Is TWF Supposed To Be Worse To Account For Surprise Rounds? [Archive]

GoodbyeSoberDay

2015-10-10, 07:43 AM

Comparing using DEX vs. STR in general:

DPR: As is established, this goes to STR, at least with two handers and the appropriate feats.
Ability Checks: Athletics and its uses are great, but it doesn't compete with Acrobatics, Sleight of Hand, Stealth (STR user likely has disadvantage here), Thieves Tools, and oh, right, Initiative. DEX is far superior here.
Saves: STR is the most common "non-standard" save, but it can't compete with DEX saving throws.
AC: This is mostly a wash, so long as your character has the right proficiencies. STR based who get their hands on full plate pull ahead a bit in the low levels, but the DEX-based characters start to catch up once DEX gets maxed out, and theoretically have higher end-game AC.
Misc: DEX has a superior ranged option, even if normally melee, and can often be used to resist those pesky Athletics-based maneuvers.

I'd say it balances out all right. If TWF had superior DPR to GWF, something would be seriously wrong.

Goodberry

2015-10-10, 08:41 AM

A couple more points:

Str DPR relies heavily on -5/+10, which means it's less effective against high AC.

Twf means TWO weapons. Less vulnerable to disarm, plus two damage types. Also easier to switch to S&B if desired.

Thrown weapons have very short range, whereas dex can use bows.

High feat requirements makes it harder to take any non-combat options.

Size restrictions on heavy weapons limits your racial options.

1 level of thief can give you Expertise in athletics(for shove), 1d6 sneak attack if you use finesse weapons. One of the best melee dips.

Vogonjeltz

2015-10-10, 02:52 PM

Dexterity based fighting was designed to be less powerful because dex is a much better attribute than str. Ignoring attack/damage bonuses str is far and away the worst stat. It governs one skill, a rarely-used saving throw, is not a casting stat, and has no defensive value. By contrast, dex covers several skill, is arguably the most important save, adds to AC and initiative and can apply to both melee and ranged weapons.

Also, from a thematic standpoint, str fighters are the heavy hitters, sacrificing mobility and precision for raw damage. Dex fighting is about out-maneuvering your foe; hitting more and getting hit less.

One more note. DPR is an inherently flawed indicator. The goal is to kill your opponents before you run out of HP, not to kill them faster.

It's a rarely used saving throw for spells. I'm pretty sure it's quite common as a saving throw for monster abilities and it's incredibly useful for contests, most of which will be physical.

MaxWilson

2015-10-10, 04:54 PM

One more note. DPR is an inherently flawed indicator. The goal is to kill your opponents before you run out of HP, not to kill them faster.

Yeah, this. Surprising how often analyses go wrong on this point. DPR is meaningful when comparing two options with equivalent defense (e.g. "should I power attack this round?") but it's generally a poor metric for comparing more complicated options such as builds.

I die a little inside every time I see someone write "assume X rounds of combat in an adventuring day."

Strill

2015-10-10, 04:55 PM

I die a little inside every time I see someone write "assume X rounds of combat in an adventuring day."What are you talking about? That's what the game is explicitly designed around.

MrStabby

2015-10-10, 07:35 PM

Another common mistake is to think that personal defence is important. You can be as tough as you like but if the enemy just consistantly wipes the rest of the party no one is going to have a good time.

If you are the easiest to kill (obviously taking into account role, positioning etc. as well as AC and hitpoints) then there is an advantage to higher AC. If you are the big slow guy who not one is prioritising as a target anyway then there isn't much advantage to being better able to resist the attacks that no one is throwing against you.

But yes, to the OP. Surprise rounds and going ahead of the enemy in initiative order is huge. Fitting in an extra round of combat at the start of a fight is a huge bonus only partially offset by fewer good options of fighting style. Having two extra rounds is often a straight win there.

The problem isn't that TWF is bad vs GWF and Polearms but that it is bad vs duelist style+swords and board equipment (from mid levels onwards) which get a decent enough effect AND can use dex (and they can still have a bonus action free on top).

Mara

2015-10-10, 08:23 PM

What are you talking about? That's what the game is explicitly designed around.No not really.

Malifice

2015-10-10, 10:04 PM

No not really.

Yes it really is designed around it. The classes balance around it, encounters are rated against it etc.

You can ignore it if you wanf. But the game is designed around certain expectations of encounters/ rests/ adventuring day.

Doof

2015-10-10, 10:13 PM

yeah the general consensus is, that since people don't know what to expect about the number of encounters per day or in some cases whether they'll have a good night's rest or not. Therefore it is always imperative to maximise your damage per round to eliminate threats faster, not 'kill it just whenever before you die'.

Tanarii

2015-10-10, 10:32 PM

You can ignore it if you wanf. But the game is designed around certain expectations of encounters/ rests/ adventuring day.he said rounds per day. Not encounters/rests per day. There's a fairly huge difference between accepting the DMG guidelines for number of encounters and number of rests (which btw are variable depending on difficulty of said encounters), and assuming the length of any given encounter will be X rounds.

Tanarii

2015-10-10, 10:35 PM

Therefore it is always imperative to maximise your damage per round to eliminate threats faster, not 'kill it just whenever before you die'.
Nonsense. It's imperative to finish an encounter with the least expenditure of resources. A shorter number of rounds per encounter absolutely can and usually will make that happen. But it's by no means ensured. Higher defense / crowd control / mobility, with lower offense, and taking more time, may achieve a better result.

Malifice

2015-10-10, 10:45 PM

he said rounds per day. Not encounters/rests per day. There's a fairly huge difference between accepting the DMG guidelines for number of encounters and number of rests (which btw are variable depending on difficulty of said encounters), and assuming the length of any given encounter will be X rounds.

In my experience a medium-hard encounter for 4-5 PCs lasts 3-4 rounds on average. My PCs get around 10 rounds / 2 encounters per short rest, and around 6 encounters per long rest (30 rounds per day). The AD Xp awards are hitting the recommended mark. My casters are really struggling in the longer days, and have the edge on the shorter ones.

It's balanced really well in my view.

Doof

2015-10-10, 10:48 PM

Nonsense. It's imperative to finish an encounter with the least expenditure of resources. A shorter number of rounds per encounter absolutely can and usually will make that happen. But it's by no means ensured. Higher defense / crowd control / mobility, with lower offense, and taking more time, may achieve a better result.

My wording was highly misleading and incorrect because I omitted the resource management side of the issue. However, I stand by my claim that it is not enough to just 'kill things any time before you die' is not a good way to handle adventures when you don't know what you'll get before you can recuperate.

That comment was made in response to somebody saying that DPR is meaningless and that you only need to kill things before you die.

Edit: still, thank you for your input. I appreciate it.

Doorhandle

2015-10-10, 10:56 PM

Also, another thing to note: TWF is still slightly more useful for crit fishing. On it's own it's not much, but combining say, advantage and the champion's improved/superior critical and it will provide an edge.

The main reason it may seem to be worse is that there are now a lot of class features which can duplicate it, while still allowing you to benefit from the damage bonus AND any bonuses from the weapon type; like the path of the berserker, or heck, flurry of blows.

Tanarii

2015-10-10, 10:57 PM

however, I stand by my claim that it is not enough to just 'kill things any time before you die' is not a good way to handle adventures when you don't know what you'll get before you can recuperate.I agree that generally speaking, if you have a balanced party (as opposed to one designed specifically to kite), killing faster is more effective than wearing your target down. Especially because targets tend to be fully combat effective until killed.

Edit: still, thank you for your input. I appreciate it.
sorry to toss the 'Nonsense' your way, that's aggressive wording. Appreciate you being so polite. ;)

Malifice

2015-10-10, 11:09 PM

It's good on a paladin and a rogue. Extra chances to land smites/ sneak attack. Battle masters can get some good use out of it too.

bid

2015-10-10, 11:30 PM

Also, another thing to note: TWF is still slightly more useful for crit fishing.
How can it be useful if you're doing the same average damage per attack?

Whether you are doing 3 attacks with an average of 2 damage, or 2 attacks with an average of 3 damage, it's the same thing. Your crit fishing will only do 6 damage for twf while it does 9 damage for gwf.

In fact, twf is closer to average while crit fishing seems to be hoping for extremes.

Strill

2015-10-10, 11:38 PM

How can it be useful if you're doing the same average damage per attack?

Whether you are doing 3 attacks with an average of 2 damage, or 2 attacks with an average of 3 damage, it's the same thing. Your crit fishing will only do 6 damage for twf while it does 9 damage for gwf.

In fact, twf is closer to average while crit fishing seems to be hoping for extremes.For most characters, you're right. For Paladins, crit fishing matters because you can choose to smite AFTER you see whether you get a crit or not.

bid

2015-10-11, 01:02 AM

For most characters, you're right. For Paladins, crit fishing matters because you can choose to smite AFTER you see whether you get a crit or not.
Oh, now I see. Thanks.

djreynolds

2015-10-11, 03:55 AM

Strength is awesome for shoving and not being shoved. Acrobatics only allows you not to get shoved, knockdown is huge.

