think that you hopefully trust everyone but might as well take easy safeguards especially when dealing with players new to your table.

[Tanarii]

(If you can't trust your DM to be fair and impartial your campaign is doomed already no matter what the stats are.)Yet despite this, DMs who admit to (or even advocate) fudging rolls, or adjusting monsters/combats mid-combat, are endemic.

[MThurston]

The average of 3d6 is 10.5, the average of 4d6b3 is 12.24. If you're going to give numbers at least give the right numbers.

The average of 2d6 is 7.

The average of 4d6 is 14.

Divide 4d6 by half it's 2d6 and half of 14 is 7.

So one can use statistics to come up with my answers.

Another way to look at it is this.

2d6 average is 7.
1d6 average is 3. They total 10. There is no .5 in dice rolling.

So If the average of 2d6 is 7 and the average of 1d6 is 3. Then 7-3=4.

So with 4d6b3 one could guess that 3 and 4 are the average if 2d6. 4 being the bigger number. So this is how you would come up with 2d6 of 7 and 4 being the best of 4 and 3.

This would give you average 4d6b3 of 11.

Now you have insight to my crazy, uneducated mind.

[Jakinbandw]

The average of 2d6 is 7.

The average of 4d6 is 14.

Divide 4d6 by half it's 2d6 and half of 14 is 7.

So one can use statistics to come up with my answers.

Another way to look at it is this.

2d6 average is 7.
1d6 average is 3. They total 10. There is no .5 in dice rolling.

So If the average of 2d6 is 7 and the average of 1d6 is 3. Then 7-3=4.

So with 4d6b3 one could guess that 3 and 4 are the average if 2d6. 4 being the bigger number. So this is how you would come up with 2d6 of 7 and 4 being the best of 4 and 3.

This would give you average 4d6b3 of 11.

Now you have insight to my crazy, uneducated mind.

For those of us eneducated allow me to present: anydice.com ! It allows you to easily get stats on your favorite dice rolling system and can even be configured to allow you to calculate average damage per round based on hit chances and damage rolls. It is truly the savour of people who don't have the time to work out tricky probabilities!

[MThurston]

For those of us eneducated allow me to present: anydice.com ! It allows you to easily get stats on your favorite dice rolling system and can even be configured to allow you to calculate average damage per round based on hit chances and damage rolls. It is truly the savour of people who don't have the time to work out tricky probabilities!

My math is close enough.

[JoeJ]

For those of us eneducated allow me to present: anydice.com ! It allows you to easily get stats on your favorite dice rolling system and can even be configured to allow you to calculate average damage per round based on hit chances and damage rolls. It is truly the savour of people who don't have the time to work out tricky probabilities!

How do you tell it to do the best 3 of 4d6? Everything I've tried returns a syntax error message.

[Man_Over_Game]

My math is close enough.

Rounding ends up creating a bias, though, and usually, people like us are particular and specific enough that the .5 difference between a Greatsword and a Great Axe matters.

The Devs took an interesting side on the point of bias, though. In every circumstance where they take average numbers of a dice roll, they sway in favor of the Player.

Hit Dice for players, using the Average option? That's actually .5 higher in the sway of the player.

Damage dice for monsters, using the Average option? That's a .5 lower in the sway of the player.

It's kind of interesting. It gives off this warming vibe of "When in doubt, help the player".

[Aquillion]

I think part of the reason for that is also that they want players to prefer the "lightweight" option by default - rolling dice takes a bit more time. I think they wanted to avoid situations where players would feel compelled to go for the more time-consuming option just to be optimal.

[mephnick]

Yet despite this, DMs who admit to (or even advocate) fudging rolls, or adjusting monsters/combats mid-combat, are endemic.

Haven't you heard? It's ok to invalidate choices and rolls if it WiLl heLP tHe sTOrY.

[Tanarii]

How do you tell it to do the best 3 of 4d6? Everything I've tried returns a syntax error message.
output [highest 3 of 4d6]

(Note everything must be lower case)

[Jakinbandw]

How do you tell it to do the best 3 of 4d6? Everything I've tried returns a syntax error message.
output [highest 3 of 4d6]

It's one of the inbuilt functions. It's more tricky to calculate the average of 4d6B3R1 7 times take the 6 best, but if you want I could spend some time and write up a program on anydice so we could see how it goes.

