Bringing PAM, SS, and GWM in line?

[Zman]

It came across (to me at least) from both you and Kryx that the mathematical work was "proof" of both problems and fixes. My counter was that your assumptions that underlie the math are driving the result. That is, you're misusing math--using it to say too much. It's a pain point for me, in part because I've seen it used in the MMO context to drive changes that hurt the game as a whole and caused conflict between players.

Not as strong as proof, but a mathematical analysis can provide strong evidence towards the existence of problems and efficacy of fixes.

Two things--the assumptions about the average AC are strongly dependent on the model applied. Do you use single-target deadly fights? Do you have a more natural mix of big and small fights? That's gonna completely change the value of things that affect accuracy (such as the -5/+10 features).

The main topic of discussion is martial combat balance and that is withing the scope of martials.
It is much harder to balance martials against say casters, but when discussing these feats and martials against themselves it becomes much easier to use. I understand what you are saying about AC, but it is far less impactful than you are making it out to be most of the time, the big exception is looking at the -5/+10 of GWM.

Most of the time AC matters far far less than you are making it out to. Why? Because we are comparing martials to other martials and their to hit values are within a very small range to the point we are often using their "expected"
value. Ie a 1st level fighter has a primary attribute of 16 and a proficiency of a +2. So, when I am comparing a Greatsword with GWF to a Greataxe with GWF to hit is largely irrelevant because in both cases it will be the same value. If the AC of the target is 13 or 17 it doesn't matter, each identical fighter goes from having a 65% chance to hit and 5% chance of a crit to having a 45% chance of hitting with the same 5% chance of a hit. Sure, with critical hits becoming a larger portion of hit chances the attack method that benefits criticals more gets a very minor edge.
But, overall this really doesn't matter much when we are trying to determine which weapon will deal more damage over time. Now, if you are attempting to simulate combat and are doing these calculations against enemies then things get much much more sticky. Did that hit kill an enemy or just hurt it etc. Everything you are saying becomes much much more relevant. I mean, if we use enemies with only a single hit point we can make damage irrelevent with an ability to hit and number of attacks being most critical, if we pick a single enemy with a massive amount of hit points, then average damage becomes much much more important. The assumptions required for that scale of a simulation analysis is pretty staggering, it is one of the reasons math is a useful tool and cannot tell of us everything. But, I don't think we need a simulation of that magnitude to tell us the Greatsword is superior to the Greataxe or that GWF benefits the Greatsword to a larger degree than a Greataxe. My point estimates will do that just fine and allow us to draw strong conclusions about the balance or more importantly lack thereof of GWF.

As I said GWM with its -5/+10 is a definite exception and more complicated. But, using an aggregate of different point estimates from the range of expected opposing ACs ie High, Low, and Median gives us quite a bit of usable data to draw a conclusions. Vs Median AC a big swing is marginally better over time than not using it, but it is swingy and people as you pointed out are risk averse. We know it is vastly superior vs Low ACs and the lower AC the better the tradeoff becomes. We know that Advantage and Disadvantage greatly alters this calculus. I mean just looking at what a big swing does vs stock at the three AC ranges and with advantage or disadvange we can draw some strong conclusions. We know it is a terrible choice against high ACs.

What you're producing are called point estimates. A fundamental theme in quantitative work is that point estimates are not worth much. The distribution of results matters much much much more. This is especially true with people, as most people are risk averse. People in general prefer to have sure results, even if that reduces their average result. Higher variance with a higher mean is usually considered worse than a lower mean with a tighter spread. "A bird in the hand is worth two in the bush" goes the old saying.

Of course the distribution of results matters more, it is why we look at the standard deviation of an expected result, its range, and its median result. But, given the nature of the dice involved using a point estimate with the mean can be useful.

Yes, people are risk adverse, but predicting people's behavior gets very tricky. For instance in my Greatsword vs Greataxe example risk aversion of rolling minimum damage may further push a player away from a Greataxe.
Since most of our analysis is using identical to hit chances this point is a bit less prominent as you make it out to be.
Sure, GWM and -5/+10 are an exception. It does matter when comparing classes or styles that can get additional attacks etc. which likley pushes in favor of GWM and PAM being even more desireable.

Second, yes, yes you do need those numbers of data points or your statistics are wrong and you're breaking the assumptions made for them. That is, your thought experiments/mental math/spreadsheets don't reflect reality and may (or may not) be seen at the table.

Strictly speaking, of course they don't. Neither would your super simulation either, but it'd be closer. The real question is are they close enough to draw firm and usable conclusions? Is my method usable enough to draw conclusions about GWF having poor internal balance? I bet my extremely quick calculations will come up to the same conclusion your extremely complex simulation will. Do we really need ten thousand runs to tell us that GWM's -5/+10 is imbalanced when there is easy access to advantage ie Reckless Attack? Do we need ten thousand runs to tell us that when I limit GWM to only a single big swing a turn that is greatly curbs that imbalance? Nope,
in each of those cases my method is close enough to draw usable conclusions. It doesn't have to perfectly capture reality to be useful, but it's damn sure close enough to be useful and inform our decisions.