TWF is good option, for rogue's who have one attack and are now jumping in on the guy who just got knocked down because his strength save sucks or his acrobatics did "not" beat the guy with proficiency in athletics. Two strikes with advantage gives you a better opportunity to perhaps land a sneak attack that could be 8d6, that's fireball's worth of damage with no save or evasion or resistance, but it is still one strike with advantage. But with TWF, it could be two and its costs you nothing (no feat) to have two daggers drawn in combat, who cares about the ability modifier. One swing, sneak attack lands, bonus action disengage. One swing miss, bonus action off-hand, sneak attack lands.

To me, IMHO, it is about options for the use of your bonus action. Its not perfect and TWF does not scream blender of death, which is sad.

Also expertise in athletics, even for a dexterity based fighter can be huge. Its a minimum of +12 at level 20 with 10 in strength. That same martial type, excluding the barbarian, with a 20 in strength for +5 is getting +11 at level 20. So a dexterous valor bard with a shield and shield master can still shove, especially with expertise in athletics and the bless spell "stolen" from the cleric class.

And the alert feat is the same initiative bonus as +5 from a 20 in dex.

What I like best about 5E are the nooks and crannies, and not the overall optimization of "builds". TWF comes off more realistic than it did in 3.5 instead of ninja's, and not the one fighting samurai, but the blender on TV. Squeezing out all the nutrients.

GoodbyeSoberDay

2015-10-11, 04:01 AM

Yeah, this. Surprising how often analyses go wrong on this point. DPR is meaningful when comparing two options with equivalent defense (e.g. "should I power attack this round?") but it's generally a poor metric for comparing more complicated options such as builds.

I die a little inside every time I see someone write "assume X rounds of combat in an adventuring day."It's a metric. Take some amount of rounds you might expect to occur in a day and see how much damage you do in that time. People might focus on it too much, but it does say something about your option to bring things to 0 HP over the day in a way that is easy to interpret.

Yes, an ideal metric for build effectiveness would measure likelihood of victory in a representative sample of combats each level, with some sensitivity analysis for party composition and encounter distribution. Unless you care to do that we're going to have to measure things each piece at a time.

Kryx

2015-10-11, 04:30 AM

Yes, an ideal metric for build effectiveness would measure likelihood of victory in a representative sample of combats each level, with some sensitivity analysis for party composition and encounter distribution. Unless you care to do that we're going to have to measure things each piece at a time.
Exactly. There are better metrics to be made, but they are significantly more complicated and a lot more work.
Max keeps repeating this same quote in most threads about DPR. And yet, as pointed out, it is the balance factor that WotC uses themselves.

On topic: Dex isn't better than Str.

Saves

There are 22 abilities/spells based on Str. There are 55 saves from monsters.
There are 53 abilities/spells based on Dex. There are 51 saves from monsters.

Dex is almost always half damage. Strength is almost always avoiding prone/restrained.
Numbers here (https://docs.google.com/document/d/1jH2H_6ijTkQzdMUz8bUvKYUQtwfViqHiiwHk2oSgMvo) (dark yellow is stuff I'm changing for consistency - not finalized)

AC
Dex starts out at 12+3 = 15 AC. Heavy starts at 16 (Chain mail). Medium starts at ~14-16 (14+2).
Around level 3-4 heavy gets plate which adjusts his AC to 18. Dex gets 16 AC at 4 if they pump dex and then 17 AC at 8 if they pump dex again. They can only get 18 AC if they take TWF. That's 2 ASIs and a feat to equal a gold cost.

Damage
Damage is heavily in Strength's favor. Dex Fighter/Paladin does significantly less than a str paladin.
S&B Fighter does ~75-85% the damage that a Polearm Fighter does.
TWF Fighter does ~71-84% the damage that a Polearm Fighter does.
S&B Paladin does ~80-85% the damage that a Polearm Paladin does.

There really is no competing in damage.

Initiative
Dex gets it, Strength doesn't.

TL;DR: Strength heavily wins out on damage. Strength wins on AC. Dexterity wins slightly on saves. Dexterity wins on Initiative.
Overall Strength is definitely stronger.

TWF needs buffs to be competitive. I'm experimenting with several changes to adjust the math:

Swap TWF Fighting Style and Dual Wielder. So Fighting style gives bigger weapons (+1 dmg) and dual wielder gives stat on offhand
TWF gets Rend at level 11 (Double proficiency extra damage if both the main and offhand hit)
TWF Fighting style gets +1 hit

The math works out much better.

djreynolds

2015-10-11, 05:25 AM

Since you only get one sneak attack per turn, TWF is perfect rogues and costs them nothing. You don't need a feat to duel daggers. You just want that sneak attack to land.

But it is all relative. It loses steam at higher levels, but that okay. But for guys without feats yet in early levels, it adds some potency to their strikes and use for a bonus action. No reason a 3rd level barbarian can't duel short swords with strength while he's waiting on GWM or whatever. And for rangers its thematic and accurate. Watch the Patriot or Last of the Mohicans

Mr Kryx's idea on two weapon rend is sweet. I would allow them an option, damage, shove, or disengage if both strikes hit. That way dexterity guys have two options, and strength guys have two options. Like a free leg kick to push off or push down.

MrStabby

2015-10-11, 05:33 AM

On topic: Dex isn't better than Str.

[/SPOILER]

Skills as well are important. Particularly skills that help you sneak about to get a surprise round.

Kryx

2015-10-11, 05:46 AM

Skills as well are important. Particularly skills that help you sneak about to get a surprise round.
True.

Athletics & Strength checks vs Acrobatics, Stealth, Sleight of Hand, and Dex checks

Athletics is Climb, Jump, Swim, shove(prone or forced movement)
Strength checks is breaking doors or chains, dragging
Acrobatics is balance
Sleight of hand is stealing (very niche)
Stealth is stealth (harder to do in a group setting, depends on DM/players)
Dex checks are to pick locks/disable traps and other nimble hands stuff.

Overall I find those comparable. Athletics more so if the DM asks for those type of checks often.

djreynolds

2015-10-11, 05:47 AM

Skills as well are important. Particularly skills that help you sneak about to get a surprise round.

Very true, you cannot have it all. But that's fun of it. Getting what you need. Gritty.

Tanarii

2015-10-11, 11:01 AM

And yet, as pointed out, it is the balance factor that WotC uses themselves.you keep repeating this is various threads, and it's still not true. The "balance factor" that WOTC uses only gives guidelines for rests per day, and a way to balance the total number of encounters. Not rounds per encounter.

Your metric uses your personal guesses about % of which difficulty encounters happen, and length (in rounds) of each type of encounter. That makes it a *pointless* metric. It doesn't matter if other people suggest alternative values for those, as they'll be guessing just as much as you are.

Furthermore your encounters per day (5.something) *explicitly* disagrees with what WOTC says in their documentation, which is 6-8 encounters per day. I agree that number doesn't take into account variability in encounter difficulty, but again, that just furthers my point: you are guessing an unguessable variable (in this case distribution of encounter difficulty), making your metric pointless.

Kryx

2015-10-11, 03:21 PM

you keep repeating this is various threads, and it's still not true.
5e is balanced on an adventuring day. That's what I keep repeating. Many people try to ignore this.

The "balance factor" that WOTC uses only gives guidelines for rests per day, and a way to balance the total number of encounters. Not rounds per encounter.
This has been discussed in other threads. Please see the DMG guidelines:

Assuming typical adventuring conditions and average luck, most adventuring parties can handle about six to eight medium or hard encounters in a day. If the adventure has more easy encounters, the adventurers can get through more. If it has more deadly encounters, they can handle fewer.
It recommends 6-8 medium or hard encounters in a day. Since they recommend mostly medium or hard my metric assumes that the majority are medium or hard. That means 40% medium, 40% hard with some easy (10%) and deadly (10%). You could argue for some small adjustments, but it has very little effect on the outcome unless you stray far from the guidelines.

Now if you look on the same page (DMG 84) it also shows how much XP should be handled in a day. For example if we take 10th level it recommends 9,000 adjusted XP per day.
Look above to DMG 82 and you'll see how much XP is given per each encounter type. Easy is 600, medium 1,200, hard 1,900, and deadly 2,800. Using the 10/40/40/10 metric that equals out to 1580 xp per encounter on average (.1*600 + .4*1200 + .4*1900 + .1*2800). To get to 9000 XP per day it would take 5.7 encounters in that day (9000/1580). Do that again for all levels and then average them out and you'll end up at 5.13 encounters per day.
The only way to get more encounters per day would be to have more easy encounters and less of everything else. Using the DMG's guidelines you're going to be between 4-6 encounters per day on average.

