D&D 3E/3.5 Statistical Analysis of 3.5 Power-Attack: UNDERpowered?

[Plane Sailing]

One other note-power attack is most useful when you don't otherwise do much damage. This is because of the fact that your average damage is a product of hit probability and average damage per hit; so, a greater percentage increase in average damage per hit makes the power attack adjustment more useful. And when your average damage per hit is low, your percentage increase in average damage for a flat +x is high.

I've seen Andy Collins use this reasoning, and I have to say that I consider it misleading at best. Percentage increases make it sound bigger, but in real terms it is adding exactly the same amount of damage whether it is used on a dagger or a greatsword - and it is real terms that the creature you are fighting is bothered by.

It sounds like a justification, but it isn't really - it doesn't help the dagger wielder to kill the creature faster than the greatsword wielder because he is getting a "bigger percentage" increase in his weapon damage! In fact because of the bigger damage die the greatsword wielder is still doing better than the dagger wielder at every point in the PA scale.

This means that your second less obvious corollary is actually wrong. It isn't more useful for low strength characters using light weapons. High Str characters have a greater "disposable" addition to their BAB than low Str characters - the 18 Str fighter could put 3 into PA and still be getting the same overall bonus as the 13 Str fighter to hit... Using PA he can always choose to use more (within the BAB/level limits, of course)

In the example you gave, the halfling is making two attacks at -2 (when he gets a full attack, which isn't all the time), and PA for -3. He gets +1 from STR So he is making two attacks at -4, for 1d4+4 damage each. (ave 13) instead of 2x1d4+1 (ave 7)

The Orc is uses PA for 9 (base 3PA, then reducing his Str bonus from +5 to +1, then adds in an extra 2pts to equal the TWF penalty, and otherwise the same amount as the halfling), - so the half-orc is making one attack at -4 and is doing 2d6+14 (ave21) damage instead of 2d6+5 (ave 12).

edit: I forgot the half-orcs 1.5str bonus, so even in 3e rules he was doing 2d6+7 normally (ave14) and 2d6+16 with that equivalent attack chance PA (ave 23). Also the halfling doesn't get full Str on his offhand attack, so his damage is 2d4+7 and 2d4+1 with and without the equivalent PA respectively.

For the sake of simplicity I'm not comparing the percentage effect of not reducing chance to hit on the base damage, but even so I think the results are clear.

So which would you rather be attacked by?

I would say that the power attack is far and away a better choice in absolute terms for the half-orc.

The silly thing is, that TWF is still stupidly brilliant for rogues - and in 3.5e a 20th level rogue with TWF, ITWF, GTWF can be getting 6 attacks *each* at +10d6 if he flanks someone subject to critical hits. In the face of that, why do they worry about Power Attack and light weapons?

I've got nothing against you personally Christian, but I've seen this percentage maths trotted out before and it just doesn't work out the way some people seem to think it does.

Cheers

 

[Darklone]

Thanks Plane, I have to print that post out and put it in a golden frame right here next to the altar of my first RPG books...

 

[Eccles]

Correct me if I'm wrong, but doesn't your maths show that 3.5 Power Attack is far more effective than 3.0 Power Attack?

If your character already had the feat at 3.0, and you're converting to 3.5, then what's the problem?

Power Attack's still there for what it's designed for - smashing apart hordes of low AC irritants (with Cleave), or doing unbelievable amounts of damage to low AC targets.

If you're power attacking something with AC 50, you're either doing something wrong, or you've made the character decision to power attack *everything*!

 

[Zhure]

I concur that percentage increases is a bad way to look at Power Attack. I also feel the change to PA makes light weapons an even worse choice than they were before. In 3e, we saw a lot of 2-handers, some sword&board, but the only people using light weapons were small characters and TWF'ers. One of the ways to make light weapons worthwhile was to power attack with them, converting a piddling 1d4+1 into a 1d4+4. Even so it was substandard compared to the 2d6+6 the great sword wielder was bashing about.

Now in 3.5, I foresee the only ones using light weapons will be the TWF'ers, who still give up a feat and take a -2 to try to keep up with the great sword guy.