Edit:
R: d{2..6}
output [highest 3 of 4dR]

Will give you 4d6B3R1. (Average 13.43)

[JoeJ]

output [highest 3 of 4d6]

(Note everything must be lower case)

It works! Thank you. (Now why didn't I think of trying brackets?)

[Jakinbandw]

Sometimes I am a bit crazy. This block of anydice code will give you your probabilities for 4d6 reroll 1s 7 times. Drop the lowest dice on each roll, drop the lowest roll. It was how I made my first character ever (without the reroll 1s to be fair)




function: random SONE STWO STHR SFOU SFIV SSIX X:n{
if X=1{result: SONE}
if X=2{result: STWO}
if X=3{result: STHR}
if X=4{result: SFOU}
if X=5{result: SFIV}
if X=6{result: SSIX}}


R: d{2..6}
ONE: [highest 3 of 4dR]
TWO: [highest 3 of 4dR]
THR: [highest 3 of 4dR]
FOU: [highest 3 of 4dR]
FIV: [highest 3 of 4dR]
SIX: [highest 3 of 4dR]
SEV: [highest 3 of 4dR]

SONE: [highest of ONE and TWO]
DROP: [lowest of ONE and TWO]
STWO: [highest of DROP and THR]
DROP: [lowest of DROP and THR]
STHR: [highest of DROP and FOU]
DROP: [lowest of DROP and FOU]
SFOU: [highest of DROP and FIV]
DROP: [lowest of DROP and FIV]
SFIV: [highest of DROP and SIX]
DROP: [lowest of DROP and SIX]
SSIX: [highest of DROP and SEV]


output [random SONE STWO STHR SFOU SFIV SSIX 1d6]

[MaxWilson]

Rounding ends up creating a bias, though, and usually, people like us are particular and specific enough that the .5 difference between a Greatsword and a Great Axe matters.

The Devs took an interesting side on the point of bias, though. In every circumstance where they take average numbers of a dice roll, they sway in favor of the Player.

Hit Dice for players, using the Average option? That's actually .5 higher in the sway of the player.

Damage dice for monsters, using the Average option? That's a .5 lower in the sway of the player.

It's kind of interesting. It gives off this warming vibe of "When in doubt, help the player".

It's not the simple.

Use caltrops against a monster resistant to nonmagical piercing damage? Round down, in favor of the monster. 1 HP/2 = 0.5 HP = 0 damage, so this monster is immune to caltrops. (Time to bust out the ball bearings instead!)

[Arcangel4774]

Point buy is better most of the time for when people want to create a particular character. Otherwise i think its tons of fun to roll in order, and try to make whatever you end up with work.

[ZorroGames]

0D&D, AD&D/1st and 2nd briefly only rolled.

“Weakest” playing personal character was mostly 8-10 in stats. Had two wargamer/FRPGer friends who managed 3 x 18s rolling 3D6.

I much prefer point buy. I think I could not have experimented with my character mix as easily with rolling Duce.

But all my 5e has been AL by choice. AL requires Standard Array or Point Buy. Most of my PCs have been Mountain Dwarf of course - but a mix of straight up classes and MC characters.

Starting when you you only had Fighters, Clerics, and Mages I have come to believe a PC should reflect a player’s Design at the start. After that anything can happen.

So far my one class characters have been Mountain Dwarf Fighter (ST and DE builds,) War Wizard, Monk; KIA in Chult Variant Human Ranger, and my (only) third tier “this will just be an experiment” Earth Genasi Fighter. Multiple MC Mountain Dwarf characters, one Wizard/Cleric Forest Gnome, and a Variant Human Ranger about to MC into Rogue rounds out my current mix.