I'm working on a script to show this, but preliminary numbers indicate that for a flat distribution, replacing the actual flat (or nearly so) distribution with the averages (or medians) requires thousands of rolls to approach 5% precision. This is basic math--flat distributions converge to normal distributions very very slowly. You cannot replace a flat 1dX roll with the average value unless you have more than (approximately) 1k data points or unless you're willing to put up with a 10% or greater swing between real and theoretical. Over the course of a single combat (~3-5 rounds), the swing is more like 50% in either direction from the numbers I'm seeing. And this is just from the dice roll. Any change of the distribution (either in target AC or modifiers) invalidates all previous data from a statistical perspective. If you never leveled up and sit hitting a single target for thousands of rounds, you might get numbers comparable to what you're quoting as numerical truth. Otherwise, the "evidence" is well inside the natural variance of perfect dice, not to mention the irregular dice people actually play with or the widely varying tactics employed (which change the value of options by large amounts).

How many active tables does it take to cover those thousands of those flatly distributed d20 rolls?
Human perception matters, it often won't take thousands of rolls for a human being to notice, or "feel" an imbalance.
I wonder how long it'll take a simulation comparing a character attacking(with identicaly to hit modifiers) with a d8 weapon and a d10 weapon to determine statistically significantly superior damage from the d10. I bet it won't take people at a table that long to figure it out.

All in all, numerical analysis is a useful thing in the abstract as long as we recognize its limitations. For myself, if two options are within about 20% theoretical DPR, I need strong in-play evidence to worry about imbalance. Numerical analysis also (by its nature) focuses on one of the least important statistics--raw damage done. The other factors even in combat make a bigger difference to how the class/option feels.

20% theoretical DPR.... I bet comparing that D10 weapon and the D8 weapon I mentioned above will fall withing 20% theoretical DPR. Or can we use a point estimate and forgo the thousands of simulated runs and the strong in play evidence in that case? /sarcasm

[PhoenixPhyre]

Replying inline to a quote makes responding much harder. Please don't do so--split those quotes up. As a result, I'm not going to respond to the bulk of your responses. They were mostly non-responsive, so I don't feel too bad.

20% theoretical DPR.... I bet comparing that D10 weapon and the D8 weapon I mentioned above will fall withing 20% theoretical DPR. Or can we use a point estimate and forgo the thousands of simulated runs and the strong in play evidence in that case? /sarcasm

You continue to make the mistake of confusing the map for the territory--the model for the thing being modeled. White-room, training-dummy calculations (which is what happens when you don't actually simulate things, by necessity) give false impressions about balance to players and hurt the game. There. I flat out said it.

The numbers presented require all else to be the same--all is rarely if ever the same. Those other factors are larger than the observed effect. Thus, the calculations have reduced meaning. Yes, a d10 vs d8 weapon is, for all practical purposes except theoretical DPR, a nullity. You're prematurely optimizing and in doing so (and presenting the results as fact, with heavy attacks and sarcasm toward those who disagree) giving the impression that if you don't optimize you're failing as a player. This has the same type of distortionary effect on player behavior that we've been decrying all along--artificially enhancing the numbers of people who pick options (feats, etc) for mechanical reasons, not for character reasons.

A large fraction of players pick options because they're any of:

• something they've never tried
• something that looks cool
• something that fits a character concept (regardless of optimization)
• something that for whatever ideosyncratic reason works for them.

Telling these players that they're playing wrong (which comes up every time someone asks for advice on a martial that either uses TWF or anything other than GWM/SS/PAM/CBE) is a disservice to them, to the community, and to the game. Making it all about "math" and "huge differences in DPR" (which rarely if ever will actually matter in play) makes matters worse. The truth is that for most games (where no one is intentionally seeking to exploit the system) the presence or absence of these feats will not make a lick of difference in the fun had by the players. I have never actually seen the feats in question in play myself. Many more people have picked the "fluff" feats (Actor, Keen-minded, Linguist, etc). Very few people in my experience know or care about "optimization." Making mechanical optimization the primary deciding factor makes people more cabined to the "correct" options. That's no fun in my opinion.

I agree very much that those specific examples are problems, but I disagree that the math is necessary or sufficient to show that. I also disagree purely as someone who cares about math itself with the way you and Kryx calculate and present the numbers. You're doing violence to statistics and math. This is entirely separate from the feats themselves. For example, you can't combine numbers from different distributions (different tables, different fights). Those numbers belong in completely different problem spaces. Combining them is bad math and makes those who know and care about such things cringe.