It's a completely nonsensical rules change as far as I'm concerned. While on the surface it looks like the change was made to balance out the big weapons vs the small weapons, in truth all it did was make the small weapons worth even less than they were before and the big weapons worth even more than they were before. A percentage shift in damage means nothing in a system where hit points are linear. So what if you do 200% more with PA and a light weapon.

I like a lot of the changes in 3.5, but this certainly isn't one of them. More and more I find myself seeing my games as 3.0 with a few 3.5 houserules rather than 3.5 games with a lot of 3.0 houserules.

Greg

 

[Plane Sailing]

Interestingly, with the 3.5e version of power attack in that same situation the Half-Orc will be doing 2d6+7Str+18PA for 2d6+25. The halfling has his power thankfully toned down so he is only doing 1d4+1, 1d4+0

 

[Christian]

I've got nothing against you personally Christian, but I've seen this percentage maths trotted out before and it just doesn't work out the way some people seem to think it does.

No offense taken. And you make a good point.

 

[Staffan]

I've seen Andy Collins use this reasoning, and I have to say that I consider it misleading at best. Percentage increases make it sound bigger, but in real terms it is adding exactly the same amount of damage whether it is used on a dagger or a greatsword - and it is real terms that the creature you are fighting is bothered by.

The thing is that the big-weapon fighter is giving up more, and the light-weapon fighter (especially one fighting with two weapons) is giving up less and getting more in return.

Let's compare two characters using 3.0 power attack. Both are 5th level fighter-types chopping at a regular troll.

Fighter 1 has Strength 16, Dex 14, Weapon focus (greatsword), Weapon specialization (greatsword), Power Attack, and Cleave. He's wielding a greatsword +1, so his total attack bonus is +10 for 2d6+8 damage.

Fighter 2 has Strength 14, Dex 16, Ambidexterity, Two-weapon fighting, Weapon Focus (shortsword) and Power Attack. He's wielding a shortsword +1 and a masterwork shortsword, so his attack bonus when fighting with both weapons is +7/+7 and a damage of 1d6+3 and 1d6+1.

Their target is a troll at AC 18, which seems somewhat average for the CR (5). There were at least a few monsters below that (zombies, dire lions) and some above (green hag), so I chose that as a pretty normal AC. For the purpose of this exercise, let's ignore crits (they would benefit both fighters equally). Let's see what average damage they do at various levels of power attack.

Power attack: 0
Fighter 1 hits the troll on an 8 - that's 65% chance. Each attack does 2d6+8 damage (average 15). That's 9.75 points per round.
Fighter 2 hits on a 11 - 50% chance. The first attack does an average of 6.5 points and the second 4.5 points. That's 5,5 points per round.

Power attack: 1
Fighter 1 hits on 9 (60%) for average damage 16. That's 9,6 points per round. Note: Any amount of power attack is useless here. It's not going to get better with more power attack.
Fighter 2 hits on 12 (45%) for average 7.5 and 5.5. That's 5,85 points per round. Note that unlike the greatswordsman, the two-weapon fighter benefits from power attack.

Power attack: 2
Fighter 2 hits on 13 (40%) for 8.5 and 6.5, total of 6 points per round.

Power attack: 3
Hit on 14 (35%) for 9.5 and 7.5, total 5.95 points. Power attack is no longer useful.

Still, the example shows that the greatswordsman didn't benefit from power attack at all, whereas the two-weapon fighter had a sweet spot at 2 points of power attack. This means that the two-weapon fighter benefits more (in absolute terms), even if he starts out in a worse position.

 

[Zhure]

Power attack: 0
Fighter 1 hits the troll on an 8 - that's 65% chance. Each attack does 2d6+8 damage (average 15). That's 9.75 points per round.
Fighter 2 hits on a 11 - 50% chance. The first attack does an average of 6.5 points and the second 4.5 points. That's 5,5 points per round.

Power attack: 1
Fighter 1 hits on 9 (60%) for average damage 16. That's 9,6 points per round. Note: Any amount of power attack is useless here. It's not going to get better with more power attack.
Fighter 2 hits on 12 (45%) for average 7.5 and 5.5. That's 5,85 points per round. Note that unlike the greatswordsman, the two-weapon fighter benefits from power attack.

Power attack: 2
Fighter 2 hits on 13 (40%) for 8.5 and 6.5, total of 6 points per round.