Now the only questionable part that I do is estimate # of rounds per encounter type. I assume Easy is either 1,2,3 (average 2), Medium is 3,4,5 (average 4), Hard is 5,6,7 (average 6), and deadly is 7,8,9 (average 8).
I would love to see further analysis around this area, but I highly doubt that these numbers are too low. If anything I'd expect them to be too high on average. If they are higher and were adjusted to even 1 lower for each tier then the average number of rounds in a day would drop from 25.7 to 20.5

It doesn't matter if other people suggest alternative values for those, as they'll be guessing just as much as you are.
The only part I'm guessing on is the number of rounds/encounter. The other part is based on their recommendations.
I doubt you'll take me up on the offer but I would love to see what you'd estimate for length of each encounter type - most likely the end result would be less rounds in a day than I currently have.

Furthermore your encounters per day (5.something) *explicitly* disagrees with what WOTC says in their documentation, which is 6-8 encounters per day. I agree that number doesn't take into account variability in encounter difficulty, but again, that just furthers my point: you are guessing an unguessable variable (in this case distribution of encounter difficulty), making your metric pointless.
Unfortunately the math provided by their charts disagrees with their text. If we assume a 50/50 split between medium and hard encounters only using their tables as data points then the average is 5.23 encounters per day. You'd have to do 70% easy encounters to get to 7.96 average encounters per day (assuming the other 3 are equal in frequency). You'd have to do 45% easy encounters to get to 6 encounters per day (assuming the other 3 are equal in frequency).

TL;DR: The DMG xp guidelines and adventuring day guidelines in the tables disagrees with the 6-8 in the text. I've assumed their math provided is more accurate than the words they provided. Given that it is nearly impossible to get anything outside of 4-6 encounters per day. From there it would require large round duration assumptions to get more rounds per day than I have.

Edgerunner

2015-10-11, 04:11 PM

I am seeing here that TWF is not that good for straight up Fighters but works well with Rogues.

If this assumption is correct... what does it mean for the Swashbuckler (Rogue subclass) ?

Swashbuckler
You focus your training on the art of the blade, relying
on speed, elegance, and charm in equal parts. While
some warriors are brutes clad in heavy armor, your
method of fighting looks almost like a performance. Duelists
and pirates typically belong to this archetype.
A Swashbuckler excels in single combat, and can
fight with two weapons while safely darting away from
an opponent.
Fancy Footwork
When you choose this archetype at 3rd level, you learn
how to land a strike and then slip away without reprisal.
During your turn, if you make a melee attack against a
creature, that creature can’t make opportunity attacks
against you for the rest of your turn.

If I am interpreting this correctly, the Swashbuckler should get the same benefit of Fighter's TWF Style which would give them the extra offhand attack, at no negatives, and they get to add their Dex to hit/damage as long as they are using two finesse weapons. They also get to disengage without any worries of AoO. Couple this with the statements made earlier about SA and this looks seriously legit.

If I am understanding this correctly it could be very advantageous indeed.

NNescio

2015-10-11, 04:19 PM

Now the only questionable part that I do is estimate # of rounds per encounter type. I assume Easy is either 1,2,3 (average 2), Medium is 3,4,5 (average 4), Hard is 5,6,7 (average 6), and deadly is 7,8,9 (average 8).
I would love to see further analysis around this area, but I highly doubt that these numbers are too low. If anything I'd expect them to be too high on average. If they are higher and were adjusted to even 1 lower for each tier then the average number of rounds in a day would drop from 25.7 to 20.5

I'm not sure how much this matters (small sample size), but my personal experience on both sides of the DM screen mostly agrees with your estimates (slightly longer, but not by much).

Yeah, generally, a 1 min buff (or some other effect) can last the entire combat.

The only exception are long drawn-out kiting affairs, but usually that only happens in poorly-balanced encounters (again, my experience).

Coidzor

2015-10-11, 04:25 PM

NNescio

2015-10-11, 04:28 PM

What you've posted here just means that if they hit an opponent they can use their bonus action to TWF and not have to worry about instead wanting to have used it to Disengage because if their opponent can't take an Opportunity Attack against them, there's no need to Disengage.

They don't actually have to hit. Just attempt to. That's the beauty of it.

Edgerunner

2015-10-11, 04:32 PM

What you've posted here just means that if they hit an opponent they can use their bonus action to TWF and not have to worry about instead wanting to have used it to Disengage because if their opponent can't take an Opportunity Attack against them, there's no need to Disengage.

A Swashbuckler excels in single combat, and can
fight with two weapons while safely darting away from
an opponent.

I read this as being able to TWF and Disengage.

NNescio

2015-10-11, 04:36 PM

A Swashbuckler excels in single combat, and can
fight with two weapons while safely darting away from
an opponent.

I read this as being able to TWF and Disengage.

Anybody can TWF as long as they hold the appropriate weapons. There's nothing in that part of the text that gives you the Two-Weapon Fighting fighting style (a well defined term) for free.

It's flavor text. Nothing pertaining to game mechanics is expressed within the bolded line. It certain does summarize the Swashbuckler's abilities though (but those abilities are defined mechanically later on).

Kryx

2015-10-11, 04:42 PM

I am seeing here that TWF is not that good for straight up Fighters but works well with Rogues.
Hand Crossbows work even better by RAW.

If this assumption is correct... what does it mean for the Swashbuckler (Rogue subclass) ?
Swashbuckler TWF is just fluff. The only great thing they get is permanent sneak attacking. There is nothing about getting full stat bonus on the second attack.

my personal experience on both sides of the DM screen mostly agrees with your estimates (slightly longer, but not by much).
Cool, thanks for the feedback!

Tanarii

2015-10-11, 04:45 PM

5e is balanced on an adventuring day. That's what I keep repeating. Many people try to ignore this.no they don't. They point out you make a jump from encounters / adventuring day --> rounds / adventuring day, based on a series of assumptions that are NOT contained in the rules. As the rest of your post reiterates, yet again. You make assumptions on the distribution of encounter difficulty, and avg rounds per combat.

You assume factors in the balancing, and that makes your numbers meaningless. It's not a matter of if its too high or two low. It's that there is no way of knowing how accurate or inaccurate your assumptions are, and that renders them, generally, a moot point.

Just looking at encounter difficulty distribution alone, staying within the guidelines for an average day, we can vary from 75% easy encounters (remainder medium) to 50% deadly (remainder hard). In fact, it's almost pointless to say "10%" Deadly ... a single Deadly encounter uses up far more of the daily budget than 10%. So there will be a difference between *any* adventuring day containing one or more Deadly encounters and one that does not.

This has been discussed in other threads. Please see the DMG guidelines:Read them. You are adding things to them that aren't actually there ... specific encounter difficulty distributions, and rounds per encounter.

Kryx

2015-10-11, 05:05 PM

They point out you make a jump from encounters / adventuring day --> rounds / adventuring day
Again I'd ask you to better estimate the average duration of each encounter type. You'd have to choose estimates that are vastly different than mine to get different results. If your idea is that there are more rounds per day on an average adventuring day then you'd have to estimate significantly more per encounter type. You seem to be opposed to doing so, but I think it would be very enlightening.

Overall my metric only makes assumptions based on rounds per encounters - that's it. Nothing wildly outlandish.

we can vary from 75% easy encounters (remainder medium) to 50% deadly (remainder hard). In fact, it's almost pointless to say "10%" Deadly ... a single Deadly encounter uses up far more of the daily budget than 10%. So there will be a difference between *any* adventuring day containing one or more Deadly encounters and one that does not.
It's an average. The whole point of DPR is to use an average. 10% Deadly means you'll have a deadly encounter every few days, but not all days.

Overall your argument feels nitpicky. I highly doubt your numbers would vastly differ from mine if you made a rounds/encounter type estimate. If your stance is "it can't be done", then I'll continue using the best estimates as that is the only way of measuring it given the information we have on 5e. You can choose to refuse any imperfect metric.

Tanarii

2015-10-11, 09:53 PM

Again I'd ask you to better estimate the average duration of each encounter type.I'm not going to guess and thereby create an equally invalid but different value. That's silly. Then we just have two value that are useless, instead of one.

I'm not assuming longer or shorter on ANY length in combat rounds for an encounter. I'm saying that guessing a number produces a potentially meaningless value. Furthermore it creates this illusion of being meaningful. After all, now you've got a number, must mean something, right?

It's an average. The whole point of DPR is to use an average. 10% Deadly means you'll have a deadly encounter every few days, but not all days.And using an average in many cases, like this one, means you have a useless number as a result. Sure, it's an average number. But not necessarily one with any applicable use.

