So I originally posted this on the other thread, but if you want a test case, your code should converge to the following answer if you assume no Strength bonus and that everything doubles on crit.
Assume base 5 to Hit, 15 to Crit. Assume 20 always Crits.
Assume d12 damage dice.
Assume three actions devoted to attacking.
Assume critical hit doubles dice + static bonuses.
Old PA: -1 Attack, +3 Damage.
New PA: 2 Actions, +1d12.
Old PA Attack List:
-1 Attack: 25% Miss, 50% Hit, 25% Crit = .5*(6.5+3)+.25*(13+6)
-6 Attack: 50% Miss, 45% Hit, 5% Crit = .45*(6.5+3)+.05*(13+6)
-11 Attack:75% Miss, 20% Hit, 5% Crit = .2*(6.5+3)+.05*(13+6)
Net Damage: 17.6
New PA Attack List:
-0 Attack: 20% Miss, 50% Hit, 30% Crit = .5*(13)+.3*(26)
-5 Attack: 45% Miss, 45% Hit, 5% Crit = .45*(6.5)+.05*(13)
Net Damage: 17.9
This is all raw probability, so with sufficient number of randomizations, you should converge to this result. I will note though that it seems likely that we may no longer be doubling static numbers on crit - that was a consequence of no longer confirming crits in 5e and it seems likely that they might share design philosophy there (since it helps reduce the sudden spikes on crits a little).
Considering Power Attack was a good baseline for how powerful a feat needs to be to not make martials completely irrelevant, it didn't need 'fixing' it and all other martial feats needed buffing, in most cases vastly, I know it wont compete with high level spells, but at least it would be worth showing up, not looking like fighters have a purpose yet from this, it is only GM fiat that makes monsters not ignore martial classes to go after the actual threats as it is.I think you are misunderstanding what Mark said. Power Attack needed fixing because the changes to the game's core mechanics (specifically critting on AC+10 rather than just nat 20s, plus or minus crit ranges) meant that the attack penalty was costing the attacker more damage than the feat was adding. It would have been the definition of a trap option at that point.
I think this is a good example of why getting too worked up about these previews is a bad idea until we have the entire system in front of us.
This thing doesn't quite gel for me. Specifically with the example that is being used, the Greataxe. In PF1E, the Greataxe is 20/x3. That means that my crit chances (with no feats) are:
Best Case Scenario (Confirm on 2+): 0.05 (Natural 20) x 0.95 (2+ Confirm) = 0.0475 = 4.75% Chance to Crit
Worst Case Scenario (Confirm on 20): 0.05 (Natural 20) x 0.05 (20 Confirm) = 0.0025 = 0.25% Chance to Crit
Meanwhile, from what I've heard, a Natural 20 in PF2E is still a crit, regardless of whether you beat AC by 10 or not. I don't know if a natural 1 is still always a miss, but:
Best Case Scenario (Beat AC by 10 on a 1+/2+): 1 (Chance of rolling 1+) or 0.95 (Chance of rolling 2+) = 1/0.95= 100%/95% chance to Crit, respectively.
Worst Case Scenario (Even a 20 doesn't beat AC +10): 0.05 (Natural 20) = 0.05 = 5% chance of a crit.
Notice that in PF2E, even your worst case scenario has a better chance of a critical than your best case scenario in PF1E. With the same numbers, the PF2E dude pulls ahead of the PF1 guy in crits, so Power Attack isn't costing you anything in that department. Hence why I don't see the new Critical Hit mechanics impacting Power Attack in any way.
Furthermore, if the penalty remains the same, the only way 1E Power Attack would impact your to hit in 2E more than it does in 1E would be if 2E has much less ways to increase your to-hit score. 2E iteratives are still -5 for each; which is the same as 1E, so that hasn't changed. The only way those minuses have more impact is that either the ACs got higher, you don't have as many bonuses, or both.
At that point, I can see how 1Es Power Attack could fall behind. I am curious, however, how the math turns out if you put Furious Focus in the mix. In that scenario, there's no to-hit difference between 1E and 2E on the first attack. There is on the second, of course, but then you have the detail that the second 1E attack benefits from Power Attack, while the 2E one doesn't.
Now, the minuses also increase your chance to fumble, but unless fumbles do stuff like making you drop your weapon or throw it away, or otherwise prevent you from attacking further in a round, they can't hurt you in the damage department anymore than a miss does: 0 damage is the worst you can get, getting 0 damage because you failed by 1 or by 17 changes nothing.
Unless fumbles heal the enemy. Which would be interesting if a bit weird.
|
New PA Attack List:
-0 Attack: 20% Miss, 50% Hit, 30% Crit = .5*(13)+.3*(26)
-5 Attack: 45% Miss, 45% Hit, 5% Crit = .45*(6.5)+.05*(13)
Net Damage: 17.9
That's only 95% all added together.
I'm guessing you meant the following:New PA Attack List:
-0 Attack: 20% Miss, 50% Hit, 30% Crit = .5*(13)+.3*(26)
-5 Attack: 45% Miss, 50% Hit, 5% Crit = .50*(6.5)+.05*(13)
Net Damage: 18.2
Another thing to remember, is it appears that PF2 rewards extra accuracy more than PF1. In PF1 if I succeed by 1 or by 20 it makes no difference, I do the same damage, so it was ok if I take a penalty to attack because the +X damage means more than I hit on a 5 instead of a 2. In PF2, the benefit but exceeding AC by more than 10 is worth more than an extra couple damage on each hit (especially if you would reduce your chance to go above 10 by 50% or more).
