Great Weapon Fighting Style House Rule?

[TheFirstStraw]

I wanted to get some feedback before house-ruling a change to the Fighting Style Great Weapon Fighting:

The ability states that when you roll a 1 or a 2 on a damage die using a melee weapon wielded with two hands you can reroll it.

This seems much better for 2d6 great weapons like great swords and mauls and only okay for other two handed weapons, particularly the 1d12 great axe. Even if you consider the fact that you are more likely to roll a 1 or 2 again on a d6, there are more chances to "fix" the roll.

Therefore I propose the following addendum: "...or if using a great axe, extend the range 1-4."

All the other two-handers have something already going for them like reach or versatility, so I don't see a need to modify them.

Whaddya think? Am I missing something?

[Tanarii]

Be aware your rule actually makes it fairly high probability you'll end up doing less damage. Right now you have to roll a 2 then a 1 in 12 of rolling a 1. So 1/144 of doing less damage. With your rule there are (1/12)*3*((1+2+3)/12) ways to do less damage, or 18/144 ways, or 12.5%. That's 18 times more likely, or twice as likely as getting a crit.

I think they knew they were encouraging Fighters and Paladins to use Greatswords over Greataxes. Just like Brutal Critical encourages Barbarians to use a Greataxes. IMO the primary 'fix' that's needed is to make the Maul a 1d12 weapon. It just seems more like a Barbarian weapon to me.

In so far as you don't care about that, I'd add another 'fix' while you're at it: make versatile weapons reroll a 1-3.

[DrowPiratRobrts]

I would leave it alone personally. As stated above, you're making it far more likely that you'll roll less damage the second time. Furthermore, these weapons are deceptively well balanced if you look at probability. See below:

d12 weapons: 1 in 12 chance of rolling a 12. Minimum damage 1. Average 6.5

2d6 weapons: 1 in 36 chance of rolling a 12. Minimum damage 2. Average 7

So be careful. You might just make this much worse for the d12 weapons without realizing it.

EDIT: In other words, 2d6 gives the player a slightly higher mean damage (average) whereas the d12 weapons will roll max damage much more often in comparison. It's worth noting that when you take 1s and 2s away from the d12 you're actually getting a very significant boost in your potential to roll max damage and a very significant decrease in your potential to roll minimum damage. I want to crunch some numbers on how this plays out for both with GWM, so if I come to something I'll post it below.

[LudicSavant]

I wanted to get some feedback before house-ruling a change to the Fighting Style Great Weapon Fighting:

The ability states that when you roll a 1 or a 2 on a damage die using a melee weapon wielded with two hands you can reroll it.

This seems much better for 2d6 great weapons like great swords and mauls and only okay for other two handed weapons, particularly the 1d12 great axe. Even if you consider the fact that you are more likely to roll a 1 or 2 again on a d6, there are more chances to "fix" the roll.

Therefore I propose the following addendum: "...or if using a great axe, extend the range 1-4."

All the other two-handers have something already going for them like reach or versatility, so I don't see a need to modify them.

Whaddya think? Am I missing something?

Math time!

GWF greatsword has an average damage of 8.33 repeating. To calculate this is simple:

For a normal d6, your average would be (1+2+3+4+5+6)/6, or 3.5.
For a d6 that rerolls on certain numbers, and takes the second roll regardless of result, we can reason that in case "rolls a 1" and "rolls a 2" you get average damage. So it'd look like (3.5+3.5+3+4+5+6)/6, or 4.166 repeating.
Just multiply by two to get the average.

We can use this formula to derive the average for all of the relevant cases for your balance analysis and houserule.

Average damage (unmodified Greatsword) = 7
Average damage (unmodified greataxe) = 6.5
Average damage (GWF Greatsword) = 8.33 repeating
Average damage (GWF greataxe) = 7.33 repeating.
Average damage (GWF houserule greataxe) = 7.833 repeating (still a full 0.5 less that the greatsword's total)
Damage gain from GWF style (Greatsword) = 1.33 repeating.
Damage gain from GWF style (greataxe) = 0.833 repeating (a full 0.5 less than the greatsword's gain, and the greatsword was already 0.5 average damage ahead, and had bell curved results which are generally superior regardless of average).
Damage gain from GWF style (GWF houserule greataxe) = 1.33 repeating.