Overall your argument feels nitpicky. I highly doubt your numbers would vastly differ from mine if you made a rounds/encounter type estimate. If your stance is "it can't be done", then I'll continue using the best estimates as that is the only way of measuring it given the information we have on 5e. You can choose to refuse any imperfect metric.It's not a best estimate, it's a flawed model incorporating constants in the place of variables. If you want to create a model that gives some potentially meaningful results, you'd have to incorporate a range of encounter difficulties for the average day, along with a range of encounter rounds within each that seems meaningful, and output an array of results. Even then you're still going to run into the classic DPR flaw, assumption of target AC, more than usual, because With resources that you increase DPR with, you'll only use when you think you can actually hit. Then you have to find a way to account for battlefield movement and terrain that effect hit chance, and spells that aren't single target damage but affect your characters single target damage. And once you've done all that you've got something that, if it's effective, is only useful for classes that use physical attacks or spam cantrips.

In short ... I think it's not possible to calculate any meaningful number for DPR across a day as a singe value. Trying to do so effectively allows an appeal to mathematical authority fallacy. Making decisions on it, and promoting ideas on the forums based on it, is potentially harmful.

bid

2015-10-11, 10:49 PM

In short ... I think it's not possible to calculate any meaningful number for DPR across a day as a singe value. Trying to do so effectively allows an appeal to mathematical authority fallacy. Making decisions on it, and promoting ideas on the forums based on it, is potentially harmful.
I am going to argue that attempting to shut someone off is actually harmful. Do you have anything constructive to say or is it just whinings?

Malifice

2015-10-11, 10:56 PM

Now if you look on the same page (DMG 84) it also shows how much XP should be handled in a day. For example if we take 10th level it recommends 9,000 adjusted XP per day.
Look above to DMG 82 and you'll see how much XP is given per each encounter type. Easy is 600, medium 1,200, hard 1,900, and deadly 2,800.

I'm not sure youre doing this correct.

The XP thresholds you cite above are not how much XP is awarded in that encountner - just the adjusted XP to determine the difficulty of the encounter. Multiple creatures increase the XP for difficulty purposes - it doesnt change how much you actualy award.

Im fairly sure the 'Adventuring Day XP' table is awarded XP; not adjusted difficulty XP.

Tanarii

2015-10-11, 11:30 PM

I am going to argue that attempting to shut someone off is actually harmful. Do you have anything constructive to say or is it just whinings?Pointing out why something is potentially destructive *is* constructive. And I'm clearly not the only one who thinks so, as I've seen multiple others object to the idea of a combat rounds per day based DPR.

Ardantis

2015-10-11, 11:38 PM

To get back on topic, I have to agree with what has been stated earlier about TWF utility for non-specialized characters, namely rogues, paladins, and early-game barbarians.

The extra chance to land sneak attack, smite, and rage damage is very valuable in certain situations, and equipping yourself so does not dedicate you to a certain build. In each case, there is no pressing need for a magical offhand item and the benefit is not hampered by the smaller damage dice of light weapons.

A dedicated build for the purpose of more damage using full damage modifiers and larger dice is really only possible in fighter and ranger because of the necessity of the fighting style, and in the case of a strength-based character is singularly less effective than great weapon style or polearm.

A champion fighter who goes dex and crit-fishes with two weapons, using stealth to try and get surprise rounds as often as possible, will be very effective at taking out groups of medium-to-heavy mooks. A dex ranger who goes hunter with two weapons will be extremely front-loaded, but will level off later in the game.

Either way it's a particular style of play which has its place, but I'd still want to have a primary bonecrusher tank in the party because you need to deal with big bads and TWF just doesn't cut it for those purposes.

bid

2015-10-11, 11:48 PM

Pointing out why something is potentially destructive *is* constructive. And I'm clearly not the only one who thinks so, as I've seen multiple others object to the idea of a combat rounds per day based DPR.
Well maybe the Earth is flat too. If it is that important, why don't you start a thread to demonstrate how DPR is harmful instead of polluting every thread. I am sure it is easy to point out how variations in combat per short/long rests, monster's AC and hp, et al. makes DPR meaningless. So go ahead and demonstrate this in its own thread.

Or I could derail this thread further by going meta and demonstrate how "constructively" derailing threads is actually harmful.

BTW Kryx, you should redirect them to another thread instead of replying in-thread. Most of them will shut up if you don't help them derail threads.

Kryx

2015-10-12, 03:05 AM

Kryx, you should redirect them to another thread instead of replying in-thread. Most of them will shut up if you don't help them derail threads.
I probably should make a generic "DPR" thread to discuss these issues. I will do so in the future. This thread has basically run its course as I can show TWF being significantly uncompetitive by any DPR metric.

Im fairly sure the 'Adventuring Day XP' table is awarded XP; not adjusted difficulty XP.
To make sure we're discussing the same thing:
DMG 84 shows an "Adventuring Day XP" table. The XP in that table is labeled as "Adjusted XP per Day per Character" (emphasis mine).

That should be what you expect, right?

Either way it's a particular style of play which has its place
There is not a single point where TWF excels. It should, but it doesn't. Even S&B style beats TWF.

Malifice

2015-10-12, 03:17 AM

DMG 84 shows an "Adventuring Day XP" table. The XP in that table is labeled as "Adjusted XP per Day per Character" (emphasis mine).

Ah OK. It is adjusted for difficulty, not awarded. Was AFB.

CNagy

2015-10-12, 04:45 AM

Discussions like this are why theorycrafting takes the backseat to actual in-game experience. Actually playing the game has, in my experience (thus making this the sort of anecdotal evidence which has little place in the realm of number-crunching) shown two weapon fighting to be no less of a benefit to party success than any of the other options. Within a party round (because you don't want to let a creature live round to round unless you can help it), the TWFighter seems more often to hand the assist to heavier damage-dealing characters or lands the finish themselves on heavily-damaged creatures. Having a TWFighter on the front lines generally, but not always, means doing less overkill. Effective damage is a harder thing to measure than raw damage, but it has more in-game relevance.

So as a solo character, no, the TWFighter is not going to top any charts. I've only seen them become a major individual force in campaigns where I've gone more old school with the treasure. But as one element of a well-rounded party, I've seen them do consistently well.

Doof

2015-10-12, 05:43 AM

Discussions like this are why theorycrafting takes the backseat to actual in-game experience.

Yup, I admit 3.5e was just so much fun theorycrafting it made me into an armchair adventurer :(

Mara

2015-10-12, 06:29 AM

It seems to me that without the feat, TWF is meant to be a dex build.

Comparing dex builds eliminates GWF leaving dueling Fighters or Hunters.

Let's just compare fighters.
Dueling v TWF
At all levels dueling either has a free hand or +2 AC and a free bonus action. TWF has more attacks per round.

At level 1:
1d8+5 v 2d6+6 TWF does 3.5 more damage.

At level 6:
2d8+14 v 3d6+15 TWF does 2.5 more damage

At level 11:
3d8+21 v 4d6+20 Dueling does .5 more damage

At level 20
4d8+28 v 5d6+25 Dueling does 3.5 more damage

Now to compare optimal damages. GWM GWF vs dual lances mounted. Assume GWF is ~+2 DPR per hit. Let's say they are both mounted.

GWM v GWF alone v TWF
All have advantage against unmounted medium or smaller foes after level 4.

At 6:
46 v 26 v 31.5

At 11:
72 v 42 v 46

At 20:
96 v 56 v 57.5

It's up to you how the average advantage and the -5 effect damages. Both depends on enemy defenses/size/ect.

Kryx

2015-10-12, 06:44 AM

Comparing dex builds eliminates GWF leaving dueling Fighters or Hunters.
A TWF's role is damage. It has no other advantage over strength as I listed earlier. There is no reason it shouldn't be comparable.

A Dex Duelist has +2 AC from a shield (above plate, below plate&shield). A TWF's has +1 AC (equivalent to plate).

Let's just compare fighters.
Dueling v TWF
Your numbers seem to have forgotten to include shield master. Shield master alters the math significantly.

No other resource numbers, just straight attack DPR (though I do not agree that this is a good metric):
1
TWF does 8.9 DPR
Dex S&B does 6.4 DPR

5
TWF does 14.2 DPR
Dex S&B does 14.8 DPR

11
TWF does 25.6 DPR
Dex S&B does 28.2 DPR

17
TWF does 25.6 DPR
Dex S&B does 28.2 DPR

20
TWF Does 32.0 DPR
Dex S&B does 37.6 DPR

Both TWF and S&B have a feat cost (TWF has Dual wielder, S&B has Shield Master).

TWF loses at all tiers except 1 and it has lower AC.

Now to compare optimal damages. GWM GWF vs dual lances mounted. Assume GWF is ~+2 DPR per hit. Let's say they are both mounted.

GWM v GWF alone v TWF
All have advantage against unmounted medium or smaller foes after level 4.
Dual mounted lances is such a niche build that is entirely campaign and DM dependent. It only works against medium or smaller creatures in open terrain. It is not a viable build for most situations and therefore not a good metric.
Assuming +2 DPR per hit from GWF isn't accurate either. It would change weapon die from 7 to 8.33 for Greatsword or 6.5 to 7.3 for Lance. Much less than 2 DPR.