In conclusion--it's the means I disagree with, not the ends. These feats need to change. In this instance, the math and reality are aligned. By accident, mostly. Portraying point estimates as the best way (or even a necessary way) to understand this does both the math and the game a disservice.

[Zman]

Spoiler: Txt, lots of txt.

Show

Replying inline to a quote makes responding much harder. Please don't do so--split those quotes up. As a result, I'm not going to respond to the bulk of your responses. They were mostly non-responsive, so I don't feel too bad.



You continue to make the mistake of confusing the map for the territory--the model for the thing being modeled. White-room, training-dummy calculations (which is what happens when you don't actually simulate things, by necessity) give false impressions about balance to players and hurt the game. There. I flat out said it.

The numbers presented require all else to be the same--all is rarely if ever the same. Those other factors are larger than the observed effect. Thus, the calculations have reduced meaning. Yes, a d10 vs d8 weapon is, for all practical purposes except theoretical DPR, a nullity. You're prematurely optimizing and in doing so (and presenting the results as fact, with heavy attacks and sarcasm toward those who disagree) giving the impression that if you don't optimize you're failing as a player. This has the same type of distortionary effect on player behavior that we've been decrying all along--artificially enhancing the numbers of people who pick options (feats, etc) for mechanical reasons, not for character reasons.

A large fraction of players pick options because they're any of:

• something they've never tried
• something that looks cool
• something that fits a character concept (regardless of optimization)
• something that for whatever ideosyncratic reason works for them.

Telling these players that they're playing wrong (which comes up every time someone asks for advice on a martial that either uses TWF or anything other than GWM/SS/PAM/CBE) is a disservice to them, to the community, and to the game. Making it all about "math" and "huge differences in DPR" (which rarely if ever will actually matter in play) makes matters worse. The truth is that for most games (where no one is intentionally seeking to exploit the system) the presence or absence of these feats will not make a lick of difference in the fun had by the players. I have never actually seen the feats in question in play myself. Many more people have picked the "fluff" feats (Actor, Keen-minded, Linguist, etc). Very few people in my experience know or care about "optimization." Making mechanical optimization the primary deciding factor makes people more cabined to the "correct" options. That's no fun in my opinion.

I agree very much that those specific examples are problems, but I disagree that the math is necessary or sufficient to show that. I also disagree purely as someone who cares about math itself with the way you and Kryx calculate and present the numbers. You're doing violence to statistics and math. This is entirely separate from the feats themselves. For example, you can't combine numbers from different distributions (different tables, different fights). Those numbers belong in completely different problem spaces. Combining them is bad math and makes those who know and care about such things cringe.

In conclusion--it's the means I disagree with, not the ends. These feats need to change. In this instance, the math and reality are aligned. By accident, mostly. Portraying point estimates as the best way (or even a necessary way) to understand this does both the math and the game a disservice.

It feels like you are being condescending and arguing just to hear yourself speak. I acknowledge you know more about math and statistics than me. I fully and unabashedly acknowledge I'm using a quick and dirty method of calculation that has major limitations and am aware of many of them.

What you have done is make accusations of me that are quite unfair. I have not said people who do not optimize are playing the game wrong. I have not said people can't and shouldn't pick things for reasons of cool, concept, or character. I have never told a person that they are playing the game wrong for choosing an "unoptimized" option. I would like an apology for that as I often am arguing and playing exactly the opposite.

I have concerns for balance as I want to see as many viable options as possible, and by that I mean both mechanically balanced and conceptually. When a handful of options are vastly superior to others it is problematic and leads those who tend towards optimizing to pick from a very few options, the same scenario often leads the player who is choosing concept over optimization to contribute less in a mechanical fashion to others at the table. Those are problems that I like to see addressed. I want to see a plethora of viable options and the bastardized type of math I'm doing is usable enough to serve that purpose. You may not like I used a screwdriver to pound in a drywall nail, it might offend your refined carpenter honed sensibilities, but if it gets the job done satisfactorily in 10s and we don't have to wait an hour for you to drive to workshop to select the perfect hammer for the job, then maybe just maybe it was the right tool for that specific job under those specific circumstances.

How are you determining what is optimized and unoptimized? Or are you saying we can't know whats optimized and are pushing a perceived or self fulfilling choice? Are you using your gut? Vast play experience which you've admitted doesn't actually apply to the questioned feats? Have you performed some grand simulated analysis of the game I'm missing? Or are you relying at least in some part with the type of napkin math me and Kryx often perform because you know its coming up with the right answer? How can you be so sure I've come up to the right answer using the wrong method when you've admitted you don't have the relevant real life experience nor have you put in the work for a full analysis? How are you coming up with your opinions or conclusions?