So, with this, the greataxe is still worse than the greatsword with no extra powers, but only a little worse. However, the greataxe becomes better when factoring in abilities that let you roll an extra weapon die (which adds 1d12 for the greataxe and 1d6 for the greatsword).

[TheFirstStraw]

Be aware your rule actually makes it fairly high probability you'll end up doing less damage. Right now you have to roll a 2 then a 1 in 12 of rolling a 1. So 1/144 of doing less damage. With your rule there are (1/12)*3*((1+2+3)/12) ways to do less damage, or 18/144 ways, or 12.5%. That's 18 times more likely, or twice as likely as getting a crit.

I think they knew they were encouraging Fighters and Paladins to use Greatswords over Greataxes. Just like Brutal Critical encourages Barbarians to use a Greataxes. IMO the primary 'fix' that's needed is to make the Maul a 1d12 weapon. It just seems more like a Barbarian weapon to me.

In so far as you don't care about that, I'd add another 'fix' while you're at it: make versatile weapons reroll a 1-3.

Thanks for mathing that out. And you are totally right about brutal critical being the opposite; I never realized. That is indeed balanced well. I guess it makes sense that each class has their ideal gear and builds, and while variety/novelty is welcome, it doesn't make sense to arbitrarily optimize it.

The RAW, in retrospect, seem good enough to roll with. Were I to revise, I might say 1-3? There is the chance of doing less damage overall, but maybe that is a fun gamble rather than the constancy of 2d6?

Thanks again. :smallsmile:

[LudicSavant]

The "chance that you do less damage is bigger" thing doesn't change the fact that the reroll option is still mathematically superior. It doesn't "make it worse for greataxes."

[youtellatale]

I wanted to get some feedback before house-ruling a change to the Fighting Style Great Weapon Fighting:

The ability states that when you roll a 1 or a 2 on a damage die using a melee weapon wielded with two hands you can reroll it.

This seems much better for 2d6 great weapons like great swords and mauls and only okay for other two handed weapons, particularly the 1d12 great axe. Even if you consider the fact that you are more likely to roll a 1 or 2 again on a d6, there are more chances to "fix" the roll.

Therefore I propose the following addendum: "...or if using a great axe, extend the range 1-4."

All the other two-handers have something already going for them like reach or versatility, so I don't see a need to modify them.

Whaddya think? Am I missing something?

I don't think it is game changing. Really getting into the probability behind the rolls and understanding what it does helps. Doubling the re-roll range for the d12 is interesting:

By changing the re-roll to 1-4 instead of 1-2 then your chances of rolling a 1 or 2 move from 1.4% to about 2.8% but your chances of a 3 or 4 go down pretty dramatically from 9.7% to 2.8%. Your chances of rolling a 5-12 go from 9.8% to about 11.1%. Essentially you're setting a higher floor for yourself because you get to re-roll so often. With this change your likelihood of rolling a 1-4 is only at 11.1% while rolling anything above that is at 88.9% (give or take some rounding). With standard GWM your chances to roll a 1-4 is 22.2% and your chances of rolling a 5 or higher is at 77.8% (give or take).

If the goal is to raise the floor of the roll and increase odds of rolling higher than this does that quite well. You could try bumping it to a re-roll on 1-3 instead as that aligns a little better (in my opinion).

[Tanarii]

EDIT: In other words, 2d6 gives the player a slightly higher mean damage (average) whereas the d12 weapons will roll max damage much more often in comparison. It's worth noting that when you take 1s and 2s away from the d12 you're actually getting a very significant boost in your potential to roll max damage and a very significant decrease in your potential to roll minimum damage. I want to crunch some numbers on how this plays out for both with GWM, so if I come to something I'll post it below.Yup. Distribution is an oft overlooked aspect of the mathy part of numbers. Everyone focuses on DPR, variance is important too.

Especially when discussing things like Great Weapon Fighting or TWF or Savage Attacks.

[LudicSavant]

Yup. Distribution is an oft overlooked aspect of the mathy part of numbers. Everyone focuses on DPR, variance is important too.