Are we calculating this based off two attacks or one attack?
I'm having a hard time truly understanding the actual variables that determine the base values, not in the equations themselves.
To elaborate on my thoughts:
In RAE in PF1, it is common for martial characters to be able to make 2 attacks. -5 to hit on the second is not detrimental to calculating reliable hitting.
In PF2, you will be making one attack, with the intention being that your chance to land a crit is more important than getting in multiple hits.
Arithmetically, in all cases where a crit is not considered, it was always better in RAEPF1 to make two attacks rather than using Vital Strike.
So, the question now is: does the game's assumed math propensiate critical hits so often that the average assumed damage on the extra die supersedes double the flat modifier thus making the second attack worthless anyway?
If the answer is yes, then the mechanics behind Power Attack as far as damage scaling and/or optimization are fine. However, we are rolling so many samples that we aren't really getting a decent sample of what to reasonably expect to see as far as variance at the table, I suspect. In practice, my brain is telling me that unless that extra d12 I'm getting for landing a crit reliably overcomes my flat modifier, then in more cases than not it will be better to roll two separate attacks and never use power attack.
Practical table use formula:
2(2d12+4)
vs
2(d12+4) + d12+4
This means I'm relying on the one extra d12 I get to roll exceeding 4 (2/3 of the time) for Power Attack to be better.
At the table I'm not seeing any real advantage to using Power Attack unless the game's math means that the attack at -5 is so much more unlikely to hit that it's not worth taking. This of course is before magic weapons or Power Attack scaling become a thing.
+1 greataxe
2(3d12+4)
vs
2(2d12+4) + 2d12+4
In this example, the mathematical advantage lies in power attack only if the two extra dice result higher than 4 and the first attack always crits while the second never does. Otherwise, without landing exactly one critical hit every turn it is not advantageous to use Power Attack ever, and even then no matter what is going on (until Power Attack adds two dice) I am reliant upon that one extra die in each scenario to be the entire purpose of the feat and decision to use it.
This is also based on the assumption that everyone uses a weapon that deals d12 damage. Once we start using weapons other than that, then the odds of Power Attack being better theoretically than just making two attacks drops significantly. Once we start using a d8 weapon, then statistically Power Attack becomes less viable than just making two attacks.
What this means to me is that Power Attack is meant to be used nigh exclusively on bigger weapons by design (this might not be bad). The assumptions about the relationship with more critical hits and how detrimental taking -5 to hit with your second attack is based solely on conjecture however, but given how it's presented it seems to be the logical inevitable conclusion.
So, unless the -5 to hit literally makes it not worth it to make that attack at all (which I suspect it does not given that it's only 25% less a chance to crit on a d20) then my simple algebra would lead me to believe that Power Attack is worthless until it adds more than one die.
To be certain, my math here does not factor in the odds of critical hits, but having 100% chance to crit with one attack means having 75% chance to crit twice. In simple math world, this means 75% of the time Power Attack is useless because:
2(d12+4) + 2(d12+4) = 2(2d12+4)
and with magic weapons
2(2d12+4) + 2(2d12+4) > 2(3d12+4)
This means a variance of 5 on AC must be more significant to your odds to hit at all than it is to land a crit on a single attack?
Where is my math wrong?
EDIT: since it may matter, this does not include any rolls of a natural 1 or 20, meaning this basically covers 90% of d20 rolls when I have to make the decision of whether or not to use Power Attack when sitting at the table.
Where is my math wrong?
Power Attack takes two actions. That leaves a third action to attack at -5.
It doesn't have to be better than an attack at -5, it has to be better than that third attack you're giving up, the one at -10. (Or something else you're doing with that action.) If you have to move up to them, you might want to hold off on Power Attack except for muscling through DR or something.
Where is my math wrong?Power Attack takes two actions. That leaves a third action to attack at -5.
It doesn't have to be better than an attack at -5, it has to be better than that third attack you're giving up, the one at -10. (Or something else you're doing with that action.) If you have to move up to them, you might want to hold off on Power Attack except for muscling through DR or something.
Then what is the point of giving me more freedom of actions, if I'm going to spend all those actions making 'full' attacks anyway?
It seems counter intuitive to the new action system, especially when I've been playing RAE for 4 years.
Where is my math wrong?Power Attack takes two actions. That leaves a third action to attack at -5.
It doesn't have to be better than an attack at -5, it has to be better than that third attack you're giving up, the one at -10. (Or something else you're doing with that action.) If you have to move up to them, you might want to hold off on Power Attack except for muscling through DR or something.Then what is the point of giving me more freedom of actions, if I'm going to spend all those actions making 'full' attacks anyway?
It seems counter intuitive to the new action system, especially when I've been playing RAE for 4 years.
Maybe you're not going to spend all your actions making full attacks. But people who want to spend all their actions attacking should get options to make that better- catch up with the more tactical sword-and-board a little. Power Attack is one of those options.
Maybe you're not going to spend all your actions making full attacks. But people who want to spend all their actions attacking should get options to make that better- catch up with the more tactical sword-and-board a little. Power Attack is one of those options.Where is my math wrong?Power Attack takes two actions. That leaves a third action to attack at -5.