[Eradis]

Point buy remove the unbalance aspect of a party. In every single game I had to deal with rolled stat (which can be interesting for many reason), there was always that one guy that rolled better than the rest if just slightly and one guy under the average rarely slightly. Makes one character feel like a Nephalem, a chosen one to be a hero among heroes and one character feel lackluster, under-par, as a commoner traveling with heroes at best. And this is with ignoring ones. Even if someone's stat block is decent, if the rest of the party is awesome, you feel apart and it feels unfair. This is mainly why I impose point buy and constant hit points.

Honestly, only time I will accept rolled stat at a table is if A. I'm a player, B. if it is a one-shot or C. if it is an odd concept implying death is being common in games and that you often change character either à la Darkest Dungeons (video game) or à la Gloomhaven (boardgame).

[Barebarian]

At my table we allow 30 point buy but ONLY if you bump your Charisma to 12. Everyone must be sexy. EVERYONE. :smallbiggrin:

Seriously though, apparently rolling gives you more points than point buy on average? I don't know if that balances out by making it so you can't pick where they go but like, I think 27 is a bit low and I kinda don't like that stats start at 8?

[Cybren]

At my table we allow 30 point buy but ONLY if you bump your Charisma to 12. Everyone must be sexy. EVERYONE. :smallbiggrin:

Seriously though, apparently rolling gives you more points than point buy on average? I don't know if that balances out by making it so you can't pick where they go but like, I think 27 is a bit low and I kinda don't like that stats start at 8?

https://kaitlyn-burnell.tumblr.com/post/140138643493/in-5th-edition-dd-is-it-better-to-roll-or-use I think the array/point-buy winds up slightly better than the average 4d6b3 array would generate, but obviously you lose the chance to spike having a 16+ before racial modifiers.

[Coffee_Dragon]

At my table we allow 30 point buy but ONLY if you bump your Charisma to 12. Everyone must be sexy. EVERYONE. :smallbiggrin:

Not sure if only joking, but CHA 12 means you have a smooth and/or forceful personality, not that you're sexy. Everyone gets to be drawn by Larry Elmore or someone just from being an adventurer.

[thereaper]

I personally see no reason to ever roll for stats or hp.

If you roll, one of three things will happen:

1) You end up with a point buy array, in which case you might as well have used point buy.
2) You roll worse than a point buy array, in which case you would have been better off with point buy.
3) You roll better than a point buy array, and end up with an overpowered character.

In all three scenarios, point buy is as good or better for the game.

In addition, point buy introduces the exploit of deliberately killing off your characters if they have poor stats, so that you can get another chance at rolling an overpowered character.

[Mad Max]

When I build a character by myself for a campaign, I tend to use point buy, because it allows me to precisely fine tune the character I have in mind. On the other hand, when I build a character with a group, especially if I have no specific character in mind, I tend to prefer rolling for it. I think it's because the randomness adds to the fun when playing with a group. You'll all laugh about a player rolling a low stat, and cheer when someone rolls that lucky 18. Overall, I enjoy both ways of stat generation! The only one I don't really enjoy is Standard Array, because it's like point buy, but more restricting.

[Barebarian]

https://kaitlyn-burnell.tumblr.com/post/140138643493/in-5th-edition-dd-is-it-better-to-roll-or-use I think the array/point-buy winds up slightly better than the average 4d6b3 array would generate, but obviously you lose the chance to spike having a 16+ before racial modifiers.

I'll look into that :O

[Barebarian]

CHA 12 means you have a smooth and/or forceful personality,

Exactly! Sexy!:smallamused:

[MaxWilson]

At my table we allow 30 point buy but ONLY if you bump your Charisma to 12. Everyone must be sexy. EVERYONE. :smallbiggrin:

Seriously though, apparently rolling gives you more points than point buy on average? I don't know if that balances out by making it so you can't pick where they go but like, I think 27 is a bit low and I kinda don't like that stats start at 8?

I speculate that it's a deliberate design decision to encourage people to be happy with their rolls. WotC encourages taking fixed HP slightly better than the average roll; they encourage rolling stats by making the standard array slightly weaker than rolling. (Point buy is its own thing, arguably as powerful as rolling if you're willing to min max and dump stats to 8 and not purchase any ability scores that will be odd after racial mods. But point buy is an advanced variant anyway.)