The thing is that counting distribution strengthens the case for 2d6 even more, rather than helping 1d12 catch up. For more discussion of why this is, see threads like this one (https://www.dndbeyond.com/forums/dungeons-dragons-discussion/tips-tactics/21204-1d8-or-2d4).

The only cases 1d12 is better are when either
A) You absolutely must roll near max damage in order to take out the enemy this turn.
B) You have extra abilities that add an extra weapon die (since 1d12 is a bigger bonus than 1d6).

[TheFirstStraw]

So I think the consensus is that the Great Weapon Fighting Style is objectively better for 2d6 weapons by design. But that's okay because barbarians are meant to specialize in greataxes, and that specialization is inherent in the game.

More importantly, my "fix" would be the same damage output or worse.

Thanks all.

[Willie the Duck]

So, with this, the greataxe is still worse than the greatsword with no extra powers, but only a little worse. However, the greataxe becomes better when factoring in abilities that let you roll an extra weapon die (which adds 1d12 for the greataxe and 1d6 for the greatsword).

This little quirk of the rules seems to be the real culprit, as it were. It seems like greataxe does 1d12 and the other 2-handed-non-reach-weapons do 2d6 exclusively for this little exploit. If I were to try to make greataxes better for everyone except barbarians, instead of messing with GWF, I would simply change great axes to 2d6.

[CTurbo]

I actually think GWF style is the worst Fighting Style and I have chosen Defense every time for my GWM builds. The only niche situation that I think I would take it in is if I was an unarmored Fighter/Barb mix with GWM.


I think GWF should have instead been what the Savage Attacker feat is. You basically roll for damage with advantage and take the higher total.


Side note: I like how 4e had the "brutal 1" or "brutal 2" weapon tags and I wish 5e would have kept it. GWF would be better if it just had the brutal 1 feature where you rerolled a 1 as many times as it takes until you got something else. I think brutal 2 would be TOO good although I'd love to see the math comparing brutal 2 vs my Savage Attacker idea. 4e had some 2d6 weapons with the brutal 2 tag and the minimum damage you could do was 6+ Str. I could see allowing the Maul and Greatsword to have brutal 1 and the Greataxe brutal 2. That still makes minimum damage from the 2d6 weapons 4+ Str while the Greataxe would be 3+Str.

Off topic side note: I think the Savage Attacker feat is kind of terrible for a feat, and have long houseruled that you could use it on any and every attack instead of just once per turn.

[DrowPiratRobrts]

Yup. Distribution is an oft overlooked aspect of the mathy part of numbers. Everyone focuses on DPR, variance is important too.

Especially when discussing things like Great Weapon Fighting or TWF or Savage Attacks.

Exactly! So I don't think it'll be game breaking, but the math shows that the mean slightly increases for the d12 which stilts the distribution even more. Because one thing you have to remember is that statistics only apply strictly in very large quantities compared to the number of possible outcomes. And the above point about d12 being far superior for things like Brutal Critical is well said.

[DrowPiratRobrts]

So I think the consensus is that the Great Weapon Fighting Style is objectively better for 2d6 weapons by design. But that's okay because barbarians are meant to specialize in greataxes, and that specialization is inherent in the game.

More importantly, my "fix" would be the same damage output or worse.

Thanks all.

No. You're actually increasing damage slightly still, even though there's also an increased chance that the reroll could end up worse. The question is if you really want to go through with it when d12 weapons have other aspects that make them better in some ways than 2d6 weapons. Namely the Brutal Critical and similar cases, as well as the distribution tendency to increase even more towards rolling max damage.

Just keep in mind that the 2d6 weapons will roll max/high damage roughly 3 times less often than the d12. That matters when you think about any single roll (such as a crit) or any added weapon dice.

[Tanarii]

The thing is that counting distribution strengthens the case for 2d6 even more, rather than helping 1d12 catch up. For more discussion of why this is, see threads like this one (https://www.dndbeyond.com/forums/dungeons-dragons-discussion/tips-tactics/21204-1d8-or-2d4).
This is not the case. Many people like the chance of getting higher values more often, even if the cost is rolling lower more often as well.

Consistency is not inherently "better". Its just what some people prefer.

Once you add rerolls in, the difference in variance gets fairly interesting.