Citan

2015-10-12, 09:59 AM

On topic: Dex isn't better than Str.

Saves
Dex is almost always half damage. Strength is almost always avoiding prone/restrained.

AC
Dex starts out at 12+3 = 15 AC. Heavy starts at 16 (Chain mail). Medium starts at ~14-16 (14+2).
Around level 3-4 heavy gets plate which adjusts his AC to 18. Dex gets 16 AC at 4 if they pump dex and then 17 AC at 8 if they pump dex again. They can only get 18 AC if they take TWF. That's 2 ASIs and a feat to equal a gold cost.

TL;DR: Strength heavily wins out on damage. Strength wins on AC. Dexterity wins slightly on saves. Dexterity wins on Initiative.
Overall Strength is definitely stronger.

A TWF's role is damage. It has no other advantage over strength as I listed earlier. There is no reason it shouldn't be comparable.

Hi!

I'm sorry but that is your personal opinion, yet you state it as a definitive fact.

On Two-Weapon Fightning
First, TWF's role is not "damage", it's "versatility": bringing an extra attack from the get-go, as a bonus action, for classes which don't get a good use for their bonus action until much later and/or don't get (m)any Extra Attacks.

As others said already, sometimes it's great to inflict huge damage to a target (point to two-handed), other times it's greater to be able to hurt and finish off two targets.
Sure, once you get the GWM feat AND are potent enough to a) hit reliably b) get the bonus attack on a consistent basis, it becomes strictly superior to TWF. But it requires maxed attack stat, maybe self or ally's buff on hit (advantage or to hit buff) and a feat. Not coming online so early.
TWF in comparison is online from lvl 1 onwards, the Feat being actually pretty dispensable damage-wise imo. And is much better in first tier with some classes such as Rogue or Ranger who benefit early from an extra attack.

Secondly, in your last post, you compare with Shield Master feat. I'd like the detail because I don't understand, do you suppose the shove always work?

On STR / DEX
Thirdly, not the STR doesn't "win" on AC. Because you have many, many easy ways to get extra defense, either permanent or temporary, which are based on Dexterity, making it real easy to get actual defense as good or better as being in heavy armor, without its usual speed, stealth or vulnerability drawbacks: feats, class abilities, spells...

Also, on saves, sure STR makes you prone which is annoying and can be dangerous, however abilities that makes you restrained are actually rare.
And DEX-based classes usually comes with enough mobility to actually avoid being taken in such spells.
Whereas DEX-based damage dealing spells are often long-range spells and among the ones you'll "meet" the most often...

All this makes a correctly built DEX-based character as viable as the heavy armor tank in most occasions, as long as he thinks a minimum about priorities and positioning. So, generally speaking, STR is not better than DEX for survivability, by far.
And, frankly, a sentence such as "Overall Strength is definitely stronger" has no meaning whatsoever (maybe you were thinking about the damage aspect when writing this?). :)

----
By the way, I peeked at your sheet, it's great. Thank you very much for sharing this, must have required a a good dozen hours of work. Hat down to you for this. :)

Kryx

2015-10-12, 10:17 AM

I'm sorry but that is your personal opinion, yet you state it as a definitive fact.
Please correct any "opinion" you see. Nothing you quoted in the saves/AC/TL;DR section is opinion. If you're talking about DPR then I can show that via any metric - "straight DPR" or 5e's system of the adventuring day. In both TWF loses to GWM and S&B (Str or Dex).

First, TWF's role is not "damage", it's "versatility"
What "versatility" does TWF get? Compare to Polearm. They get less attacks (no provoke), less damage, less AC until later levels, less reach.
Splitting damage among multiple targets? Hardly worth any more than 5% in overall damage imo. Currently the numbers are 20%+.
The only potential benefit is level 1-3 on a non-variant human.

As others said already, sometimes it's great to inflict huge damage to a target (point to two-handed), other times it's greater to be able to hurt and finish off two targets.
This would matter if TWF actually had more attacks than another fighting style. It doesn't.

Secondly, in your last post, you compare with Shield Master feat. I'd like the detail because I don't understand, do you suppose the shove always work?
Nope. Shove has a 60% chance to work and can be used before the main attacks. (I should likely assume prof in athletics vs standard save instead of flat 60%, but it'll be nearly the same I think - I'll do that now)

Thirdly, not the STR doesn't "win" on AC. Because you have many, many easy ways to get extra defense, either permanent or temporary, which are based on Dexterity, making it real easy to get actual defense as good or better as being in heavy armor, without its usual speed, stealth or vulnerability drawbacks: feats, class abilities, spells...
Strength has a higher AC. Dex has a few like defensive duelist, but all of those come at a cost. If there was no cost to some of those options you'd have an argument. If you still think Dex is better somehow then compare a Duelist Dex vs a TWF dex and you'll see a Duelist gets more damage and more defense than a TWFer (the topic of this thread).

Also, on saves, sure STR makes you prone which is annoying and can be dangerous, however abilities that makes you restrained are actually rare.
No they aren't. There are 10 Restrained Strength saves in the game.

And DEX-based classes usually comes with enough mobility to actually avoid being taken in such spells.
That is not how spells work... They allow a Strength save, not a Dex save to avoid.

Whereas DEX-based damage dealing spells are often long-range spells and among the ones you'll "meet" the most often...
RAW Strength has 78 Saves from spells/abilities. Dex has 116. Dex is bloated by 20 just on Dragons and another 5 on beholders. Strength has lots that are long range.

All this makes a correctly built DEX-based character as viable as the heavy armor tank in most occasions
Not at all. They still have lower AC even if they have 1 reaction per turn to raise AC or diminish damage at a high cost.

a sentence such as "Overall Strength is definitely stronger" has no meaning
I outlined the criteria. You seem to disagree with that criteria, but based on that criteria Strength makes a more useful character for a party.

By the way, I peeked at your sheet, it's great. Thank you very much for sharing this, must have required a a good dozen hours of work. Hat down to you for this. :)
Thanks!

MrStabby

2015-10-12, 10:31 AM

I think initiative is underrated here. The chance to get an extra round of combat in from having higher dex is a huge bonus. I find most combats to be about 3 rounds - getting an extra turn is then a 25% increase in damage.

I also find it odd to assume that Str adds to armour. My latest campaign are now all at level 7 and no one has full plate armour yet. The assumption that any character can get whatever equipment they want is probably reasonable for anything under 100gp but I wouln't be so sure beyond that.

toapat

2015-10-12, 10:33 AM

RAW Strength has 78 Saves from spells/abilities. Dex has 116. Dex is bloated by 20 just on Dragons and another 5 on beholders. Strength has lots that are long range.

How many of these saves, each, are exclusive to Only Dex or Only Str?

Also, claiming that Dexterity has lower armor then Strength: No. Mage Armor exists, High Elf and variant Human exist, Anyone can take Magic Initiate., and unless some massive change in the unwritten rules happened since the DMG came out, you cant assume magical items, which means no +3 armor.

Citan

2015-10-12, 11:21 AM

Please correct any "opinion" you see. Nothing you quoted in the saves/AC/TL;DR section is opinion. If you're talking about DPR then I can show that via any metric - "straight DPR" or 5e's system of the adventuring day. In both TWF loses to GWM and S&B (Str or Dex).

What "versatility" does TWF get? Compare to Polearm. They get less attacks (no provoke), less damage, less AC until later levels, less reach.
Splitting damage among multiple targets? Hardly worth any more than 5% in overall damage imo. Currently the numbers are 20%+.
The only potential benefit is level 1-3 on a non-variant human.

Strength has a higher AC. Dex has a few like defensive duelist, but all of those come at a cost. If there was no cost to some of those options you'd have an argument. If you still think Dex is better somehow then compare a Duelist Dex vs a TWF dex and you'll see a Duelist gets more damage and more defense than a TWFer (the topic of this thread).