It doesn't have to be better than an attack at -5, it has to be better than that third attack you're giving up, the one at -10. (Or something else you're doing with that action.) If you have to move up to them, you might want to hold off on Power Attack except for muscling through DR or something.Then what is the point of giving me more freedom of actions, if I'm going to spend all those actions making 'full' attacks anyway?
It seems counter intuitive to the new action system, especially when I've been playing RAE for 4 years.
Then Power Attack is useless, because it's always better to roll two attacks 75% of the time.
See where my problem is?
| 1 person marked this as a favorite. |
Maybe you're not going to spend all your actions making full attacks. But people who want to spend all their actions attacking should get options to make that better- catch up with the more tactical sword-and-board a little. Power Attack is one of those options.Where is my math wrong?Power Attack takes two actions. That leaves a third action to attack at -5.
It doesn't have to be better than an attack at -5, it has to be better than that third attack you're giving up, the one at -10. (Or something else you're doing with that action.) If you have to move up to them, you might want to hold off on Power Attack except for muscling through DR or something.Then what is the point of giving me more freedom of actions, if I'm going to spend all those actions making 'full' attacks anyway?
It seems counter intuitive to the new action system, especially when I've been playing RAE for 4 years.
Then Power Attack is useless, because it's always better to roll two attacks 75% of the time.
See where my problem is?
No, I don't? You get two attacks instead of three. If you don't want to "full attack" at least sometimes, then Power Attack may not be for you.
So I don't have the time to do the maths right now, but something to check would be if your second attack needs an 11 to hit making your first need a 6 to hit. This makes it so the second can't crit (or need a nat 20, still not clear). Does this change the math at all that the potential for the bigger crit on the first attack is worth more?
Or does applying the accuracy penalty and thus lower expected damage of the second swing cause the math to look better for the full attack? Like if the second hit needs a 6 to hit it's only getting 75% of it's damage to DPR right? Since it won't always hit. Maybe this is the math that makes power attack better.
The other thing to consider is that maybe it's only supposed to be good if you were "full attacking", so the power attacked swing is supposed to only be mathematically better than adding the -10 attack?
No, I don't? You get two attacks instead of three. If you don't want to "full attack" at least sometimes, then Power Attack may not be for you.Maybe you're not going to spend all your actions making full attacks. But people who want to spend all their actions attacking should get options to make that better- catch up with the more tactical sword-and-board a little. Power Attack is one of those options.Where is my math wrong?Power Attack takes two actions. That leaves a third action to attack at -5.
It doesn't have to be better than an attack at -5, it has to be better than that third attack you're giving up, the one at -10. (Or something else you're doing with that action.) If you have to move up to them, you might want to hold off on Power Attack except for muscling through DR or something.Then what is the point of giving me more freedom of actions, if I'm going to spend all those actions making 'full' attacks anyway?
It seems counter intuitive to the new action system, especially when I've been playing RAE for 4 years.
Then Power Attack is useless, because it's always better to roll two attacks 75% of the time.
See where my problem is?
Okay, but if Power Attack is only good when I make full attacks, then why not make it a 3 action feat that scales better so it arithmetically is worth the feat slot?
This seems like poor design.
Some more math, considering 'full attacks'
2(2d12+4) + d12+4 < 2(d12+4) + d12+4 + d12+4
Considering that the primary attack is supposed to reliably crit this means Power Attack is still worse. If Power Attack is guaranteed to crit, then my second attack is going to crit 75% of the time and my third attack is going to crit 50% of the time.
You make Power Attack worse not better by taking full attacks if the case is true where you're supposed to be able to bet on landing a crit, which again seems to be the design of the feat given Mark's statements about how it makes you better.
As presented the math doesn't check out, unless I'm doing something wrong.
In RAE this is why Vital Strike doesn't exist.
Your first attack only crits 100% of the time when your attack bonus is nine greater than their AC. That should be… really rare. It's a guaranteed hit on the last attack at a -10. Yeah, in those cases you should just attack extra times.
Your first attack only crits 100% of the time when your attack bonus is nine greater than their AC. That should be… really rare. It's a guaranteed hit on the last attack at a -10. Yeah, in those cases you should just attack extra times.
Okay, but the justification for power attack being what it is stems from the propensity of critical hits.
It can therefor be inferred that it is expected to happen at some point, especially since there does not seem to be anything that ever reduces your attack bonus and you get to increase your attack bonus more and more as you level, increase proficiency, and increase your ability scores.
Legendary proficiency and a legendary weapon stack, for a fighter that's level 13. Is a +6 STR too much to assume by then? Probably a low ball tbh.
Given just what they're told us, you could in theory expect to be swinging at +25.
Remember, 75% of the time that the d20 would make your primary hit a crit, the secondary will crit, and 50% of the time that same roll would make your tertiary attack hit.
So, 37.5% of the time Power Attack is guaranteed to be useless if the odds of critting increase on your primary attack.
Unless AC scales in such a way that your odds to hit/crit are always the same, then Power Attack mathematically needs to start by adding two dice. With this math, as the probability of the third attack critting approaches 1 Power Attack becomes more useless, which means we can reasonably expect Power Attack's scaling to kick in around level 10-11.
Because math.