[DrowPiratRobrts]

The thing is that counting distribution strengthens the case for 2d6 even more, rather than helping 1d12 catch up. For more discussion of why this is, see threads like this one (https://www.dndbeyond.com/forums/dungeons-dragons-discussion/tips-tactics/21204-1d8-or-2d4).

The only cases 1d12 is better are when either
A) You absolutely must roll near max damage in order to take out the enemy this turn.
B) You have extra abilities that add an extra weapon die (since 1d12 is a bigger bonus than 1d6).

C) Any time you roll a crit. I personally would rather have the highest chance possible of rolling max damage when I crit. Especially when adjusting the odds with GWM.

So for crit fishing builds in particular the 1d12 weapons are really appealing.

[Cynthaer]

Two thoughts.

First, while RAW doesn't quite say, "anybody can reskin any weapon as anything they like", it's pretty damn close and everybody allows it anyway. As others have noted, it's not coincidence that greatswords play nice with GWF while greataxes synergize with half-orc and barbarian features—but if a player wants a big axe and GWF, there's no reason not to let their axe deal 2d6.

Second, as a DM, I've found an approach to houseruling this kind of tiny, slightly annoying imbalances that works really well: Wait until a player is actually annoyed by something, then tailor the fix to whatever they want to do.

The weapons table is a great example for this, because it's chock full of minor mechanical imbalances. Everybody's got at least one thing they want to reach in and fix about it.

Now, you could homebrew your own custom list of weapons, perfectly balancing damage based on whether each weapon is martial, ranged, light, thrown, two-handed, etc. Or you could give every weapon a unique property until they all have something to offer.

My approach is simpler. I told my players the following in session zero:


"If you ever find yourself stressing on a character-building decision because the mechanically 'better' choice doesn't match what you want to do with your character, let me know. Either you're wrong about that choice being much better, and I can show you why, or you're right and we'll houserule something to make them comparable again."

So far, it's worked quite well. The fact is, most imbalances don't actually bother most players, so why spend energy in advance on "fixing" something that never would have been a problem? Plus, the fewer houserules you have, the more your players can continue to use their PHB as written.

But the biggest benefit is that the fix can be crafted in reaction to whatever the precise thing is that is making this specific player unhappy. It's great, because you don't have to devise a system that handles all possible players in all possible scenarios.

The only caveat I can think of is that this might not work very well if you have immature players, or one or more players who will try to game the system. This doesn't really have any defenses against that by design—it assumes we're all going to be honest about what we want from each other.

[LudicSavant]

First, while RAW doesn't quite say, "anybody can reskin any weapon as anything they like", it's pretty damn close and everybody allows it anyway. As others have noted, it's not coincidence that greatswords play nice with GWF while greataxes synergize with half-orc and barbarian features—but if a player wants a big axe and GWF, there's no reason not to let their axe deal 2d6. This is basically how I handle it as a DM.

[stoutstien]

I just made great weapon fighting style count a single one on damage rolls as 3. Same result, no rerolls.

[Nifft]

Second, as a DM, I've found an approach to houseruling this kind of tiny, slightly annoying imbalances that works really well: Wait until a player is actually annoyed by something, then tailor the fix to whatever they want to do.

If you have thoughtful players who analyze or read up on a system before playing it, you may never see those annoyances.

Your players will just adapt to avoid them, and you'll just see a smaller number of options being used -- the ones that have no associated annoyances.

[DrowPiratRobrts]

I just made great weapon fighting style count a single one on damage rolls as 3. Same result, no rerolls.

How is this the same result in any way? You're lowering the damage output in both cases and putting the greatsword at a disadvantage compared to the greataxe.

[tonberrian]

Really the rerolling of damage dice is an obnoxious way to get what works out to slightly more than 1 extra damage per hit. I know Great Weapon fighting style adds more to greatswords than greataxes, but its only a very marginal difference, even compared to the .5 average damage (and better consistency) of Greatswords vanilla.

[stoutstien]

Great axe 1-12 reroll 1 =3 so really breaks Down to an extra 3 on the die.
Great sword 2d6 reroll one (1)=3 so extra 3 on one die.
I pull up the math if you like but it almost a 20% in average damage.

