D&D General What is two weapon fighting good for?

[Oofta]

Ok, let's ignore everything that happens above 10th level, for the reason you cited.

A 5th TWF level fighter (standard array) will have 16-17 STR or DEX, TWF fighting style and the dual wield feat. This fighter, using its attack and bonus actions, will make 3 attacks, with +6 to hit, each of them doing 1d8+3 damage. Against an average of AC 13, that's a DPR of 16,875.

Let's make a 5th polearm master level fighter (standard array). It will have 16-17 STR, defense fighting style (mostly chosen for ease of comparison) and the polearm master feat. This fighter, using its attack and bonus actions, will make 3 attacks, with +6 to hit, two of them doing 1d10+3 damage, one doing 1d4+3. Against an average of AC 13, that's a DPR of 16,875, the exact same of the TWF. They also have the same AC (dual wield vs defense style). The Polearm Master, however:

• has reach
• can make opportunity attacks when enemies enter its reach
• has a slightly more damaging opportunity attack, in general
• its action surge attack action does more damage
• if both characters find a single powerful magic weapon, the polearm master will benefit more from it

I'd argue the polearm master is ahead of the TW fighter, and would get more ahead if higher levels see play.

Let's try with an hypothetic greatsword user. It has have 18-19 STR (it picked ability score increase at 4th) and defense fighting style (again for ease of comparison). This fighter, using its attack action, will make 2 attacks, with +7 to hit, each doing 2d6+4 damage. Against an average of AC 13, that's a DPR of 17,6 slightly more than the TW fighter. They also have the same AC.
This Greatsword fighter, however:

• doesn't have to waste its bonus action every round. It can freely second wind without losing damage, for example
• has a substantially more damaging opportunity attack
• its action surge attack action does considerably more damage
• if both characters find a single powerful magic weapon, the greatsword user will benefit more from it

Only thing the TW fighter has over this greatsword fighter is, I believe, less wasted damage on overkills. I'd still pick this over the TW fighter everyday tbh (and so have all my players for almost 10 years of campaigns - rogue excluded ofc).


If I'm building the TW fighter wrong, please tell me, but what I'm seeing is it falling off as early as 5th level.

First? The main point is that what you're calling a massive difference is around a single point of DPR per round (correct me if I misread). When that accounts for 1-3% of the monster's total HP it's not particularly relevant. If you're hitting mobs of low level creatures, carrying over damage absolutely matters. I also think overkill comes into play, another thing not accounted for.

In addition, while back I was on a "I'm going to create monster database" for personal use before DndBeyond was released so I have a pretty extensive list of thousands of entries from the MM. The average for all levels is 14, but the picture changes if you break them down by CR. The average AC is 13 at CRs 1-4, but climbs to 19 at CRs greater than 16. The actual numbers: <5 = 13, 5-10 = 15, 11-16 = 17, >16 = 19. Of course you could also break those down by proficiency bonus, etc..

But then there are decision points. What attack ability modifier do you expect? Does the PC go for higher ability score or feat at 4th level? With TWF for example, adding to the ability is giving you a +1 to attack and damage. Dual wielder only adds +1 damage for going to a d8 weapon but it also adds +1 to AC (not to mention drawing two weapons at once). A lot of people ignore the defensive side of things. If you're going PAM (which I'd consider a variation of TWM) you have to take that feat at 4th and then somehow take into consideration extra opp attacks. Not sure how you do that. Finally, the extra bonus action may or may not matter depending on type of fighter.

The analysis for this is complicated and depends on many, many things. If I'm playing a tabaxi strength based fighter, I can just max out my strength until 8th level because I'm using my claws anyway. If magical swords are as rare as hen's teeth, a single weapon may make more sense. If magic weapons are relatively common but creatures immune to specific damage types are also common, a frost blade and fire blade combo for my dual weapon fighter may make sense.

Maybe someday I'll resurrect my fight program and try to figure some of this out because that's based on survivability not just DPR. I just think that spreadsheet analysis doesn't give you the whole picture. I can tell you that in Solasta, which tracks various stats over the course of a campaign, the TWF does better for every campaign I've run but that game has modified feats so it's not a direct comparison. Most of the time the TWF does better than even the casters, but that depends on the design of the specific campaign.