New PA Attack List:
-0 Attack: 20% Miss, 50% Hit, 30% Crit = .5*(13)+.3*(26)
-5 Attack: 45% Miss, 45% Hit, 5% Crit = .45*(6.5)+.05*(13)
Net Damage: 17.9
That's only 95% all added together.
I'm guessing you meant the following:New PA Attack List:
-0 Attack: 20% Miss, 50% Hit, 30% Crit = .5*(13)+.3*(26)
-5 Attack: 45% Miss, 50% Hit, 5% Crit = .50*(6.5)+.05*(13)
Net Damage: 18.2
Yep, you're right. I realized I made a mistake in percentages somewhere and must have not put the 5% back into hit chance for the New Power Attack. Pesky math.
Remember, 75% of the time that the d20 would make your primary hit a crit, the secondary will crit, and 50% of the time that same roll would make your tertiary attack hit.
That's not correct, I'm afraid. Say you crit on a 16 with your primary attack (hit on a 6). Then you only crit on a 20 with your second and third attack. That's only 20% as often as your primary critting, and it's only because 20s always crit.
If you're basing your results on that, they're going to be substantially off. They only hold for cases where you have much higher attack than their AC.
Remember, 75% of the time that the d20 would make your primary hit a crit, the secondary will crit, and 50% of the time that same roll would make your tertiary attack hit.That's not correct, I'm afraid. Say you crit on a 16 with your primary attack (hit on a 6). Then you only crit on a 20 with your second and third attack. That's only 20% as often as your primary critting, and it's only because 20s always crit.
If you're basing your results on that, they're going to be substantially off. They only hold for cases where you have much higher attack than their AC.
Okay, but that doesn't make Power Attack better.
That makes Power Attack the same arithmetically as hitting them twice.
So Power Attack isn't actively worse, but it doesn't improve the feat.
My math holds.
| 1 person marked this as a favorite. |
Got bored, wrote Python code.
I can test arbitrary cases now, but here is how New PA compares to old PA for +3 damage and -1 attack, for a d12 weapon.
https://i.imgur.com/dQLAjeT.png
Plot for fun! Forgot labels in this version, but red is old PA and green is new PA.
| 2 people marked this as a favorite. |
Remember, 75% of the time that the d20 would make your primary hit a crit, the secondary will crit, and 50% of the time that same roll would make your tertiary attack hit.That's not correct, I'm afraid. Say you crit on a 16 with your primary attack (hit on a 6). Then you only crit on a 20 with your second and third attack. That's only 20% as often as your primary critting, and it's only because 20s always crit.
If you're basing your results on that, they're going to be substantially off. They only hold for cases where you have much higher attack than their AC.
Okay, but that doesn't make Power Attack better.
That makes Power Attack the same arithmetically as hitting them twice.
So Power Attack isn't actively worse, but it doesn't improve the feat.
My math holds.
All right, here we go. Primary attack hits on 6. That's a 75% chance to hit, 25% chance to crit. We'll use a d8 (if you need a hand free) and a d12 (if you want to focus on killing. We'll use average damage because that's useful and easy and I don't want to list out a probability distribution. We'll assume a strength mod of +4, so +4 damage on d8s, and +6 damage on d12s. We'll also try it out against some DR to see if certain situations improve it.
No DR:
Without Power Attack:
d8: 8.5 * (.75 + .25 + .5 + .05 + .25 + .05) = 15.725
d12: 12.5 * (.75 + .25 + .5 + .05 + .25 + .05) = 23.125
With Power Attack:
d8: 13 * (.75 + .25) + 8.5 * (.5 + .05) = 17.675
d12: 19 * (.75 + .25) + 12.5 * (.5 + .05) = 25.875
Move, and either two attacks or Power Attack:
Without Power Attack:
d8: 8.5 * (.75 + .25 + .5 + .05) = 13.175
d12: 12.5 * (.75 + .25 + .5 + .05) = 19.375
With Power Attack:
d8: 13 * (.75 + .25) = 13
d12: 19 * (.75 + .25) = 19
Against DR 5:
Without Power Attack:
d8: 3.5 * (.75 + .25 + .5 + .05 + .25 + .05) = 6.475
d12: 7.5 * (.75 + .25 + .5 + .05 + .25 + .05) = 13.875
With Power Attack:
d8: 8 * (.75 + .25) + 8.5 * (.5 + .05) = 12.675
d12: 14 * (.75 + .25) + 12.5 * (.5 + .05) = 20.875
Move, and either two attacks or Power Attack:
Without Power Attack:
d8: 3.5 * (.75 + .25 + .5 + .05) = 5.425
d12: 7.5 * (.75 + .25 + .5 + .05) = 11.625
With Power Attack:
d8: 8 * (.75 + .25) = 8
d12: 14 * (.75 + .25) = 14
It looks like Power Attack at low levels (assuming a generous hit-on-six) is a small improvement on full attacks with no DR, a miniscule loss over two attacks with no DR, a substantial improvement for full attacks against DR, and a small improvement over two attacks against DR.
Contributor, RPG Superstar 2010 Top 8
|
| 2 people marked this as a favorite. |
Okay, but if Power Attack is only good when I make full attacks, then why not make it a 3 action feat that scales better so it arithmetically is worth the feat slot?
Because, like Quid said, it's also good at the end of a move if you need to muscle through someone's DR.