If you haven't simulated thousands of runs how are you coming up with any conclusions at all? Or are you making some pretty sweeping claims based on nothing but your hypotheses? How are any of us to discuss balance at all without doing some kind of analysis? How could Wizards possibly create a remotely balanced game without some kind of analysis. I'm betting my kind of quick and dirty math was used for much of the framework of the game by the devs, and I'm betting money they did not rely on complex simulations and your kind of proper math.

[PhoenixPhyre]

Spoiler: Specific responses

Show

I have tried to make sure to note where we agree and that my points are not personally directed at you or Kryx. I see that I have let my irritation get the best of me and have failed in that attempt. I believe your work is valuable. I also believe that it is frequently taken out of context and applied where it can't apply and used to make claims it cannot support. That is the problem I see.

What you have done is make accusations of me that are quite unfair. I have not said people who do not optimize are playing the game wrong. I have not said people can't and shouldn't pick things for reasons of cool, concept, or character. I have never told a person that they are playing the game wrong for choosing an "unoptimized" option. I would like an apology for that as I often am arguing and playing exactly the opposite.

My comments were directed at those who use yours (and Kryx's math) to argue such things. If you feel that it doesn't apply to you, I'm sorry that I unfairly tarred you with that brush.

I have concerns for balance as I want to see as many viable options as possible, and by that I mean both mechanically balanced and conceptually. When a handful of options are vastly superior to others it is problematic and leads those who tend towards optimizing to pick from a very few options, the same scenario often leads the player who is choosing concept over optimization to contribute less in a mechanical fashion to others at the table. Those are problems that I like to see addressed. I want to see a plethora of viable options and the bastardized type of math I'm doing is usable enough to serve that purpose. You may not like I used a screwdriver to pound in a drywall nail, it might offend your refined carpenter honed sensibilities, but if it gets the job done satisfactorily in 10s and we don't have to wait an hour for you to drive to workshop to select the perfect hammer for the job, then maybe just maybe it was the right tool for that specific job under those specific circumstances.

Except where that rough and dirty math is used as a weapon against others, making claims it can't in truth support. My original attempt was to clarify that the numerical work has rather strong limitations when applied beyond a limited region.

How are you determining what is optimized and unoptimized? Or are you saying we can't know whats optimized and are pushing a perceived or self fulfilling choice? Are you using your gut? Vast play experience which you've admitted doesn't actually apply to the questioned feats? Have you performed some grand simulated analysis of the game I'm missing? Or are you relying at least in some part with the type of napkin math me and Kryx often perform because you know its coming up with the right answer? How can you be so sure I've come up to the right answer using the wrong method when you've admitted you don't have the relevant real life experience nor have you put in the work for a full analysis? How are you coming up with your opinions or conclusions?

I'm not making any determinations other than "I don't like it the way it is." I think that optimization (especially for DPR) is significantly overvalued and needlessly restrictive. I care much more about perceived imbalance--how it feels in play. If that's "using my gut," so be it.

If you haven't simulated thousands of runs how are you coming up with any conclusions at all? Or are you making some pretty sweeping claims based on nothing but your hypotheses? How are any of us to discuss balance at all without doing some kind of analysis? How could Wizards possibly create a remotely balanced game without some kind of analysis. I'm betting my kind of quick and dirty math was used for much of the framework of the game by the devs, and I'm betting money they did not rely on complex simulations and your kind of proper math.

By not relying on math at all. I'm relying on the reports of people across many threads who hone in on and push these feats as "the one and only true way" of being a martial. That in and of itself is enough for me to decide I don't like how the feats feel. I fully admit that this is pure personal taste and others may differ. I have hunches about why it feels wrong, but since I've never seen the problem myself I'm not willing to take it beyond that.

The devs may have not used "real" math for balancing, but they also didn't rely on math as the primary sense of balance. I don't believe that your math is useless, just not formally correct nor valid beyond a rough cut. My personal cut-off for caring is between 10 and 20% differences in white-room situations--others may have different view points. I just don't personally care all that much about theoretical balance beyond that point, as my experience tells me that the other factors are way more important than the theory suggests or allows for.

Zman, I'm sorry that I got contentious. This will be my last post on the matter, as I see I cannot discuss this matter without becoming angry, and that anger is clouding my perceptions and tempting me to be much harsher than I rightfully should be.

I believe that we're actually much more in agreement than in disagreement (especially about the direct subject of the OP). I want to stress that I respect the output that those like you produce and find it useful (within its limits).

[Rebonack]

Absolutely. And such design requires much more than just numerical analysis. It requires intent on the designers--numerical balance in DPR is a good thing, but it's not the only thing nor is it an essential thing.

My preferred option would be to spread the wealth. Make the "power attack" part (and other similar large numerical boosts) either class features that appropriate classes had access to (like fighting styles) or just include them in the base rules. This prevents all "optimized" archers from being battlemaster fighters with SS wielding hand crossbows (for example). Or all "optimized" paladins from wielding a QS and a shield (or a halberd to combine GWM and PAM).