Thanks!
Sorry, only quoted the parts I react on to avoid bloating.
1. Do you realize that you actually target 80% of the builds? Not EVERYONE wants to have to take the human JUST to be able to take Polearm Master feat.
And without feat, your argument does not hold, you only gain reach from wielding this kind of weapon (unless I forgot something).
Also, again, unless you're really lucky with rolling stats, you may very well want to increase your stats first before taking any feats. So that's Polearm or GWM coming online at lvl 6 for a Fighter at the earliest, 8 for other characters.
You actually prove that it takes a very specific level 1 choice to make any build better than Two-Weapon Fighting for most classes except Monk (which gets its own bonus action at lvl 1 also), and that it stays a very potent choice until end of first tier, maybe later for some builds. Thanks. :)

2. Ok... Do I REALLY have to list everything? Please pity me...
Mirror Image (non-concentration), Blur (concentration), Mage Armor, permanent Mage Armor (Draconic Sorcerer, Warlock), Evasion, Deflect Arrows, Parry Manoeuver, Shield Master's improved save, Barbarian and Monk's Unarmored Defense, and I probably forget a few...
Sure, if you compare a maxed build towards only permanent AC, STR- based will prove better in most cases (except maybe lvl 20 DEX-based Barb), I won't argue there.
But your point was that DEX-based characters quickly fall behind STR-based in terms of AC. And that is false. Some DEX builds actually start with better AC than some STR builds (not even talking about the "defense" you actually get by having good initiative and stealth capability). :)
And you can make pretty strong DEX-based AC if you really wish so, you just have to plan a bit more than with STR-based character who justs have to don the armor.
By the way, using Duelist is not a good point of comparison on the STR / DEX topic since you can also use finesse weapons with it. :)

Ardantis

2015-10-12, 12:20 PM

I have to come out in support of Kryx in terms of base DPR- every other damage-based build exceeds TWF past level 5 and the acquisition of feats and extra attack. Even archers outdo TWF because of the strength of the archery style and the Sharpshooter feat.

That said, two-weapon fighting still has a place for the following reasons:

- At low levels, it's the only way to get bonus actions attacks asides from Variant Human Polearm Master and Frenzy Barb.

- It enables extra chances at sneak attack and smite with NO FEAT INVESTMENT.

- It's the only way for a non-Monk dex-based character to gain a bonus action attack. Granted, it is competing with dueling style for dex-based characters, to which it loses in DPR, but it does spread the damage across one more attack, which is relevant mostly for Rangers. I would consider it a valid option for dex-based melee.

That said, with only those specific examples of excellence, it is a style with fairly limited application. Granted, this seems reasonable to me, because from a simulationist perspective, it is a very difficult way to fight which sees very little benefit but for the specially trained, and is very rarely seen outside of certain Western and Japanese dueling traditions.

Kryx

2015-10-12, 12:45 PM

I also find it odd to assume that Str adds to armour. My latest campaign are now all at level 7 and no one has full plate armour yet. The assumption that any character can get whatever equipment they want is probably reasonable for anything under 100gp but I wouln't be so sure beyond that.
See https://www.reddit.com/r/dndnext/comments/2r8kci/deconstructing_5e_typical_wealth_by_levelGiven normal treasure the party would have enough gold at level 6 at the latest given the standard treasure tables.

How many of these saves, each, are exclusive to Only Dex or Only Str?
Nearly every save is one or the other. There are a handful that will do Dex or one part and Str for another, but none or very few that allow multiple saves for the same effect.

Also, claiming that Dexterity has lower armor then Strength: No. Mage Armor exists, High Elf and variant Human exist, Anyone can take Magic Initiate., and unless some massive change in the unwritten rules happened since the DMG came out, you cant assume magical items, which means no +3 armor.
Mage armor = 13+5 = 18 max. For a martial to get that they'd need a feat or race investment as you've said. No magical items assumed at all.

Variant Human
100% agreed that variant human is not ideal. I don't allow it personally.

Early Levels
Level 1-3 is not 80% of builds. 1-3 is considered the warm up. 1-2 is supposed to take less than 3 sessions. 3 is quite short as well. No one should care much about level 1-3.
Level 4+ TWF gets crushed by Polearm or GWM.

Feats
Without a feat the math is different on all sides. Without feats S&B likely loses to TWF, but TWF still falls short of Greatsword as I outlined above.

Stats over feats
It is never mathematically advantageous to take a stat boost over Polearm for any class that can use Polearms.

Mirror Image (non-concentration), Blur (concentration), Parry Maneuver, Shield Master's improved save
Both Dex and Str can use these

Mage Armor, permanent Mage Armor (Draconic Sorcerer, Warlock)
This would tie heavy armor after you invest a feat or race and 2-4 ASIs depending on build.

Barbarian and Monk's Unarmored Defense
Cool features, but don't stack up until very late game with lots of investment in secondary stats.

Evasion
Save against dex saves = no damage instead of half.
First real dex benefit, but a Barbarian's resistance via totem would outweigh that if we're comparing Str vs Dex.

Deflect Arrows
Good benefit. Costs reaction, but really nice. However the Monk chasis takes a long time to ramp up into his defenses.

That said, with only those specific examples of excellence, it is a style with fairly limited application. Granted, this seems reasonable to me, because from a simulationist perspective, it is a very difficult way to fight which sees very little benefit but for the specially trained, and is very rarely seen outside of certain Western and Japanese dueling traditions.
The specific examples only reign true at levels 1-3. Every other build has better options:
Rogue - Hand Crossbows
Paladin - Polearm
Ranger - Heavy Crossbow or Hand Crossbows

TWF has existed in fantasy for a long time. Whether it's a dual wielding Barbarian or Drizzt. It shouldn't be uncompetitive because it's niche.

Vogonjeltz

2015-10-12, 12:52 PM

Citan

2015-10-12, 01:47 PM

Early Levels
Level 1-3 is not 80% of builds. 1-3 is considered the warm up. 1-2 is supposed to take less than 3 sessions. 3 is quite short as well. No one should care much about level 1-3.
Level 4+ TWF gets crushed by Polearm or GWM.
You didn't understand what I meant. I meant that 80% of the builds are based on anything else than Human Variant. You condition the race choice to a mechanical advantage during the early levels. It's your conception of character building. I find it very sad but to each its own... Also, again, it crushes only if you can hit and either kill or crit consistently. Meaning heavy luck on stat roll or a very good buffer in your party. +10 damage is useless if you miss.

Feats
Without a feat the math is different on all sides. Without feats S&B likely loses to TWF, but TWF still falls short of Greatsword as I outlined above.
Because of the 1&2 reroll? Well, it's pretty nice indeed. But the difference isn't that great in the first levels, because you get 2 attacks instead of 1 on the other side.

Stats over feats
It is never mathematically advantageous to take a stat boost over Polearm for any class that can use Polearms.
I'd like to see a detailed argument on this point, because it seems very surprising to me. Thanks. :)

Both Dex and Str can use these
Sure, but if you have low Dex, your Mirror Images will fall much quickier, greatly diminishing the interest.

This would tie heavy armor after you invest a feat or race and 2-4 ASIs depending on build.
I really don't see how. Mage Armor is a given for Bard, Sorcerer, Warlock, Eldricht Knight and Arcane Trickster, throughout their carrier, without any additional feat or race investment. As for ASI investment, well it's exactly the same with Strength, again... You are maxing your attack stat in both cases.

Cool features, but don't stack up until very late game with lots of investment in secondary stats.
Agreed on that if you want to get up to 20, but getting AC up to 18 isn't so hard. You can easily begin with 16 AC at level 1 without great rolls, maybe 17 with luck. And you are still doing what you would do anyways, maxing your primary stats (Widom for Monk is as important as Dexterity. Constitution for Barbarian is debatable indeed).

Save against dex saves = no damage instead of half.
First real dex benefit, but a Barbarian's resistance via totem would outweigh that if we're comparing Str vs Dex.
Requiring a specific class to get as well, and limits a bit your action (rage limitations), although for a martial it's not a big hassle. Unless you take Shield Master of course, which benefits any build.

Good benefit. Costs reaction, but really nice. However the Monk chasis takes a long time to ramp up into his defenses.
Confer the previous point. :) Also I'm surprised you see this feature so high, I thought it was one of the less interesting of Monk. :)

The specific examples only reign true at levels 1-3. Every other build has better options:
Rogue - Hand Crossbows.
Except that you need a feat to avoid disadvantage in melee AFAIK, and another one to really bring in damage.
Paladin - Polearm.
For reach I suppose? Because beyond that you again enter the field of feats.
Ranger - Heavy Crossbow or Hand Crossbows
Except that you suffer the exact same limitations as with Rogue example.

TWF has existed in fantasy for a long time. Whether it's a dual wielding Barbarian or Drizzt. It shouldn't be uncompetitive because it's niche.
>>> Answered directly in quote (in italic) except for the last part because I'm really a lazy guy, and without remorse even. :)

It's not "uncompetitive". It's less interesting damage-wise at higher levels and is not as easy to build with as would be a Sword & Board. Doesn't mean it's lesser in any and all situations. (Although I agree with most that it could have benefitted from more love).

By the way, I'm personally a Sword & Board lover and take it whenever in doubt. In case you were wondering... ^^

bid

2015-10-12, 02:06 PM

I think initiative is underrated here. The chance to get an extra round of combat in from having higher dex is a huge bonus. I find most combats to be about 3 rounds - getting an extra turn is then a 25% increase in damage.
Assuming equal initiative, you would go first 50% of the time, with +3 66% and +5 75%. That increases your early DPR by 32% to 50% which might be enough to kill another monster before it acts. In a way, every DPR should be increased by that early damage spread over the expected number of rounds.

This doesn't change a thing on how TWF loses to S&B, though.