I have noticed that it's very easy to be content with 5E stat rolls compared to AD&D stat rolls, maybe because the distribution of bonuses is flatter and the stakes are lower. In AD&D the difference between Str 18 and Str 16 is enormous; in 5E it's minimal, so it doesn't hurt to roll less than 18. Stat rolling is one of the things 5E gets right IMO.

[LordCdrMilitant]

The article about rolling vs. points buy is interesting... but I don't entirely agree with it

I've done similar simulations when I was on a kick of breaking down the advantages of PB and rolling, by "brute force" [running off a couple thousand examples], and got somewhat similar results, but came to the opposite conclusion .


First off, I don't think her methodology of measuring the top three bonuses and summing them is an entirely valid way of assessing the array. There are multiple reasons for this; one of which is that I don't think D&D is a 3-stat game. Going over the classes, most really need 1 to get to 20 for combat, and an extra one to either add additional power or make you not-a-failure with skills [and some classes, notably Sorc and 'lock, only need CHA]. And you're not going to spend ASI's on any but the combat skill. Obviously, this seems to favor PB, because it's less punishing in your off stats to buy 2 15's than 3 while rolling has about 45% odds of getting at least 2 15+'s. The second stat though is usually the "skills" stat, and isn't essential to be awesome, just like decent. After that, you're really just looking for 10's or 8's so you don't absolutely suck at saves.

However, you get a fixed number of feats and ASI's, and they come from the same pool in 5e. You want to get your combat stat to a 20, and then have as many feats as possible, to have a more efficient and more powerful character. Starting with a +4 [16+2 racial] is not only a considerable early-game boost, but it also lets you take a feat earlier, since you'll be hitting 20 at level 4 and taking feats at every ASI afterwords leading to a net more powerful character, while characters who only rolled a 15 [thus having a 17] require 2 ASI's to hit their 20, placing them as many as 4 levels behind in terms of modifiers and feats, or more practically speaking, leaving them with 1 less feat overall than characters starting with a 16. This, to me is more significant that the total summative modifier from the first three stats.

Now: rolling for stats gives you about a 55% chance of getting a 16 or higher for one of your stats [and like a 10% odds of getting an 18, which is effectively 2 extra feats over a points-buy array]. The point at which you're doing [i]worse than your bought array is if your total starting mod in your key stat drops below a +3 [14+2=16], thus requiring an extra ASI and effectively losing a feat. You only have about a 10% chance of doing worse [not rolling a 14+] than having bought your stats.

Ergo, with a 55% chance of winning and only a 10% chance of really failing, if I have a different race I want to play, it seems pretty obvious to me that I want to roll my stats with 4d6b3 [even more so if there are other dice-fixing measures in play, like re-roll 1's, roll 2 arrays and pick the highest, or roll 7 stats and pick the best 6]. And there are still good odds [notably the 42% chance of getting 2 15+ overlaps somewhat with the 55% chance of getting a 16+] of getting 2 15+ stats, which would lead to a very strong character.


Now, there's an exception: Humans get 2 +1's, and no +2's, so they'd need a 17 [30%] to save themselves an ASI and get their 20 at level 4, and already get a bonus feat. This means, that, if you want to be human, you don't want to roll, since you can get 2 16's off the start and a feat with PBE, since you have only a 30% chance of "winning" but a 58% chance of "losing".


[Also, one further observation: Most of us in my group have lots of Citadel D6's. There is an article I read indicating that these dice [and the similar Chessex brand ones] are generally poorly balanced and have like a 2% higher-than-correct chance of rolling a 1 ;)]

[Tanarii]

That logic assumes players should want to get to a 20 in the primary score.

If you do want to, its not bad logic though. Similarly if you want to play an off-stat class (Non-Str race fighter) rolling is the only way you can start with a 16+ in your primary ability score.

[Lance Tankmen]

The article about rolling vs. points buy is interesting... but I don't entirely agree with it

I've done similar simulations when I was on a kick of breaking down the advantages of PB and rolling, by "brute force" [running off a couple thousand examples], and got somewhat similar results, but came to the opposite conclusion .