I don't have a problem with the overall level of power--it's the concentration into a few feats and associated weapon types that is a problem for me.

Another way to look at it is this: if toning down these feats makes martials severely underpowered, then we need to do so and buff them elsewhere. This is because feats are optional. If, in a no-feat (no PAM/GWM/SS/CBE) game, martials can't even compete in their specialty, then buff martial classes in their class features. Let feats give you new options--variety, not necessity. No one should feel "compelled" to take a feat/adopt a certain style/avoid another style to remain competitive.

Thank you for this.

This general take on the issue is more or less how I feel about it. The Big Three fighting style feats aren't a problem just because they're strong. They're a problem because they have the ability to make other options much less attractive by contrast. Polearm Master especially because it tramples all over the identity of two weapon fighting as the 'moar attacks!' style. If these were to be reigned in, I really don't think it would be a terrible idea to simply make power attacking (or overdrawing) an inherent weapon feature that scales with Prof bonus. That would prevent the effect from being overpowering early in the game while also giving other fighting styles another tool to work with. Something like:

Power attack: When you make an attack with a weapon you're proficient with you may decline to apply your proficiency bonus ToHit before making the attack roll. If the attack hits, you deal bonus damage equal to twice your proficiency modifier.

[FinnS]

Thank you for this.

This general take on the issue is more or less how I feel about it. The Big Three fighting style feats aren't a problem just because they're strong. They're a problem because they have the ability to make other options much less attractive by contrast. Polearm Master especially because it tramples all over the identity of two weapon fighting as the 'moar attacks!' style. If these were to be reigned in, I really don't think it would be a terrible idea to simply make power attacking (or overdrawing) an inherent weapon feature that scales with Prof bonus. That would prevent the effect from being overpowering early in the game while also giving other fighting styles another tool to work with. Something like:

Power attack: When you make an attack with a weapon you're proficient with you may decline to apply your proficiency bonus ToHit before making the attack roll. If the attack hits, you deal bonus damage equal to twice your proficiency modifier.

And again, Martial characters are already behind casters.
There shouldn't be any nerfing at all of Martial classes and feats.
Especially if the main reason is because dual weilders feel butt hurt.
Find a reasonable buff to TWF then.

[Easy_Lee]

And again, Martial characters are already behind casters.
There shouldn't be any nerfing at all of Martial classes and feats.
Especially if the main reason is because dual weilders feel butt hurt.
Find a reasonable buff to TWF then.

It's not just TWF. The issue is that every weapon that isn't a bow, polearm, or great weapon is inferior for damage. Not sure when the last time you looked at the weapon table was, but FYI, it includes longswords. Longswords are about as iconic as swords get. And for martial characters, they're one of the worst options. Even Shield Masters are better off using Dexterity, so they'd rather have a rapier.

Regarding casters being superior, that's true in social encounters and at later levels. But that isn't what it's about. It's about making all weapons viable.

I have a simple solution, one posted many times in the past: just remove the weapon limitations from PAM and GWM. Rename GWM Cleave and PAM to Stylin' On Em' or whatever, and just let them work with any melee weapon. Sharpshooter already works with any ranged attack, so you could use it with a sling or thrown weapon too if you were so inclined. That's by far the easiest thing to do.

And if you're worried about low level damage from SS and GWM, just switch the -5 to disadvantage and the +10 to 2*Prof.

[Rebonack]

And again, Martial characters are already behind casters.
There shouldn't be any nerfing at all of Martial classes and feats.
Especially if the main reason is because dual weilders feel butt hurt.
Find a reasonable buff to TWF then.

Assuming Power Attack becomes available to everyone as detailed above-

Great Weapon Style: If you roll 9 or higher on your heavy weapon's damage die, you deal 6 bonus damage.
Great Weapon Master: While wielding a heavy melee weapon without the reach property you gain several benefits.
-If you reduce a foe to zero HP you may move up to half your Movement and make a single attack as a bonus action.
-When you deal damage to a foe, you may deal damage equal to twice your proficiency bonus to an adjacent foe.

Polearm Style: While wielding a weapon with the reach property all hostile creatures treat your threatened area as difficult terrain.
Polearm Master: While wielding a weapon with the reach property you gain several benefits.
-You may make an Opportunity Attack against a foe who enter your reach.
-As a bonus action you may enter a defensive stance until the beginning of your next turn. While in this defensive stance your reach is increased by 5 feet and Opportunity Attacks do not consume your Reaction.

Two Weapon Style: You may add your ability modifier to your off-hand attack. Additionally, you may draw two weapons with a single item interaction.
Dual Weapon Master: While wielding a weapon in each hand you gain several benefits.
-Attacking with your off-hand weapon no longer requires a bonus action. Instead, you may make one additional attack when you use the Attack action. This feature stacks with the Extra Attack class feature.
-As a bonus action, you may focus on a single foe until you slay them or attack another target. Each consecutive attack against the focused target deals 1d4 bonus damage, stacking with itself (1d4 on the first strike, 2d4 on the second, ect).