Power attack: 3
Hit on 14 (35%) for 9.5 and 7.5, total 5.95 points. Power attack is no longer useful.

Still, the example shows that the greatswordsman didn't benefit from power attack at all, whereas the two-weapon fighter had a sweet spot at 2 points of power attack. This means that the two-weapon fighter benefits more (in absolute terms), even if he starts out in a worse position.

So you're saying the light weapon user is doing less damage and expending more feats and is thus receiving a better bargain for his feat expenditure because he's accomplishing less? In absolute terms, the 2WF, with or without PA - 3.0 or 3.5 - does less damage unless he has add-on dice, like sneak attack or the flaming enhancement. With 3.5 PA, he does even less than he did before.

 

[AuraSeer]

If you're power attacking something with AC 50, you're either doing something wrong, or you've made the character decision to power attack *everything*!

Actually, in that situation you should probably be power attacking at your maximum (assuming you don't run away). If the target's AC is so high that you can only hit on a 20, your chance to hit can't possibly get worse, so there's no downside to using a power attack.

Another use for the feat is in overcoming DR. Against a monster with DR 15/whatever, a hit for 17 points is twice as good as a hit for 16 points, and anything below 15 is no better than a flat miss. Increasing your average damage is therefore more valuable than one might think, especially in 3.5 where you're less likely to always have the right weapon for every monster.

 

[Mike Sullivan]

In absolute terms, the 2WF, with or without PA - 3.0 or 3.5 - does less damage unless he has add-on dice, like sneak attack or the flaming enhancement.

This is not true in all cases.

Example follows:

Sneaky McShortsword is a two-shortsword-fighter under the 3.5 rules. He is a level 12 Fighter with an 18 Str. He has the following feats: Weapon Focus (Shortsword), Weapon Specialization (Shortsword), Greater Weapon Focus (Shortsword), Greater Weapon Specialization (Shortsword), Improved Critical (Shortsword), Two Weapon Fighting, Improved Two Weapon Fighting, and Greater Two Weapon Fighting. He uses a +3 Shortsword in his on-hand, and a +1 Shortsword in his off-hand.

Sneaky's stats:

Attack bonus for full attack:
On-hand sword: +19/+14/+9
Off-hand sword: +17/+11/+7

Damage:
On-hand sword: 1d6 + 11, 17-20/x2 crit
Off-hand sword: 1d6 + 7, 17-20/x2 crit


Bashy von Bashenstein is a greatsword fighter under the 3.5 rules. Bashy has an 18 Str. He has the following feats: Weapon Focus (Greatsword), Weapon Specialization (Greatsword), Greater Weapon Focus (Greatsword), Greater Weapon Specialization (Greatsword), Improved Critical (Greatsword). He uses a +3 Greatsword.

Bashy's stats:

Attack bonus for full attack:
+21/+16/+11

Damage:
2d6 + 13, 17-20/x2 crit


Bashy and Sneaky are each attacking an AC 18 opponent.

Bashy's expected damage per round:

He hits on a 2+, 2+, 7+. His expected damage per hit is 20 * 1.2 = 24. His expected damage per round, then, is 24 * .95 + 24 * .95 + 24 * .7 = 62.4

Sneaky's expected damage per round:

With his on-hand, he hits on a 2+, 4+, 9+. His expected damage per hit is 14.5 * 1.2 = 17.4. His expected damage per round, then, is 17.4 * .95 + 17.4 * .85 + 17.4 * .6 = 41.76

With his off-hand, he hits on a 2+, 6+, 11+. His expected damage per hit is 10.5 * 1.2 = 12.6. His expected damage per round, then, is 12.6 * .95 + 12.6 * .75 + 12.6 * .5 = 27.72.

Thus, Sneaky's total expected damage per round is 69.48, some 7 points higher than Bashy's, without sneak attacks, elemental damage, or the like.

Of course, Sneaky's one-round attacks lag way behind Bashy's, and Sneaky's feat expenditure is three more than Bashy's, and Bashy will tend to do better as AC's rise. But without any kind of Power Attack, it is possible for a two-weapon fighter to out-damage a similarly built big-weapon fighter in 3.5.