Or, maybe you're expecting them to shield block, and you want to try to break their shield. A Power Attack is more likely to overcome the shield's hardness and actually put a dent in it.
Or maybe they're a high AC enemy with a nasty riposte reaction. Concentrating on one big hit after a move is better than making two attacks and risking a fumble and retribution.
So it makes you better at full attacks (focusing your damage into your high bonus attacks), while also having niche uses outside of full attacking. That sounds fine for a feat.
Okay, but if Power Attack is only good when I make full attacks, then why not make it a 3 action feat that scales better so it arithmetically is worth the feat slot?
Because, like Quid said, it's also good at the end of a move if you need to muscle through someone's DR.
Or, maybe you're expecting them to shield block, and you want to try to break their shield. A Power Attack is more likely to overcome the shield's hardness and actually put a dent in it.
Or maybe they're a high AC enemy with a nasty riposte reaction. Concentrating on one big hit after a move is better than making two attacks and risking a fumble and retribution.
So it makes you better at full attacks (focusing your damage into your high bonus attacks), while also having niche uses outside of full attacking. That sounds fine for a feat.
And this gives me some substance to consider, thank you.
It does seem the shield mechanics are going to show up more often and that we will see much more of them in game since they've been showcased so much in the small amount of reveals we have.
All right, here we go. Primary attack hits on 6. That's a 75% chance to hit, 25% chance to crit. We'll use a d8 (if you need a hand free) and a d12 (if you want to focus on killing. We'll use average damage because that's useful and easy and I don't want to list out a probability distribution. We'll assume a strength mod of +4, so +4 damage on d8s, and +6 damage on d12s. We'll also try it out against some DR to see if certain situations improve it.
This is perfect for comparison's sake.
Now that I can see the maths it is easier to understand the design space. Given the damage calcs as they are I might conject that it more or less aligns with our findings on Vital Strike in PF1RAE. The relationship with DR is substantial, and seeing the numbers now and combining that with our understanding of shields it seems likely that we will be seeing more and more abilities to actively grant DR, both on enemies and PCs. That would make Power Attack an actively better feat, even if the numbers it puts out are comparatively similar in execution. I now understand why it's worth the feat. Tactically, it is more an option to attack an enemy with DR, rather than an always on option.
Now if we increase in level and the propensity of the secondary attack's probability to crit reaches 1, we then and there have a mathematically viable place to expect the increase in dice.
This answers a lot of questions I had.
Mark confirmed that Legendary weapons stack with Legendary proficiency.Want to go to this point:Your first attack only crits 100% of the time when your attack bonus is nine greater than their AC. That should be… really rare. It's a guaranteed hit on the last attack at a -10. Yeah, in those cases you should just attack extra times.Okay, but the justification for power attack being what it is stems from the propensity of critical hits.
It can therefor be inferred that it is expected to happen at some point, especially since there does not seem to be anything that ever reduces your attack bonus and you get to increase your attack bonus more and more as you level, increase proficiency, and increase your ability scores.
Legendary proficiency and a legendary weapon stack, for a fighter that's level 13. Is a +6 STR too much to assume by then? Probably a low ball tbh.
Given just what they're told us, you could in theory expect to be swinging at +25.
Remember, 75% of the time that the d20 would make your primary hit a crit, the secondary will crit, and 50% of the time that same roll would make your tertiary attack hit.
So, 37.5% of the time Power Attack is guaranteed to be useless if the odds of critting increase on your primary attack.
Unless AC scales in such a way that your odds to hit/crit are always the same, then Power Attack mathematically needs to start by adding two dice. With this math, as the probability of the third attack critting approaches 1 Power Attack becomes more useless, which means we can reasonably expect Power Attack's scaling to kick in around level 10-11.
Because math.
Legendary proficiency and a legendary weapon stackThis applies to armour as well, legendary armour proficiency and legendary armour should also stack. Add that to the armour's own value, plus the standard level and dex bonuses…
Let's see, at level 13, AC is 10 + level(13) + dex bonus(lets say +3) + proficiency rank(let's say +2) + armour rank(let's say +2) + armour value. That's 30 before taking in...
I think that quote was from one of the other designers. However, you can absolutely use a high-quality item and be awesome yourself and stack them.Mark Seifter said something to the effect of "just wait until you see what one can do with proficiency in armor" so I think it's even more dramatic (and a legendary heavy armor user will also get +6 AC).So, how does proficiency affect things other than skills?
Is it tied to weapon groups? Can I be legendary with swords and use a legendary weapon to get level +6 to hit?