First off, I don't think her methodology of measuring the top three bonuses and summing them is an entirely valid way of assessing the array. There are multiple reasons for this; one of which is that I don't think D&D is a 3-stat game. Going over the classes, most really need 1 to get to 20 for combat, and an extra one to either add additional power or make you not-a-failure with skills [and some classes, notably Sorc and 'lock, only need CHA]. And you're not going to spend ASI's on any but the combat skill. Obviously, this seems to favor PB, because it's less punishing in your off stats to buy 2 15's than 3 while rolling has about 45% odds of getting at least 2 15+'s. The second stat though is usually the "skills" stat, and isn't essential to be awesome, just like decent. After that, you're really just looking for 10's or 8's so you don't absolutely suck at saves.

However, you get a fixed number of feats and ASI's, and they come from the same pool in 5e. You want to get your combat stat to a 20, and then have as many feats as possible, to have a more efficient and more powerful character. Starting with a +4 [16+2 racial] is not only a considerable early-game boost, but it also lets you take a feat earlier, since you'll be hitting 20 at level 4 and taking feats at every ASI afterwords leading to a net more powerful character, while characters who only rolled a 15 [thus having a 17] require 2 ASI's to hit their 20, placing them as many as 4 levels behind in terms of modifiers and feats, or more practically speaking, leaving them with 1 less feat overall than characters starting with a 16. This, to me is more significant that the total summative modifier from the first three stats.

Now: rolling for stats gives you about a 55% chance of getting a 16 or higher for one of your stats [and like a 10% odds of getting an 18, which is effectively 2 extra feats over a points-buy array]. The point at which you're doing [i]worse than your bought array is if your total starting mod in your key stat drops below a +3 [14+2=16], thus requiring an extra ASI and effectively losing a feat. You only have about a 10% chance of doing worse [not rolling a 14+] than having bought your stats.

Ergo, with a 55% chance of winning and only a 10% chance of really failing, if I have a different race I want to play, it seems pretty obvious to me that I want to roll my stats with 4d6b3 [even more so if there are other dice-fixing measures in play, like re-roll 1's, roll 2 arrays and pick the highest, or roll 7 stats and pick the best 6]. And there are still good odds [notably the 42% chance of getting 2 15+ overlaps somewhat with the 55% chance of getting a 16+] of getting 2 15+ stats, which would lead to a very strong character.


Now, there's an exception: Humans get 2 +1's, and no +2's, so they'd need a 17 [30%] to save themselves an ASI and get their 20 at level 4, and already get a bonus feat. This means, that, if you want to be human, you don't want to roll, since you can get 2 16's off the start and a feat with PBE, since you have only a 30% chance of "winning" but a 58% chance of "losing".


[Also, one further observation: Most of us in my group have lots of Citadel D6's. There is an article I read indicating that these dice [and the similar Chessex brand ones] are generally poorly balanced and have like a 2% higher-than-correct chance of rolling a 1 ;)]

Likely just nitpicking but a vhuman fighter doesnt need a 17 to get 18, if he has a 16 he takes heavy armor master and adds +1 to str for 18 str, or any class that wants to grab a feat for their main stat fighter is just easiest .

[Merudo]

https://kaitlyn-burnell.tumblr.com/post/140138643493/in-5th-edition-dd-is-it-better-to-roll-or-use I think the array/point-buy winds up slightly better than the average 4d6b3 array would generate, but obviously you lose the chance to spike having a 16+ before racial modifiers.

Extremely interesting article.

When you only value the first 3 stats, and you value them equally, the "greedy" array gives +1.35 modifier compared to rolling (9 vs 7.65), on average.

So, what happens when you value fewer stats, or more stats?