 

[Corinnguard]

When you engage in two-weapon fighting, you can add your ability modifier to the damage of the second attack. Also, Whenever you are wielding a melee weapon in one hand and a different melee weapon in the other hand, you gain a +1 bonus to AC.

Do both weapons still have to have the Light and Finesse properties in order to use them in a fight? Or can you now go with the longsword/shortsword approach with this version of TWF? You need to specify weapon properties.

• with this version of the fighting style, you get a defensive bonus for fighting with two weapons. The bonus isn't as good as a shield, unless you also invest into the Dual Wielder feat, but it should be nice to have
• you can no longer stab your opponent with both your weapons at once, but if your main weapon misses, you get a second chance with your off-hand. This should play nice with rogues and their sneak-attacking needs
• this should scale well with Extra Attack, potentially turning TWF into an high-accuracy style. You don't do as much damage as with a two hander, and you don't have the AC of shield+defense style, but you have some defense and are very likely to actually land hits
• also, no bonus actions required anymore. Should be nice for rogues, but also other classes I bet

Okay, if I am reading this correctly, this version of TWF allows you to stack the AC bonus you get with this style with the AC bonus you get from the Dual-Wielder feat. So that's a nice +2 AC bonus whenever you are dual-wielding. The part where you can no longer stab (or slash or bludgeon) your opponent with both your weapons at once is a bit confusing. So if I attack an opponent with my main weapon and miss, I can follow it up with the offhand attack. But what if I hit an opponent with my main weapon, what happens to my offhand attack then? I can imagine that it could be still used as an attack against a second opponent who is also within 5 feet of you as the first opponent, or it goes from being used for offense and is used for defense instead, hence the AC bonus. Since both attacks are now a part of the Attack action, Extra Attack will now work for the both the main weapon and the offhand weapon. This would work well with the Fighter class and it's three Extra Attacks. It could mean up to 8 attacks before you throw in two uses of your Action Surge. However, I am not sure how the part about no longer being to attack with both weapons at once will work with those Extra Attacks. First attack hits, second attack blocks, etc? This part needs some clarification.

 

[Olrox17]

First? The main point is that what you're calling a massive difference is around a single point of DPR per round (correct me if I misread). When that accounts for 1-3% of the monster's total HP it's not particularly relevant. If you're hitting mobs of low level creatures, carrying over damage absolutely matters. I also think overkill comes into play, another thing not accounted for.

In addition, while back I was on a "I'm going to create monster database" for personal use before DndBeyond was released so I have a pretty extensive list of thousands of entries from the MM. The average for all levels is 14, but the picture changes if you break them down by CR. The average AC is 13 at CRs 1-4, but climbs to 19 at CRs greater than 16. The actual numbers: <5 = 13, 5-10 = 15, 11-16 = 17, >16 = 19. Of course you could also break those down by proficiency bonus, etc..

But then there are decision points. What attack ability modifier do you expect? Does the PC go for higher ability score or feat at 4th level? With TWF for example, adding to the ability is giving you a +1 to attack and damage. Dual wielder only adds +1 damage for going to a d8 weapon but it also adds +1 to AC (not to mention drawing two weapons at once). A lot of people ignore the defensive side of things. If you're going PAM (which I'd consider a variation of TWM) you have to take that feat at 4th and then somehow take into consideration extra opp attacks. Not sure how you do that. Finally, the extra bonus action may or may not matter depending on type of fighter.

The analysis for this is complicated and depends on many, many things. If I'm playing a tabaxi strength based fighter, I can just max out my strength until 8th level because I'm using my claws anyway. If magical swords are as rare as hen's teeth, a single weapon may make more sense. If magic weapons are relatively common but creatures immune to specific damage types are also common, a frost blade and fire blade combo for my dual weapon fighter may make sense.

Maybe someday I'll resurrect my fight program and try to figure some of this out because that's based on survivability not just DPR. I just think that spreadsheet analysis doesn't give you the whole picture. I can tell you that in Solasta, which tracks various stats over the course of a campaign, the TWF does better for every campaign I've run but that game has modified feats so it's not a direct comparison. Most of the time the TWF does better than even the casters, but that depends on the design of the specific campaign.

My experience with Solasta is unfortunately zero, so I can't really add anything in that regard.

