Jonathan Tweet denounces Power Attack

[Ahglock]

But the evidence suggests that power attack, applied blindly, hurts as much as it helps, especially for attackers already doing decent damage.

I'm trying to find where in his post he implied anyone at his table uses power attack blindly. You don't need a spreadsheet to make a relatively informed decision on what to put into power attack after a couple rounds of combat.

 

[mmadsen]

If power attack is so obviously broken in that it doesn't even do what it claims to do, how did that get by the designers of 3E?

It's not obviously broken until you've done a bit of mathematical analysis -- and running the numbers isn't something everyone's inclined to do. The power attack mechanic seems elegant, and it seems like it should work. Further, people actually using power attack seem to think they're using it well and gaining from it, even when they're not.

 

[helium3]

It's not obviously broken until you've done a bit of mathematical analysis -- and running the numbers isn't something everyone's inclined to do. The power attack mechanic seems elegant, and it seems like it should work. Further, people actually using power attack seem to think they're using it well and gaining from it, even when they're not.

Right. But that's sorta my point. If you run the numbers, apparently it becomes obvious that power attack doesn't actually do what it's supposed to. So the designers of 3E didn't run the numbers when they designed the feat?

 

[Mustrum_Ridcully]

Right. But that's sorta my point. If you run the numbers, apparently it becomes obvious that power attack doesn't actually do what it's supposed to. So the designers of 3E didn't run the numbers when they designed the feat?

I don't know. Maybe they did, but they also had other considerations they put more emphasis on:
- Power Attack is the prerequisites for Cleave, Improved Bullrush and Improved Sunder. These feats might have been seen worthwile a weak feat (and it's not like Power Attack is useless in totally every way)
- Monte himself wrote that they wanted to reward Rules Mastery. Some feats where written deliberately in a way that wouldn't really make them good - Toughness, Skill Focus (3.0) and maybe also Power Attack. (Honestly, I don't mind the concept of rewarding Rules Mastery, but the type of reward should probably not be related to character effectiveness so much and more in gaining ability to express what you want to play or create in mechanical terms).

 

[glass]

So here's a tangentially related question for you. If power attack is so obviously broken in that it doesn't even do what it claims to do, how did that get by the designers of 3E?

Because it is not obvious, it is insidious. Just look at this thread!


glass.

 

[Nifft]

Right. But that's sorta my point. If you run the numbers, apparently it becomes obvious that power attack doesn't actually do what it's supposed to. So the designers of 3E didn't run the numbers when they designed the feat?

It's obvious if you know what to look for. If you're just looking to see how balanced it is, the numbers show it to be basically fine. If you're looking to see how usable it is, the numbers show "not very".

But it doesn't look hard to use...

Cheers, -- N

 

[Najo]

So here's a tangentially related question for you. If power attack is so obviously broken in that it doesn't even do what it claims to do, how did that get by the designers of 3E?

When I went to Gen Con to witness the release of 3.0 there was a seminar that had like 800+ people in it. We came to listen to Monte and Johnathan Tweet talk about the future of D&D 3.0. Tweet passed out a handout that had math on it for combat probablity and the math around hitting, damage and critical hits. He had percentage of chance to cause a critical hit based on crit range on the sheet. The math on the sheet was wrong though, it forgot to take into account having to roll to confirm the crit, so it had the chances to cause crits were more likely than they should be.

Likewise, the Mystic Theurge, regardless of what the team in WOTC said, was a huge mathematical mistake. I still have copies of the debate between Tweet and myself on this board. It was educational. I still to this day disagree with the prestige class bandaid to fix issues with multiclassing spellcasters. I don't know anyone who likes the class once they used it, and as owner of a game store I see alot of D&D players.

Every time a player finds a broken combo, that is a game design error. They do get missed, they do exist. Game designers are human. They can't cover the material like all of the playes can, espcially with forums like this where we can share our insights with each other.

The thing is, they do this as a job, day in and day out. They are less likely to make those mistakes. They are more likely to catch things. They get to benefit from years of working on those designs as a job, and recieving all of the feedback from the players directly.

What I am saying, they probablly, like all of us, like the idea and elegance that power attack seemed to have. But, somewhere along the lines, didn't really run the math on it or if they did thought it was fine for a low level starting feat.

But regardless, none of this matters. Because Tweet came forward and said the feat was flawed. So there is your designer catching the mistake, it happend after it was published and in play for 6-7 years..but hey, they caught it. At least it was sooner than any of us did

 

[BryonD]

But the evidence suggests that power attack, applied blindly, hurts as much as it helps, especially for attackers already doing decent damage.

I've done the math for actual play in my game. The facts that it works out real well trumps evidence based on incomplete data.

 

[mmadsen]

I've done the math for actual play in my game.

Would you mind sharing that math?

 

[BryonD]

1d12+10 attack with a +28 attack against a typical MM CR correct AC of 26. Crits for 19/X3.
Avg damage with no PA is 18.81.
Avg damage with a completely random PA between -1 and -14 is 24.78. Optimal PA is -7 with an avg damage of 27.45. Worst case for using PA is -14 with an avg damage of 21.36 or -1 with an avg of 21.09. No matter what number is selected, the expected damage improves.1d12+10 +28 AC 26

PA Hit % crit % avg damage
0 0.95 0.095 18.81
1 0.95 0.095 21.09
2 0.95 0.095 23.37
3 0.95 0.095 25.65
4 0.9 0.09 26.46
5 0.85 0.085 27.03
6 0.8 0.08 27.36
7 0.75 0.075 27.45
8 0.7 0.07 27.3
9 0.65 0.065 26.91
10 0.6 0.06 26.28
11 0.55 0.055 25.41
12 0.5 0.05 24.3
13 0.45 0.045 22.95
14 0.4 0.04 21.36