[LudicSavant]

I know Great Weapon fighting style adds more to greatswords than greataxes, but its only a very marginal difference, even compared to the .5 average damage (and better consistency) of Greatswords vanilla.

The difference between the amount of damage that Great Weapon Fighting style adds isn't "very marginal even compared to .5" because it literally is 0.5

GWF grants +0.83 average damage to greataxes, and +1.33 to greatswords. 1.33-0.83=0.5

[BeefGood]

To avoid rerolling, redefine GWFS as +1 damage.

[Boci]

Great axe 1-12 reroll 1 =3 so really breaks Down to an extra 3 on the die.
Great sword 2d6 reroll one (1)=3 so extra 3 on one die.
I pull up the math if you like but it almost a 20% in average damage.

Pretty sure a 1 into a 3 is 2 extra damage, not 3.

[CTurbo]

Nobody like my GWF = Savage Attacker idea? Roll for damage with advantage and take the higher. It's what I'm going to use from now own instead of rerolling 1s over and over again.

[Boci]

Nobody like my GWF = Savage Attacker idea? Roll for damage with advantage and take the higher. It's what I'm going to use from now own instead of rerolling 1s over and over again.

Its a substantial buff to GWF, a style that needed it least.

And just to be clear, you only reroll once, not over and over. Unless you mean for every attack.

[CTurbo]

Its a substantial buff to GWF, a style that needed it least.

And just to be clear, you only reroll once, not over and over. Unless you mean for every attack.

Yeah I know how it's written but as I said above, I think it's incredibly weak, so I house ruled that GWF gives the old 4e Brutal 1 property to any weapon wielded with 2 hands.

[Tanarii]

Really the rerolling of damage dice is an obnoxious way to get what works out to slightly more than 1 extra damage per hit.
What it does is tend to turn a 1 or a 2 into a more average roll for Greataxes. That's fairly nice. It may average out to almost +1 per hit, but changing your small hits into more typical hits is nice feel good effect. Just adding one damage or setting a minimum wouldn't feel anywhere as good.

For Greatswords and Mauls, it's a less obvious effect, even if the average gain is higher over time. But if you use a GS or Maul in the first place, you're probably more concerned with getting the average damage over swingy damage in the first place.

---------

(Hmmm I'm doing something wrong with anydice. I'll work on it to see if I can get us a proper damage curve for reroll 1-2)

Edit: okay here's the damage curves:
https://anydice.com/program/111f5

So Greataxes have a higher chance of getting a 12, about the same of 6 & 11, and a lower chance of doing 7-10. So I was wrong. The variance actually works against the Greataxes, not in its favor. You have a slightly greater chance of getting max damage. But nowhere near as good a chance of getting high damage.

[LudicSavant]

So I was wrong. The variance actually works against the Greataxes, not in its favor.

Yeah, that's exactly what I told you earlier. Heh.

[Tanarii]

Yeah, that's exactly what I told you earlier. Heh.
Yup. That's why I had to call myself out when I proved you right. :smallwink:

[jas61292]

If you rally want to average out the damage between the weapons, what you could do is add exactly one extra die to the roll. Then drop the lowest. This is a boost in power, but averages out to slightly less than dueling's straight +2, and makes the greataxe and greatsword nearly identical in average damage (technically the axe actually come out on top in average damage here, but the difference is a negligible .03 damage per hit). That said, the axe still stays fairly swingy in comparison to the sword.

Not sure I'd use this, but the math checks out fairly well.

[Tanarii]

That said, the axe still stays fairly swingy in comparison to the sword.
Not really. Like advantage, this significantly skews the axe to higher values. It doesn't introduce a bell curve.

The axe has a 30% chance of getting a 11-12, or a 55% chance of getting a 9+.

Meanwhile the GS retains a bell curve, but skewed.

Link: https://anydice.com/program/7c4b
Check out normal and at least

[jas61292]

Not really. Like advantage, this significantly skews the axe to higher values. It doesn't introduce a bell curve.

The axe has a 30% chance of getting a 11-12, or a 55% chance of getting a 9+.

Meanwhile the GS retains a bell curve, but skewed.