Basically, my point (and the point of my previous analysis) is this: TWF starts falling off in terms of damage compared to other fighting styles at level 5. Before level 5, it can be the strongest style damage-wise. For fighters specifically, it becomes comparatively very weak as they get more and more extra attacks.
This, of course, isn't true for pure rogues (who never get extra attack/attacks) and, I suspect, for a few specific, specialized builds (can't name any, but I don't doubt they exist).

To this, add the fact that over several campaigns at my table, only rogues ever used TWF regularly (specifically because everyone realized the un-scaling math problem early on), and also that reading a bit about historical fencing made me realize that the current incarnation of TWF really doesn't represent that type of combat style well, and here I am, looking for a fix for all these issues.

 

[Stormonu]

Skill checks in a fight generally cost an action, at least in my experience.

Eh, probably considered a house rule, but in mine most such skill checks are a bonus action (Persuasion, Perception, Intimidation, Insight) or a non-action (Athletics, Acrobatics) - sort of like the "Potion chugging as a bonus action" change.

 

[Olrox17]

Do both weapons still have to have the Light and Finesse properties in order to use them in a fight? Or can you now go with the longsword/shortsword approach with this version of TWF? You need to specify weapon properties.

The idea there is that as long as you're holding a weapon in one hand and a weapon in the other, you get a +1 AC with that style. Of course, if you are dual wielding longswords (without the Dual Wielder feat), that's all you're getting, because the TWF rule still requires light weapons.

Okay, if I am reading this correctly, this version of TWF allows you to stack the AC bonus you get with this style with the AC bonus you get from the Dual-Wielder feat. So that's a nice +2 AC bonus whenever you are dual-wielding. The part where you can no longer stab (or slash or bludgeon) your opponent with both your weapons at once is a bit confusing. So if I attack an opponent with my main weapon and miss, I can follow it up with the offhand attack. But what if I hit an opponent with my main weapon, what happens to my offhand attack then? I can imagine that it could be still used as an attack against a second opponent who is also within 5 feet of you as the first opponent, or it goes from being used for offense and is used for defense instead, hence the AC bonus. Since both attacks are now a part of the Attack action, Extra Attack will now work for the both the main weapon and the offhand weapon. This would work well with the Fighter class and it's three Extra Attacks. It could mean up to 8 attacks before you throw in two uses of your Action Surge. However, I am not sure how the part about no longer being to attack with both weapons at once will work with those Extra Attacks. First attack hits, second attack blocks, etc? This part needs some clarification.

I suppose it could be written more clearly, but the intent is this: if you hit your main hand attack, you don't get the off-hand attack at all. If your main hand attack missed, got parried or whatever, you get your second chance with the off-hand attack.

If you have Extra Attack, you can do an off hand attack for every time you missed during your attack action.
So, for example, a 20th level fighter makes 4 attacks, hits two and misses two. The fighter can now make 2 off hand attacks. If the fighter had missed 3 attacks, it could now make 3 off hand attacks. If the fighter hits all of its 4 attacks, no off hand attacks for this turn.

 

[Oofta]

My experience with Solasta is unfortunately zero, so I can't really add anything in that regard.

Basically, my point (and the point of my previous analysis) is this: TWF starts falling off in terms of damage compared to other fighting styles at level 5. Before level 5, it can be the strongest style damage-wise. For fighters specifically, it becomes comparatively very weak as they get more and more extra attacks.
This, of course, isn't true for pure rogues (who never get extra attack/attacks) and, I suspect, for a few specific, specialized builds (can't name any, but I don't doubt they exist).

To this, add the fact that over several campaigns at my table, only rogues ever used TWF regularly (specifically because everyone realized the un-scaling math problem early on), and also that reading a bit about historical fencing made me realize that the current incarnation of TWF really doesn't represent that type of combat style well, and here I am, looking for a fix for all these issues.

When "comparatively weak" is thrown around, I'm sorry but it just doesn't buy much for all the reasons I just listed so I'm not going to bother repeating most of it.

Short version:

• A point or two extra DPR is a very small increase percentage wise, if it's even accurate, is rarely going to make a difference.
• DPR analysis ignores the AC of the PC
• DPR ignores the distribution of damage, especially since weapon damage is all or nothing.
• Most calculations don't scale the average AC of opponents.
• bonus point: randomizing ability scores will have a far bigger impact on variability

But obviously I'm not going to change your mind. Maybe someday I'll fire up my fight simulator and take a crack.