I also find it odd to assume that Str adds to armour. My latest campaign are now all at level 7 and no one has full plate armour yet. The assumption that any character can get whatever equipment they want is probably reasonable for anything under 100gp but I wouln't be so sure beyond that.
Yeah, scale mail / chain mail have the same AC but as soon as you get to breastplate / splint heavy gets +1 AC. I don't think getting 200g is that hard in the middle levels but a +7 hit monster would get you 10/20 vs 11/20 -> a 10% increase in damage, so Dex should do 10% more damage to match Str.

Ardantis

2015-10-12, 02:11 PM

See I would argue that fighters are best at it because they get a fighting style for it, and they have no real bonus action conflict.

Rogues, barbarians and paladins don't have that style, and rogues have other bonus actions they probably want to use.

Rogues and Paladins don't need the Fighting Style, as they only use the offhand attack to enable bonus damage from Sneak Attack and Smite, respectively. Barbarians add rage bonus damage to all Strength attacks including offhand attacks, making TWF valuable for mook killing at low levels.

Rogues do have the most other uses for the bonus action, but they don't always conflict with offhand attack, and Swashbucklers in particular gain melee Sneak Attack against isolated opponents and are likely to gain significant benefit in achieving sneak attack from offhand attack.

Additionally, since the offhand attack is merely sneak attack insurance, hitting with your attack action then frees up your bonus for Cunning Action. It makes melee Rogues more reliable at Sneak Attack without taking away their mobility advantage. Basically, if you miss a Sneak Attack you can choose to try again or Cunning Action. More choices are good.

Note that this does not require MC or a feat to work.

Kryx

2015-10-12, 02:13 PM

That increases your early DPR by 32% to 50% which might be enough to kill another monster before it acts.
Can you explain these numbers? How does going first increase DPR? By eliminating enemies?

Also what about the factor of distance on a melee character? Just because you go first doesn't mean you can engage.

bid

2015-10-12, 02:36 PM

I really don't see how. Mage Armor is a given for Bard, Sorcerer, Warlock, Eldricht Knight and Arcane Trickster, throughout their carrier, without any additional feat or race investment. As for ASI investment, well it's exactly the same with Strength, again... You are maxing your attack stat in both cases.
Cha is the attack stat in half the cases, AT is Dex-only. That leaves EK as the only one who could go Str.

Agreed on that if you want to get up to 20, but getting AC up to 18 isn't so hard. You can easily begin with 16 AC at level 1 without great rolls, maybe 17 with luck. And you are still doing what you would do anyways, maxing your primary stats (Widom for Monk is as important as Dexterity. Constitution for Barbarian is debatable indeed).
The ultimate barbarian can start 17-14-17 and end 20-16-20 for AC18 unarmored at level 19. Any heavy can get AC18 from level 1 onward.
Most monk start 16-14 and end 20-20 for AC20 unarmored. Heavy with shield starts at AC20.

Sure you can delay access to splint/full until the mid levels, but unarmored will always be weaker.

BTW, I consider level 1-4 to be a fluke. What matters level 5 onward, when you've invested too much in your anti-optimised character to back down. Pick TWF because you want to RP that concept and because you don't care so much about DPR that an extra 10% won't matter.

Dex S&B all the way, baby!:smallbiggrin:

bid

2015-10-12, 02:57 PM

Can you explain these numbers? How does going first increase DPR? By eliminating enemies?
You need a fixed definition of "start of turn" that accounts for initiative. The easiest one is when the first monster attacks. Therefore all damage done prior to that is in a pre-round that needs to be added to overall damage. To improve, you should pick "the last monster to die" as the start of turn. Then you don't care if it had 1 hp left or almost survived.

For instance, in a 4-round combat where the monsters have no init bonus:
- a mace user would do 50% * (2d6+3) * 12/20 = 3 pre-round damage or 6.75 corrected DPR
- a rapier user would do 66% * (1d8+5) * 12/20 = 3.8 pre-round damage or 6.64 corrected DPR

Since they both need a turn to reach their targets, just assume a 5th round they spend dashing forward. As for ranged:
- a longbow user would do 166% * (1d8+3) * 14/20 = 8.7 pre-round damage or 7.43 corrected DPR

Kryx

2015-10-12, 04:36 PM

That extra "pre-fight" DPR only applies on the first round. I don't know what 12/20 is.

Very weird way of looking at things. If you consider it a "pre-round" then you have to consider that you're acting second in the "normal round" and therefore losing out some there.
It feels like a bit much. In practice Initiative doesn't change DPR, but would change damage taken if you're able to go first and kill a creature before he moves.

bid

2015-10-12, 06:09 PM

That extra "pre-fight" DPR only applies on the first round. I don't know what 12/20 is.
In the same way that assassin does, or your last slot of smite does. Yet you spread this across all rounds for an average DPR.

As for that 12/20, that's the same as hit on 9+.

Very weird way of looking at things. If you consider it a "pre-round" then you have to consider that you're acting second in the "normal round" and therefore losing out some there.
It feels like a bit much. In practice Initiative doesn't change DPR, but would change damage taken if you're able to go first and kill a creature before he moves.
Initiative order is meaningless once you've gone around. From the point of view of the char who rolled init 5, a round is from his turn to the next, anything else is pre-round. It's just a different frame of reference.

Seen backward, the last mob does its last action and then dies on a player's turn, making it the end of the round. You could frame your rounds this way and have that player's first action as the end of pre-round. It's just a different POV which is fairer, since everyone had their turn every round and "pre-round" contains the "unfair" part.

And yes, initiative changes DPR. Once the last mob is dead everyone else gets 0 DPR for their last turn. Or seen another way: the first player will attack a 5th time and do 25% more overall damage than the last player who only did 4 attacks.

endur

2015-10-12, 09:09 PM

As WOTC releases more books, we can expect to see more feats, more fighting styles, more sub-classes, etc. Just because TWF is behind GWF & Archery right now, doesn't mean it will always be that way.

Yakk

2015-10-12, 09:35 PM

What is +4 initiative worth?

If you have 5 fights per day and an average of 3.5 rounds/fight, and the foes initiative is never rolled so low that you auto-beat them, nor so high that you auto-lose... (this needs to be modeled better)

You get 4/20 = 0.2 extra rounds/fight compared to a foe. At 5 fights/day, that is 1 extra round. In addition, this extra round is "front loaded".

So you get 18 rounds, compared to 17 rounds for someone with lower dex (3.5*5=17.5, gave you +.5 and them -.5).

That is a 5.9% increase in your per-day damage output. I was generous in the "your foes don't roll really bad, nor really good" model, but less generous in "front loaded damage in a fight is worth more than back loaded damage" (as dead foes don't cause damage, typically).

JoeJ

2015-10-12, 09:48 PM

Splitting damage among multiple targets? Hardly worth any more than 5% in overall damage imo.

How did you calculate that? Wouldn't you have to make some sort of assumption about the number of enemies per fight?

Malifice

2015-10-12, 10:07 PM

PC's with access to Hunters Mark and Hex can do some decent damage with it at low and mid levels.

Rangers, Vengance Paladins, Warlocks, magic initiate feat.

A DW/ TWF Dex based ranger with Colossus slayer can deal some very solid damage starting at third level.

Twin rapiers (sabers?) dealing (1d8+3+1d6) and (2d8+3+1d6) is nothing to sneeze at.

bid

2015-10-12, 10:53 PM

You get 4/20 = 0.2 extra rounds/fight compared to a foe.
That's not true.
That 4/20 is when you roll 17+ and are automatically first, it does not handle the 4/20 cases when they roll 4- and are automatically last.
Counting that way, it should be {4/20 + 4/20 - 16/400 = 0.36}

If you make a 20x20 grid for both rolls and don't apply any init bonus, you'll be first 190/400, last 190/400 and equal 20/400.
If you have a +4, you'll be last 190 - (19 + 18 + 17 + 16) = 120/400, equal 15/400 and first 265/400.
That's (265 - 120)/400 = 0.36 extra round/fight.

Your Str fighter will get 3.5*5 = 17.5 rounds
Your Dex fighter will get (3.5+0.36)*5 = 19.3 rounds
That's a 10% increase in your per-day damage output.

Coidzor

2015-10-13, 01:04 AM

PC's with access to Hunters Mark and Hex can do some decent damage with it at low and mid levels.

Isn't the issue there with having dead rounds where one can't TWF due to having to use one's Bonus Action to change targets?

Malifice

2015-10-13, 01:15 AM

Isn't the issue there with having dead rounds where one can't TWF due to having to use one's Bonus Action to change targets?

Yeah, but by the time you hit third it's not that big a deal; most CR 2/3 monsters have 50+ hp anyway.

A CR 3 manticore has 68 HP. a knight has 52. A killer whale has 90. A hobgoblin captain has 39 and AC 17.

A single CR 3 creature is an 'easy' encounter for 5 x 3rd level PC's. 1 X CR 3 and 1 x CR 2 is 'hard'.