Link: https://anydice.com/program/7c4b
Check out normal and at least

Oh, I know. I ran those exact outputs myself. Yes, it does skew higher, but it is overall less consistent, being more likely to roll very low. This is easily seen in its higher standard deviation. Maybe not as swingy as it was,
but still more than the sword. Now, whether or not that makes it worth using or not is up to opinion, but when the average damage is near identical, I'm not too worried about that answer.

[Cynthaer]

If you have thoughtful players who analyze or read up on a system before playing it, you may never see those annoyances.

Your players will just adapt to avoid them, and you'll just see a smaller number of options being used -- the ones that have no associated annoyances.

That's absolutely true, and that's why I make a point of mentioning it in session zero. It's important that they know I'm willing to work with them in advance, because otherwise they'd just silently avoid trying their two-weapon barbarian or whatever.

I should note that when I say "wait until a player is actually annoyed by something", I include "being annoyed during character creation because the thing I want to do is weaker", and not just "being annoyed during gameplay after realizing something doesn't work".

[Cynthaer]

This doesn't contradict anything that's been said in this thread, but I think it's worth mentioning:

From a player experience perspective, there's real value in being more likely to roll max damage.

Like, mathematically 1d12 has worse average output than 2d6, and it's more likely to waste damage on overkill. But, you're also much more likely to roll a 12 and watch everybody else at the table raise their eyebrows as the DM says, "Jesus, you chopped that guy in goddamn half".

For a lot of players, that's way more valuable than dealing more damage on average over the course of the campaign. Not coincidentally, those players are likely to play barbarians, which rewards the greataxe, etc.

"Impressing your friends with max damage" can't be calculated into mathematical balance, but it is a balancing factor in overall game enjoyability across different types of players. Just something to keep in mind.

[Willie the Duck]

there's real value in being more likely to roll max damage.

Like, mathematically 1d12 has worse average output than 2d6, and it's more likely to waste damage on overkill. But, you're also much more likely to roll a 12 and watch everybody else at the table raise their eyebrows as the DM says, "Jesus, you chopped that guy in goddamn half".

For a lot of players, that's way more valuable than dealing more damage on average over the course of the campaign. Not coincidentally, those players are likely to play barbarians, which rewards the greataxe, etc.


I suspect that that is the 'why' as to why this mechanic exists at all, rather than a simple static damage boost. Rolling a 2 on an attack with a greataxe is a psychological loss (perhaps worst than having missed outright), and the opportunity to try again is powerful.

[DrowPiratRobrts]

What it does is tend to turn a 1 or a 2 into a more average roll for Greataxes. That's fairly nice. It may average out to almost +1 per hit, but changing your small hits into more typical hits is nice feel good effect. Just adding one damage or setting a minimum wouldn't feel anywhere as good.



Yeah, this is the whole idea of "loss aversion." Numerous researchers have shown that people typically prefer to avoid a loss two times as much as they prefer an equal gain. So would you rather gain a small boost to all your rolls that slightly raises your average, or would you rather reroll your bad rolls to slightly raise your average. Typically the latter will "feel" twice as satisfying, which is a key factor in any game mechanic.

[PhoenixPhyre]

Yeah, this is the whole idea of "loss aversion." Numerous researchers have shown that people typically prefer to avoid a loss two times as much as they prefer an equal gain. So would you rather gain a small boost to all your rolls that slightly raises your average, or would you rather reroll your bad rolls to slightly raise your average. Typically the latter will "feel" twice as satisfying, which is a key factor in any game mechanic.


I suspect that that is the 'why' as to why this mechanic exists at all, rather than a simple static damage boost. Rolling a 2 on an attack with a greataxe is a psychological loss (perhaps worst than having missed outright), and the opportunity to try again is powerful.

This is an important thing to keep in mind in many areas. Optimizers and theory-crafters love averages, but averages aren't exciting like rolling big numbers. Rolling dice is fun, rolling big dice is even funner. Except when they roll low. A 1 on a d12 is much more painful than a 1 on a d6. Combine this with most people being horrible at probability and statistics and...

So designers need to take that into consideration when designing mechanics--it's not just the averages that matter. There's a tradeoff between rolling more dice (always fun) and rolling bigger dice (also fun).