Against DR 5:
"Full Attack":
Without Power Attack:
d8: 3.5 * (.75 + .25 + .5 + .05 + .25 + .05) = 6.475
d12: 7.5 * (.75 + .25 + .5 + .05 + .25 + .05) = 13.875
With Power Attack:
d8: 8 * (.75 + .25) + 8.5 * (.5 + .05) = 12.675
d12: 14 * (.75 + .25) + 12.5 * (.5 + .05) = 20.875Move, and either two attacks or Power Attack:
Without Power Attack:
d8: 3.5 * (.75 + .25 + .5 + .05) = 5.425
d12: 7.5 * (.75 + .25 + .5 + .05) = 11.625With Power Attack:
d8: 8 * (.75 + .25) = 8
d12: 14 * (.75 + .25) = 14
Subtracting the DR directly from the average damage is going to throw off your numbers for crits. It should be something like this:
"Full Attack":Without Power Attack:
d8: 3.5 * (.75 + .50 + .25) + 8.5 * (.25 + .05 + .05) = 8.225
d12: 7.5 * (.75 + .50 + .25) + 12.5 * (.25 + .05 + .05) = 15.625
With Power Attack:
d8: (8 * .75 + 3.5 * .50) + (13 * .25 + 8.5 * .05) = 11.425
d12: (14 * .75 + 7.5 * .50) + (19 * .25 + 12.5 * .05) = 19.625
Move, and either two attacks or Power Attack:
Without Power Attack:
d8: 3.5 * (.75 + .50) + 8.5 * (.25 + .05) = 6.925
d12: 7.5 * (.75 + .50) + 12.5 * (.25 + .05) = 13.125
With Power Attack:
d8: (8 * .75 + 13 * .25) = 9.25
d12: (14 * .75 + 19 * .25) = 15.25
Oh, you’re right! Thanks, that was a silly mistake.
Average on a d12 is 6.5, might skew the math.
Average on a d12 is 6.5, might skew the math.
Damage values were based on the average plus the static bonus from strength.
power attack at the normal -1/+2 ratio was really not that great of a trade and only worth it when you had lots of extra accuracy.
Actually, for characters with mainly bonuses that apply equally to attack and damage rolls (strength, weapon training, inspire courage, etc.), the ratio of equilibrium was something like 1:1.5 to 1.8 (i.e. +1 to attack rolls increases the average damage the same amount as a 1.5 or 1.8 to damage rolls). One-sided bonuses shift that ratio, obviously.
So it was generally useless for TWF, OK for one-handed weapon, and great for two-handed weapon.Considering Power Attack was a good baseline for how powerful a feat needs to be to not make martials completely irrelevant, it didn't need 'fixing' it and all other martial feats needed buffing
If martials need certain feats to be non-useless, the game is broken. Mandatory feats always stifle variety.
power attack at the normal -1/+2 ratio was really not that great of a trade and only worth it when you had lots of extra accuracy.Actually, for characters with mainly bonuses that apply equally to attack and damage rolls (strength, weapon training, inspire courage, etc.), the ratio of equilibrium was something like 1:1.5 to 1.8 (i.e. +1 to attack rolls increases the average damage the same amount as a 1.5 or 1.8 to damage rolls). One-sided bonuses shift that ratio, obviously.
So it was generally useless for TWF, OK for one-handed weapon, and great for two-handed weapon.Considering Power Attack was a good baseline for how powerful a feat needs to be to not make martials completely irrelevant, it didn't need 'fixing' it and all other martial feats needed buffingIf martials need certain feats to be non-useless, the game is broken. Mandatory feats always stifle variety.
My experience in RAE showed that Power Attack was only really optimal on two handed builds, because TWF was actually viable.
|
| 1 person marked this as a favorite. |
Continuing a bit from the other thread.
[Power Attack]= 2dx+ damage/ dx+ damage = 3dx+ 2damage
[Critical hit Power Attack]= 4dx + 2damage/ dx+ damage = 5dx+ 3damage
[no Power Attack]= dx+ damage/ dx+ damage/ dx+ damage = 3dx+ 3damage
[no Power Attack critical hit]= 2dx+ 2damage/ dx+ damage/ +dx +damage = 4dx+ 4damage[comparison] If 5dx+ 3damage > 4dx +4damage, then Power Attack is good
If 5dx+ 3damage < 4dx+ 4damage, then Power Attack is bad.
I'm interested in how that comparison changes with different variables. QuidEst did an analysis for primary attacks that hit on 6. How does the math change when you hit on 11? 16?
If your third attack can only hit on a twenty (and can't crit), is it really worth taking over bonus damage on your primary weapon?
I put QuidEst's example into excel to find out more.
Hit on 13, d12+6
Power Attack:
Two Actions 12.5 * (0.4 + 0.05) + 6.5 * (0.4) = 8.255
Three Actions 12.5 * (0.4 + 0.05) + 6.5 * (0.4) + 12.5 * (0.15 + 0.05) = 10.725
No Power Attack:
Two Actions 12.5 * (0.4 + 0.05) + 12.5 * (0.15 + 0.05) = 8.125
Three Actions 12.5 * (0.4 + 0.05) + 12.5 * (0.15 + 0.05) + 12.5 * (0.05) = 8.75
I found an inflection point when you need a 13 to hit on your first attack. Against harder to hit enemies, you'll want to Power Attack even if that is your only attack. Between 3 and 13, you want to take your second attack instead.
If you are rolling d12+4 instead, the inflection point moves down to hit on AC 11. If a monster takes more than an 11 to hit, Power Attack is better.
Continuing a bit from the other thread.
[Power Attack]= 2dx+ damage/ dx+ damage = 3dx+ 2damage
[Critical hit Power Attack]= 4dx + 2damage/ dx+ damage = 5dx+ 3damage
[no Power Attack]= dx+ damage/ dx+ damage/ dx+ damage = 3dx+ 3damage
[no Power Attack critical hit]= 2dx+ 2damage/ dx+ damage/ +dx +damage = 4dx+ 4damage[comparison] If 5dx+ 3damage > 4dx +4damage, then Power Attack is good
If 5dx+ 3damage < 4dx+ 4damage, then Power Attack is bad.