I did the computations and got these results for each number of valued attributes:


1 attribute: rolling gives +.59 modifier compared to point-buy (3.59 vs 3)
2 attributes: rolling gives a +.17 modifier compared to point-buy (6.17 vs 6)
3 attributes: pointbuy gives a +1.35 modifier compared to rolling (9 vs 7.65)


Interestingly, in the case of the Mountain dwarf, the results are a bit different:


1 attribute: rolling gives +.59 modifier compared to point-buy (3.59 vs 3)
2 attributes: point-buy gives a +.33 modifier compared to rolling (7 vs 6.67)
3 attributes: point-buy gives a +.85 modifier compared to rolling (9 vs 8.14)


Now, what about Variant human? We get these results:


1 attribute: rolling gives +.08 modifier compared to point-buy (3.08 vs 3)
2 attributes: point-buy gives a +.33 modifier compared to rolling (6 vs 5.67)
3 attributes: point-buy gives a +.85 modifier compared to rolling (8 vs 7.14)


This is a good case for point-buy, especially with 1 or 2 attributes valued. One reason is that to roll a better primary attribute with VHuman you need a 17 or a 18, while with other races a 16 is enough.

And what about the Half-Elf?


1 attribute: rolling gives +.59 modifier compared to point-buy (3.59 vs 3)
2 attributes: rolling gives a +.17 modifier compared to point-buy (6.17 vs 6)
3 attributes: point-buy gives a +.85 modifier compared to rolling (9 vs 8.14)


which is the worse case for point-buy: it behaves the same as a +2/+1 race when 1 or 2 attributes are valued, but performs worse when 3 attributes are valued.

The previous analysis assumed we picked the race ahead of time and then rolled for stats. However, for pure min-maxing it may be preferable to pick the race depending on the stats we rolled.

To simplify this analysis, I assume that we either pick a +2/+1 race, or we pick VHuman. I assume the feat from Vhuman is roughly equivalent to .75 of an ASI. With point-buy we pick VHuman if one or two attributes are valued, and a +2/+1 race if 3 attributes are valued. With rolling we make our choice depending on what our stats are. This gives the following results:


1 attribute: rolling gives a +.20 modifier compared to point-buy (3.95 vs 3.75)
2 attributes: point-buy gives a +.21 modifier compared to rolling (6.75 vs 6.54)
3 attributes: point-buy gives a +.98 modifier compared to rolling (9 vs 8.02)

[LordCdrMilitant]

Likely just nitpicking but a vhuman fighter doesnt need a 17 to get 18, if he has a 16 he takes heavy armor master and adds +1 to str for 18 str, or any class that wants to grab a feat for their main stat fighter is just easiest .

Yeah, point. I've always started with GWM for the added damage. Rolling might also just be better for humans too. I've taken Silver Tongued or Diplomat from UA before to achieve the same effect for casters.

[Merudo]

I'll also consider the case where three attributes are valued: Con, Dex, and another attribute, and the character is using medium armor. This applies to Barbarians, Hexblades, and some Clerics. In those cases, Dex doesn't provide additional AC over 14 - so we won't value any increase of dex over 14.

Then, we get


3 attributes: point-buy gives a +.94 modifier compared to rolling (8 vs 7.06)


So rolling for stats becomes just a little bit worse.

[dave2008]

Now, I don’t have anything AGAINST rolling for stats, but I personally feel point buy is more interesting. You have to make hard choices to build your character. Almost every character ive seen has rolled a good array of stats, to the point that nothing is lackluster. This seems a bit dull to me. Am I the only one in 5e that actually implements point but at their table?

We've done it both ways. I like point buy because I feel I get to make the decisions as opposed to the luck of the dice.

However, I'm wondering if anyone has played in a group where some people in the group used point buy and some rolled stats? I've never seen that before.

[ZorroGames]

If (when?) I make a third world for gaming, “You've got to ask yourself one question. Do I feel lucky? Well, do ya, punk?” - it will require point buy or 3D6 in order (Oldest school.)

And it will be AL like/Lite in a different physical setting.

[Temperjoke]

You know, there's probably some sort of Unified Game Stat theory that could be derived from a combination of Point Buy, Rolled Stats, and Standard Array, where you include the options to pay to roll your stats or utilizing the standard array in with Point Buy. But I'm not into mathematics enough to figure it out.

[Lyracian]

However, I'm wondering if anyone has played in a group where some people in the group used point buy and some rolled stats? I've never seen that before.
Yes.
In the game I am running 4 players rolled the last one went with triple 15’s point buy Barbarian.

I am just starting another game where we were given the option of point buy or rolling. Not sure if anyone took point buy though.