Guardian: If you are wielding a shield, all allies within 5 feet of you benefit from your shield bonus to armor class. A creature can only benefit from one shield bonus to armor class.
Sentinel: You have mastered techniques to take advantage of every drop in any enemy’s guard, gaining the following benefits.
-When you hit a foe with an opportunity attack their speed is reduced to zero for the rest of the turn.
-Creatures provoke opportunity attacks from you even if they take the Disengage action before leaving your reach.
-When a creature within 5 feet of you would hit with an attack against a target other than you (and that target doesn’t have this feat), you may use your Reaction to make an interrupt attack against that foe. If you are wielding a shield and this weapon attack hits, the foe's attack against your alley is resolved at disadvantage.

And so on and so forth. Exact numbers used subject to change, ect ect.

As said already, my goal here isn't a net nerf of martial classes. That would be silly. Just knocking power attack out of the game would be too disruptive. Making it an option for everyone makes sense already. The Great Axe user has his scary leap attack. Guy with a rapier has his lunge. Both follow the same general concept. Dialing back the risk/reward at lower levels also makes sense due to low HP values coupled with how bound accuracy works. The goal is instead to provide each fighting style with its own unique tactical niche while aiming for a net shift in power that leaves everything more or less the same. Great Weapon style is for smashing groups of foes. Polearm is for zone control. Two Weapon is for hacking down single large targets. Guardian/Sentinel is for protecting your buddies.

All of these are, of course, just rough ideas. But I hope they get the point across. 'Nerfing' the three big style feats would allow fighting styles to be more diverse.

[Cybren]

It's not just TWF. The issue is that every weapon that isn't a bow, polearm, or great weapon is inferior for damage. Not sure when the last time you looked at the weapon table was, but FYI, it includes longswords. Longswords are about as iconic as swords get. And for martial characters, they're one of the worst options. Even Shield Masters are better off using Dexterity, so they'd rather have a rapier.

Regarding casters being superior, that's true in social encounters and at later levels. But that isn't what it's about. It's about making all weapons viable.

I have a simple solution, one posted many times in the past: just remove the weapon limitations from PAM and GWM. Rename GWM Cleave and PAM to Stylin' On Em' or whatever, and just let them work with any melee weapon. Sharpshooter already works with any ranged attack, so you could use it with a sling or thrown weapon too if you were so inclined. That's by far the easiest thing to do.

And if you're worried about low level damage from SS and GWM, just switch the -5 to disadvantage and the +10 to 2*Prof.

how are shield masters better off using dexterity, the ability score that most specifically does not apply to their bonus action shove?

[mephnick]

Sharpshooter already works with any ranged attack, so you could use it with a sling or thrown weapon too if you were so inclined.

Nope, ranged weapon attacks. Throwing weapons unfortunately do not work with SS as they are all considered melee weapons with the thrown property.

[Zman]

Spoiler: Specific responses

Show

I have tried to make sure to note where we agree and that my points are not personally directed at you or Kryx. I see that I have let my irritation get the best of me and have failed in that attempt. I believe your work is valuable. I also believe that it is frequently taken out of context and applied where it can't apply and used to make claims it cannot support. That is the problem I see.



My comments were directed at those who use yours (and Kryx's math) to argue such things. If you feel that it doesn't apply to you, I'm sorry that I unfairly tarred you with that brush.



Except where that rough and dirty math is used as a weapon against others, making claims it can't in truth support. My original attempt was to clarify that the numerical work has rather strong limitations when applied beyond a limited region.



I'm not making any determinations other than "I don't like it the way it is." I think that optimization (especially for DPR) is significantly overvalued and needlessly restrictive. I care much more about perceived imbalance--how it feels in play. If that's "using my gut," so be it.



By not relying on math at all. I'm relying on the reports of people across many threads who hone in on and push these feats as "the one and only true way" of being a martial. That in and of itself is enough for me to decide I don't like how the feats feel. I fully admit that this is pure personal taste and others may differ. I have hunches about why it feels wrong, but since I've never seen the problem myself I'm not willing to take it beyond that.

The devs may have not used "real" math for balancing, but they also didn't rely on math as the primary sense of balance. I don't believe that your math is useless, just not formally correct nor valid beyond a rough cut. My personal cut-off for caring is between 10 and 20% differences in white-room situations--others may have different view points. I just don't personally care all that much about theoretical balance beyond that point, as my experience tells me that the other factors are way more important than the theory suggests or allows for.

Zman, I'm sorry that I got contentious. This will be my last post on the matter, as I see I cannot discuss this matter without becoming angry, and that anger is clouding my perceptions and tempting me to be much harsher than I rightfully should be.