I'm interested in how that comparison changes with different variables. QuidEst did an analysis for primary attacks that hit on 6. How does the math change when you hit on 11? 16?
If your third attack can only hit on a twenty (and can't crit), is it really worth taking over bonus damage on your primary weapon?
I put QuidEst's example into excel to find out more.
Hit on 13, d12+6
Power Attack:
Two Actions 12.5 * (0.4 + 0.05) + 6.5 * (0.4) = 8.255
Three Actions 12.5 * (0.4 + 0.05) + 6.5 * (0.4) + 12.5 * (0.15 + 0.05) = 10.725No Power Attack:
Two Actions 12.5 * (0.4 + 0.05) + 12.5 * (0.15 + 0.05) = 8.125
Three Actions 12.5 * (0.4 + 0.05) + 12.5 * (0.15 + 0.05) + 12.5 * (0.05) = 8.75I found an inflection point when you need a 13 to hit on your first attack. Against harder to hit enemies, you'll want to Power Attack even if that is your only attack. Between 3 and 13, you want to take your second attack instead.
If you are rolling d12+4 instead, the inflection point moves down to hit on AC 11. If a monster takes more than an 11 to hit, Power Attack is better.
Notes:
20's auto hit and auto crit, so 5% of the time, Power Attack is mathematically pointless.The existence of Certain Strike guarantees a hit with flat damage and all variables treated as 1's. This affects Power Attack's usefulness immensely as before the argument was that it had value because that third attack was so unreliable. Now we have to find out how reliable Power Attack's additional damage is to be useful.
To do this, we need to start setting up and solving some algebra:
f(full attacks with Power Attack)= 5dx + 3damage
f(full attacks without Power Attack)= 4dx+ 4damage
Now, Certain Strike comes into play and breaks down the full formula into something like this:
f(Power Attack)= 2dx + damage/ dx + damage
f(Certain Strike)= dx+ damage/ dx + damage/ 1 + damage
so we can simplify this to be something more like
3dx + 2 damage ~ 2dx + 3 damage + 1
Reducing common terms means that if (dx > 1 + damage) then Power Attack is good.
If (dx < 1 + damage) then Power Attack is almost always a worse choice than Certain Strike.
We can look at a 3rd level fighter with STR 18 as a start, since we can feasibly imagine this to be a real character. We do not know about the benefits of two-handing a weapon, if there even are any in this edition. 3rd level fighters improve their proficiency to master, which means they get +2 to hit and damage, not including weapon qualities.
So we swing on a +9/+4/-1 ([+3]3rd level, [+4]18 STR, [+2]master proficiency)
Longswords seem pretty standard as far as expected loot, so let's go ahead with a moderate d8.
damage now looks like this:
f(Power Attack)= (2d8+6) + (d8+6) = [3d8+12]
f(Certain Strike)= (d8+6) + (d8+6) + 7 = [2d8+19]
Reduce common terms and solve the system:
d8 = 7
If using a longsword with Power Attack at 3rd level, you need to always roll a 7 or better on the extra die for it to be worth trading that attack.
The only damage dice that seem to be worth it are the ones on the higher end of the damage spectrum. Even then, average damage on a d12 is 6.5, and on 2d6 it's 7, making the options equal. Except Certain Strike is listed as a fighter ability, it may not even be a feat.
Based on this data, we can reasonably conclude that Power Attack will need to change during the play test.
20's auto hit and auto crit, so 5% of the time, Power Attack is mathematically pointless.
What? No, it's not mathematically pointless. It's still doing stuff, like giving you a larger base to crit on with the attack most likely to crit. Yes, you're passing up a 5% chance to crit on your last attack, but only in situations where you have at least a 50/50 chance of hitting with your main attack. (If your 20 would miss, it's a regular hit instead.)
The existence of Certain Strike guarantees a hit with flat damage and all variables treated as 1's. This affects Power Attack's usefulness immensely as before the argument was that it had value because that third attack was so unreliable. Now we have to find out how reliable Power Attack's additional damage is to be useful.
Power Attack is available at first level, while Certain Strike is available at double-digit levels. Certain Strike seems really good at low levels, because it's really strong in situations where your static makes up at least half your expected damage.
math involving a d8
I probably wouldn't bother grabbing Power Attack for less than a d10. If I had Certain Strike, I'd probably be focusing on smaller weapons with useful properties, like a scimitar (which does additional damage when used against a single target repeatedly). They're for different builds.
math ignoring probability and assuming regular hits/misses
I mean, I get why you're doing it. But you can't then say we can reasonably conclude Power Attack needs to change. We can reasonably conclude that if the math shortcuts iron out and bonus damage is that high, that Certain Strike would be a better feat that Power Attack at third level if you're using a d8 weapon. But, Certain Strike isn't available at third level and using a d8 weapon doesn't synergize well with Power Attack.
20's auto hit and auto crit, so 5% of the time, Power Attack is mathematically pointless.What? No, it's not mathematically pointless. It's still doing stuff, like giving you a larger base to crit on with the attack most likely to crit. Yes, you're passing up a 5% chance to crit on your last attack, but only in situations where you have at least a 50/50 chance of hitting with your main attack. (If your 20 would miss, it's a regular hit instead.)