Oh, and Solasta is a turn based CRPG based on the core D&D rules. They have feats, but most are slightly tweaked from base D&D because they can only use the free versions. You can play custom campaigns with can go up to level 11 (soon to be 13 I think). At the end of the campaign it gives summary of damage given, taken, how many enemies each PC killed, etc. Their version of 2WF regularly comes out on top, with wizards sometimes slightly in the lead. Two-handed weapon fighters are next. Rogues are typically towards the bottom, even though it's typically easy to hide every round.

I trust the numbers I see in Solasta more than I would any spreadsheet, but that's me.

 

[Neonchameleon]

To add to that list, our paladin leverages TWF as it gives more opportunities for smite. And Improved Smite does make the offhand more useful with the extra d8 radiant.

Don't forget that the Paladin basically has no abilities that use their bonus action.

 

[Olrox17]

When "comparatively weak" is thrown around, I'm sorry but it just doesn't buy much for all the reasons I just listed so I'm not going to bother repeating most of it.

Short version:

• A point or two extra DPR is a very small increase percentage wise, if it's even accurate, is rarely going to make a difference.
• DPR analysis ignores the AC of the PC
• DPR ignores the distribution of damage, especially since weapon damage is all or nothing.
• Most calculations don't scale the average AC of opponents.
• bonus point: randomizing ability scores will have a far bigger impact on variability

But obviously I'm not going to change your mind. Maybe someday I'll fire up my fight simulator and take a crack.

Oh, and Solasta is a turn based CRPG based on the core D&D rules. They have feats, but most are slightly tweaked from base D&D because they can only use the free versions. You can play custom campaigns with can go up to level 11 (soon to be 13 I think). At the end of the campaign it gives summary of damage given, taken, how many enemies each PC killed, etc. Their version of 2WF regularly comes out on top, with wizards sometimes slightly in the lead. Two-handed weapon fighters are next. Rogues are typically towards the bottom, even though it's typically easy to hide every round.

I trust the numbers I see in Solasta more than I would any spreadsheet, but that's me.

I just ran some numbers by hand (not as good as a software, but it is what it is) for levels between 5 and 11, using the fighter class.

Great Weapon Masters are DPR kings when AC is low or they have to hit bonuses outside the baseline (Bless, magic weapons, advantage, precision strike). Polearm Masters with a d10 weapon are always equal or better than TWF in DPR, and can get into crazy territory if they also take GWM. Plus, Polearm Masters get a number of other useful combat perks, and can easily afford to take defense style for an AC boost.
TWF wins the DPR race over sword & board with duelist style (only barely at level 11), but sword & board has more AC and an extra feat to spend for whatever they want (Shield Mastery, Sentinel, Resilient or perhaps a Con increase?).

I'm honestly comfortable in saying that TWF is currently the worse combat style in D&D 5e, for level 5+, outside of rogues and the odd build (I saw people mentioning paladins, possibly, and perhaps they're right).
I remain fully ready to change my mind if solid opposing evidence emerges.

Btw, what's your fight simulator?

 

[Yora]

It's a fashion statement.

 

[Oofta]

...
Btw, what's your fight simulator?

It's a program I wrote that simulates fights, pitting different fighter builds against level appropriate monsters using a long array of randomized D20 rolls. Runs through a thousand fights or so and figures out how often the fighter wins or the monster does. Been a while since I pulled it up and still depends on assumptions, but duelist with shield master generally came out on top if you allow the shield bash before other attacks, then TWF dual weapon then GWM. I don't think I ever got around to PAM, but that gets a bit tricky. With PAM do you want to win initiative in a cage match? It also did some comparisons with champion and battle master (assuming the battle master just did extra damage).

I used it for this kind of analysis but also comparisons of rolling for ability scores which showed that random rolls make much more of a difference than fighting style. Of course it's also limited because I can't really emulate things like all of the BM's maneuvers and I assumed the fighter had action surge available 1/3 of the time.

If I were really ambitious I'd link it to my monster database and run them through a few hundred different monsters each and let it crunch. I'm not that ambitious.