I believe that we're actually much more in agreement than in disagreement (especially about the direct subject of the OP). I want to stress that I respect the output that those like you produce and find it useful (within its limits).

No worries, I have a tendency to get snarky myself and it's easy to get overly invested and harsher than necessary. Apology accepted and respected.

I agree, I think we agree far more than we disagree, especially about the direct subject of the OP.

[Easy_Lee]

Nope, ranged weapon attacks. Throwing weapons unfortunately do not work with SS as they are all considered melee weapons with the thrown property.

By that logic, the dueling fighting style should apply when you throw a one-handed weapon and aren't wielding anything else. As a weapon attack doesn't require a weapon (unarmed strike), I argue that a ranged weapon attack doesn't require a ranged weapon.

I'm fully aware that SS isn't intended to apply to thrown weapons. But that doesn't mean that it isn't what the RAW says. Now, we can get into it and derail this whole thread, but I'd rather not. In a thread that's clearly about changing rules, does it really matter what WotC intends?

Besides, it's beside the point. This is only a small portion of my post, and if you were going to change the other feats anyway, may as well update sharpshooter.

[jas61292]

And again, Martial characters are already behind casters.
There shouldn't be any nerfing at all of Martial classes and feats.
Especially if the main reason is because dual weilders feel butt hurt.
Find a reasonable buff to TWF then.

And again, this entirely misses the point. All these feats do is bump up damage. The one thing Martials already are the best at is... you guessed it, damage. You could bump up their damage even more than these feats already do and they would still be considered weaker than casters because damage is just a single element of the game. Acting like these feats are necessary for game balance is completely dishonest, because they do absolutely nothing to change the balance between casters and martials. What they do is utterly ruin the game balance between different types of martials, while not changing a thing in how they compare to casters. The fact that the classes that best use them are on the weaker side (in theory) does not excuse their incredibly poor game design.

[Sindeloke]

I have a simple solution, one posted many times in the past: just remove the weapon limitations from PAM and GWM. Rename GWM Cleave and PAM to Stylin' On Em' or whatever, and just let them work with any melee weapon. Sharpshooter already works with any ranged attack, so you could use it with a sling or thrown weapon too if you were so inclined. That's by far the easiest thing to do.

While simple (a significant virtue that all houserules should strive for), this only solves half the problem. Iconic weapons stop being directly inferior to other weapon styles (minus TWF, which still competes for the bonus action), but the feats are still so powerful and so much better than the +2 Str/Dex they compete with (or idk, a trash feat like Savage Attacker) that they feel somewhat obligatory.

I think the approach Rebonack advocates is the best. Burn the feats, make Power Attack a default option (my table actually uses "If you don't already have disadvantage, accept disadvantage to roll weapon damage an extra time," which hits on several goals but mostly is about that same concept of trading accuracy for damage), leave bonus action attacks for unique situations like Frenzy and War Magic or whatever (although the idea of a pure Cleave feat is appealing). If you feel like the lack of access to the hilt strike from PAM leaves martials behind the acceptable damage curve, put it back through the base class chassis where it belongs, with a rage damage bump for barbarians or a baseline proficiency bump for fighters or a stronger/earlier improved smite for pallies, but even RAW with no feats they're still well above casters for DPR (even the poor sorceror) so I wouldn't call it necessary.

[Easy_Lee]

I think the approach Rebonack advocates is the best. Burn the feats, make Power Attack a default option (my table actually uses "If you don't already have disadvantage, accept disadvantage to roll weapon damage an extra time," which hits on several goals but mostly is about that same concept of trading accuracy for damage), leave bonus action attacks for unique situations like Frenzy and War Magic or whatever (although the idea of a pure Cleave feat is appealing).

I like this idea. Many players feel martials don't have enough options. Power attack as a default option on weapon attacks would be good. Some thoughts:

• I would limit this only to martial character classes and change it to disadvantage for +2*prof. Otherwise monks and the revised beast conclave are going to do more damage than anyone wants to see at low levels. Trust me on the beast conclave: it can get four attacks per round, two of them with advantage, at level 5 with the wolf pet. You don't want that wolf pet power attacking for +10, or even +2*prof.
• Sharpshooter and crossbow expert could lose their power attack / bonus attack and become half feats.
• Possible Cleave feat: when you kill a creature with an attack, you may immediately make another attack against a creature within your reach. This is more like the 3.5e version, and could set up some funny chains that, I think we can all agree, wouldn't make martials too strong.

[ZorroGames]

So Polearm Master, Sharp Shooter, and GWM are grossly out of line with the general power level of feats. PAM and GWM both step on Two Weapon Fighting's toes by adding an extra attack as a bonus action. And SS is simply overloaded. Any suggestions on reigning these feats in so they don't completely overshadow everything else? My kneejerk reaction is split GWM and SS and simply give PAM's bonus attack the axe. Would that do it? Would game health in general be improved if these feats weren't quite so centralizing? Or are these grossly over-tuned fighting style feats required for martial folks to compete with the not-so-martial folks?