Bonus action hunters mark then greatsword attack on round 1 isnt that different from bonus action hunters mark then longsword attack on round 1. From round 2 however, things get dire for the monster.

Also; a lot of creature like to monologue or do fluff stuff pre-battle. Hex and Hunters mark the crap out of them as they do so.

Kryx

2015-10-13, 03:16 AM

As WOTC releases more books, we can expect to see more feats, more fighting styles, more sub-classes, etc. Just because TWF is behind GWF & Archery right now, doesn't mean it will always be that way.
TWF shouldn't have to take more feats to be competitive. The cost is already the same:
GWM Fighter: 1 feat (polearm)
TWF Fighter: 1 feat (dual wielder)

What is +4 initiative worth?
This seems more legit. Though the value would change based on the creature. You'd have to just multiple .36*normal DPR to get initiative value. (used Bid's updated number)

Your Str fighter will get 3.5*5 = 17.5 rounds
Your Dex fighter will get (3.5+0.36)*5 = 19.3 rounds
That's a 10% increase in your per-day damage output.
3.5?

How did you calculate that? Wouldn't you have to make some sort of assumption about the number of enemies per fight?
I didn't calculate it. That's my opinion. Even then Reach is more valuable I think.

PC's with access to Hunters Mark and Hex can do some decent damage with it at low and mid levels.
This would be true if TWF and Hunter's Mark didn't share the same resource (bonus action).

If we use basic DPR:

Normal damage:
TWF: d6+3 + d6+3 = 7+6 = 13
HM: d6+d6+3 + d6+d6+3 = 14+6 = 20
Main hand: d6+d6+3 = 7+3 = 10

Assuming a 5 round encounter and switching enemies twice:
Round 1: HM: 10 DPR
Round 2: HM: 20 DPR
Round 3: HM: 10 DPR
Round 4: HM: 20 DPR
Round 5: HM: 10 DPR

TWF gets 13*5 = 65
HM gets 70

That's a typical encounter imo. Now if you're in a boss fight then it would be 65 vs 90.

It's not great. It's only great 1-3. After that it scales to fall behind by a large amount.

Malifice

2015-10-13, 03:42 AM

If we use basic DPR:

Normal damage:
TWF: d6+3 + d6+3 = 7+6 = 13
HM: d6+d6+3 + d6+d6+3 = 14+6 = 20
Main hand: d6+d6+3 = 7+3 = 10

Assuming a 5 round encounter and switching enemies twice:
Round 1: HM: 10 DPR
Round 2: HM: 20 DPR
Round 3: HM: 10 DPR
Round 4: HM: 20 DPR
Round 5: HM: 10 DPR

TWF gets 13*5 = 65
HM gets 70

That's a typical encounter imo. Now if you're in a boss fight then it would be 65 vs 90.

It's not great. It's only great 1-3. After that it scales to fall behind by a large amount.

That doesnt accurately account for single target DPR IMO.

Add the dual wielder feat and account for the reduction in returned DPR via +1 AC. I find (in practice its more than competitive with GWM fighters till around 11th and 3rd attack comes online. Action surge makes GWM better for Fighters (TWF BM fighters going nova being an exception). TWF is perfectly valid for Rangers and Paladins and Rogues for melee though. Rangers are clearly better ranged however.

Kryx

2015-10-13, 04:47 AM

That doesnt accurately account for single target DPR IMO.
I wrote it below. 65 vs 90. Once you add dual wielder (4) then polearm dominates it and GWM beats it. Hand Crossbow wins for everything else.

I find (in practice its more than competitive with GWM fighters till around 11th and 3rd attack comes online. Action surge makes GWM better for Fighters.
11+ is when TWF falls behind miserably according to the math. Action surge actually helps GWM more as their attacks are bigger.

TWF is perfectly valid for Rangers and Paladins and Rogues for melee though. Rangers are clearly better ranged however.
By RAW no it's not. -5/+10 makes archery a significantly better choice.
For Paladins GWM or Polearm is a significantly better choice.

There is no niche level 4+

toapat

2015-10-13, 06:08 AM

Nearly every save is one or the other. There are a handful that will do Dex or one part and Str for another, but none or very few that allow multiple saves for the same effect.

Count. Actual Numbers. Because every instance ive read allows Dex/Str or Athletics/Acrobatics to resist of Str Saves barring One.

Strill

2015-10-13, 06:10 AM

Count. Actual Numbers. Because every instance ive read allows Dex/Str or Athletics/Acrobatics to resist of Str Saves barring One.

You're thinking of ability checks, not saving throws. If it lets you use Athletics/Acrobatics, it's not a saving throw.

Kryx

2015-10-13, 06:11 AM

Count. Actual Numbers.
I gave them above. Here is my saving throw document (https://docs.google.com/spreadsheets/d/17ZeFuwQVvb9DsMseUU8Pb0KxDU7sizhmebp-U7FuzLY/edit) that counts them all (with some houserules in progress). Still porting Con and Wis over from my other doc:

I know of 1 that allows either Strength or Dex and that is Paladin OotA Nature's Wrath.

Now please share the others that allow either so I can add them to my doc or say "You're wrong!" :P

bid

2015-10-13, 12:05 PM

3.5?
dixit Yakk: "If you have 5 fights per day and an average of 3.5 rounds/fight"

Kryx

2015-10-13, 12:38 PM

average of 3.5 rounds/fight
That's quite short. As you likely saw I got a lot of flak for estimating 5 per encounter.

I'd be curious what method you used to get to 3.5.
Are you assuming Medium and Hard as DMG recommends and then assuming medium is 3, hard is 4?

bid

2015-10-13, 04:20 PM

I'd be curious what method you used to get to 3.5.
Don't ask me, ask Yakk.:smallwink:

Vogonjeltz

2015-10-13, 08:23 PM

How did you calculate that? Wouldn't you have to make some sort of assumption about the number of enemies per fight?

We could always calculate the number of possible enemies for each type of fight difficulty by each level and then determine the average number of enemies.

ie at level X a medium difficulty fight is a combined adjusted XP value of say, 3200; which allows a maximum of 80 opponents worth 10 XP (because of the factors involved in determining overall difficulty) or a single opponent worth 3200 XP or some mixture in between there.

Of course, this is just the average possibility we'd be working towards, in practice it might be that fights usually have a similar number of enemies as PCs, or between 4-8 enemies. Sometimes it's just going to be an individual DM style (your mileage may vary)

Mara

2015-10-13, 08:26 PM

Don't ask me, ask Yakk.:smallwink:
I really hope Yakk's justification is, "well those are what the guidelines in the DMG say. It's not my fault if you play differently then what the whole game was balanced around".

BigONotation

2015-10-14, 12:18 AM

Swap TWF Fighting Style and Dual Wielder. So Fighting style gives bigger weapons (+1 dmg) and dual wielder gives stat on offhand
TWF gets Rend at level 11 (Double proficiency extra damage if both the main and offhand hit)
TWF Fighting style gets +1 hit

The math works out much better.

Kryx, thanks so much for doing so much work educating us to DPR, really appreciated it! To address the part of your post I quoted, do you have that calculation on your homebrew? I would so love to see and am hoping it exists already or you could easily add it. Thanks again for all the hard work, I can say the 20 or so odd people I play with find it invaluable.

Kryx

2015-10-14, 01:47 AM

Kryx, thanks so much for doing so much work educating us to DPR, really appreciated it! To address the part of your post I quoted, do you have that calculation on your homebrew? I would so love to see and am hoping it exists already or you could easily add it. Thanks again for all the hard work, I can say the 20 or so odd people I play with find it invaluable.
I'm glad you find it useful as well!

Looking at the actual math has made me reconsider a lot of assumptions that I had and has made me change several balancing houserules to be better.

What calculation are you referring to? I compare twf vs polearm and GWM on RAW and on the houserules tab with the graph. RAW TWF is horrendously behind.

djreynolds

2015-10-14, 05:05 AM

Is TWF weaker, of course it is.

Its in the middle of someone with pole arm for damage because he can use GWM and get a bonus action and guy with a shield who has more AC and can shield bash.

But is has its perks.

First with two light weapons its free. For finesse guys caught in melee, not bad.

Many players have bonus actions and can't afford a feat yet. So till 5 its awesome.

For rogues it is a lifesaver to get a second chance to land a sneak attack. Its a cheap investment.

Second, I find magic daggers all the time. Some people have resistance to normal weapons. I may want a magical hammer, but if all I have is magic short sword and magic dagger, I may have to use it.

Donning a shield is an action, with duel wield feat drawing two weapons is free. Disarm my bow, I'll draw two finesse weapons and fight.

TWF is situational, but I like the game like that. I don't like perfect, its boring

But I like two-weapon rend, its an easy fix. Double proficiency for damage. Or free disengage or shove. Your choice if both weapons hit.