Have not read all 223 posts at this time but after several pages my answers to your OP question/statements in sentence order are:
WTF, really?

So what, 1D4? Cheese but it is not alone in that aspect.

Deal with it, ban it from your games if it bothers you, it's powerful but so are casters, are you going to ban them?

No, I like them as they are. I do not use those feats but they have a sense of cool.

No, just no.

Would it eviscerate Martials - maybe. Unlikely, but maybe.

No.

Opinion does not require an answer. This Martials players has not chosen to use any of them but it does provide flavor to other PCs. Let them have their fun.

just go to leave this thread since the arguments (and that is what they seem to be) seem to be conjecture nd opinion.

[Rebonack]

I like this idea. Many players feel martials don't have enough options. Power attack as a default option on weapon attacks would be good. Some thoughts:

• I would limit this only to martial character classes and change it to disadvantage for +2*prof. Otherwise monks and the revised beast conclave are going to do more damage than anyone wants to see at low levels. Trust me on the beast conclave: it can get four attacks per round, two of them with advantage, at level 5 with the wolf pet. You don't want that wolf pet power attacking for +10, or even +2*prof.
• Sharpshooter and crossbow expert could lose their power attack / bonus attack and become half feats.
• Possible Cleave feat: when you kill a creature with an attack, you may immediately make another attack against a creature within your reach. This is more like the 3.5e version, and could set up some funny chains that, I think we can all agree, wouldn't make martials too strong.

If I recall correctly, the unarmed strikes don't count as weapons, so RAW with how the 'power attack as a weapon feature' works it wouldn't be usable with the Monk's two bonus-action attacks.

For the beast ranger pet, I suspect specifying 'when you use the Attack action with a manufactured weapon' would rule out natural weapon shenanigans. Barring that, you could make it available only to classes with the Extra Attack feature. That means no one would get it until level 5, which wouldn't be much of a difference to anyone aside from vHumans.

I would be pretty iffy about making it disadvantage, though. That feels like it would be a much bigger hit early on and potentially become irrelevant later in the game when ToHit gets massive while AC hasn't moved much.

[Pex]

I would be pretty iffy about making it disadvantage, though. That feels like it would be a much bigger hit early on and potentially become irrelevant later in the game when ToHit gets massive while AC hasn't moved much.

Also, there are ways to get advantage countering each other to allow for a normal roll to hit with a net bonus of damage. Spells like Faerie Fire become stronger. Whether this is a good thing or bad thing is up to you.

Edit: Barbarians are now practically always power attacking to get more damage on top of their rage.

[Easy_Lee]

Also, there are ways to get advantage countering each other to allow for a normal roll to hit with a net bonus of damage. Spells like Faerie Fire become stronger. Whether this is a good thing or bad thing is up to you.

Edit: Barbarians are now practically always power attacking to get more damage on top of their rage.

In my experience, most barbarian players are already doing that. Advantage offsets the -5 effectively in most situations.

[Sindeloke]

In my experience, most barbarian players are already doing that. Advantage offsets the -5 effectively in most situations.

Exactly. We found that, rather than trying to guess enemy AC and then do the math, players with GWM would just default to "if I have advantage, power attack, if I don't, don't." Disadvantage is, on average, worth a -5 to hit in any situation where your odds of hitting are close to 50%, so the penalty is essentially the same any time that it matters.

So if it's the same why do it then? Well #1 it's more in line with the general "avoid flat bonuses" philosophy of 5e, #2 it creates tactical decisions because there are a couple other things we've added that you can also accept disadvantage to do (like target a specific body part, like a dragon's wings or a specific hydra head), and hooking them to the same trigger makes them mutually exclusive, and #3 it means that even at high levels when your to-hit is +11 and you're still swinging at tons of AC 13 enemies and there's no reason not to take a -5 ever, there are still situations where you're just going to be at disadvantage no matter what you do (RAW Reckless Attack might cancel out invisibility+underwater with a bludgeoning weapon+fighting defensively+a magic dizziness spell, but it's a pretty reasonable ruling to say otherwise), so it remains an interesting option at end game instead of basically a boring always-on damage boost.

I would limit this only to martial character classes and change it to disadvantage for +2*prof

Flat prof-based scaling damage is sensible. The idea behind making it weapon dice was that it gives a nice interaction with crits and helps make a player's choice of weapon slightly more meaningful, but if you've got a lot of low-level sources of advantage that your players are going hog wild with, I agree you'd want to restrict it. I actually really like "requires extra attack" for that, but you'd probably want to put a caveat in Thirsting Blade in that case if you want Str-based bladelocks to be worth it.