The existence of Certain Strike guarantees a hit with flat damage and all variables treated as 1's. This affects Power Attack's usefulness immensely as before the argument was that it had value because that third attack was so unreliable. Now we have to find out how reliable Power Attack's additional damage is to be useful.Power Attack is available at first level, while Certain Strike is available at double-digit levels. Certain Strike seems really good at low levels, because it's really strong in situations where your static makes up at least half your expected damage.
math involving a d8I probably wouldn't bother grabbing Power Attack for less than a d10. If I had Certain Strike, I'd probably be focusing on smaller weapons with useful properties, like a scimitar (which does additional damage when used against a single target repeatedly). They're for different builds.
math ignoring probability and assuming regular hits/missesI mean, I get why you're doing it. But you can't then say we can reasonably conclude Power Attack needs to change. We can reasonably conclude that if the math shortcuts iron out and bonus damage is that high, that Certain Strike would be a better feat that Power Attack at third level if you're using a d8 weapon. But, Certain Strike isn't available at third level and using a d8 weapon doesn't synergize well with Power Attack.
I supposed I phrased my statement incorrectly to assert the message I meant.
We know that dx+6 is almost assured to be the minimum expectation for a fighter of at least 3rd level. Any point after that your flat numbers just increase, either by your STR increasing, proficiency, weapon quality, or all of these and potentially more. This starts at level 3.
This means we already know the extent to which Power Attack is useful, even though we don't have the other parts of the equation. Because we know this, we can reasonably predict where the rest of the math of the system has to come from otherwise the game would break.
I'm not sure where you got the info for Certain Strike, in the blog post proper there is no mention of what level fighters get it, so I'm assuming there must have been a post by a developer that spoiled it, unless you're making an assumption to disprove my math.
My math is written out in the same way I would expect to write it on a character sheet, but to be fair when we start calculating in critical hits, we end up with the difference being [2dx = 1+damage], which might matter, but again only when comparing 3 attacks to power attack plus one other attack. In the cases where you're only making two attacks compared to a single power attack, the existence of Certain Strike alone negates how useful Power Attack is.
You pointing out the issue with a d8 weapon not having synergy with Power Attack means I did my math right in my book, we now understand that, with math. Good stuff.
I'm not sure where you got the info for Certain Strike, in the blog post proper there is no mention of what level fighters get it, so I'm assuming there must have been a post by a developer that spoiled it, unless you're making an assumption to disprove my math.
Developer post from Mark Seifter. I think you've uncovered not "why Power Attack needs to be fixed", but "why Certain Strike isn't available at low levels".
My math is written out in the same way I would expect to write it on a character sheet, but to be fair when we start calculating in critical hits, we end up with the difference being [2dx = 1+damage], which might matter, but again only when comparing 3 attacks to power attack plus one other attack. In the cases where you're only making two attacks compared to a single power attack, the existence of Certain Strike alone negates how useful Power Attack is.
Certain Strike also does no damage if you roll a 1 and that would miss, or if you miss by 10 or more.
You pointing out the issue with a d8 weapon not having synergy with Power Attack means I did my math right in my book, we now understand that, with math. Good stuff.
No? I'm saying it's not a very fair example to pick. Unless you think that Power Attack needs to work well for mid-range weapons to be worthwhile for inclusion, in which case, we still haven't looked at the math, and I disagree with your premise. I'm of the opinion that if Power Attack works well for large weapons, then it's got a valid role.
Anyway, stuff getting tossed around without actual math was bugging me.
Kludgy calculator to compare Power Attack with regular attacks.
If I wanted to compare a d12 at with 1 base die and 1 power attack die with 4 static damage, I'd enter:
12
1
1
4
If I wanted to compare a d8 at 3 base dice and 2 power attack dice with 8 static damage, I'd enter:
8
3
2
8
I'm not bothering with Certain Strike, because that's a double-digits level feat with a different focus.
You guys really need to wait till we get more info in my opinion.
| 1 person marked this as a favorite. |
You guys really need to wait till we get more info in my opinion.
That's fair.
You got your PDF so soon?
|
|
| 1 person marked this as a favorite. |
Notes:
20's auto hit and auto crit, so 5% of the time, Power Attack is mathematically pointless.
20s only crit if the attack would be a hit. If your third attack can't beat the AC of the monster, you don't crit. That's why Power Attack is better against high AC enemies.
The existence of Certain Strike guarantees a hit with flat damage and all variables treated as 1's.
Certain Strike does not guarantee a hit. It only applies in the Failure condition, not the Critical Failure condition. So against a high AC enemy, your third attack will have <50% chance of doing any damage.
Again, high AC enemies are where Power Attack becomes valuable.
Notes:
20's auto hit and auto crit, so 5% of the time, Power Attack is mathematically pointless.
20s only crit if the attack would be a hit. If your third attack can't beat the AC of the monster, you don't crit. That's why Power Attack is better against high AC enemies.
The existence of Certain Strike guarantees a hit with flat damage and all variables treated as 1's.Certain Strike does not guarantee a hit. It only applies in the Failure condition, not the Critical Failure condition. So against a high AC enemy, your third attack will have <50% chance of doing any damage.
Again, high AC enemies are where Power Attack becomes valuable.
Considering the blog specifically calls out that there is no critical failure result on attack rolls, I'd say we're looking at some bad communication on part of the developers then, because in the blog on critical success/failure it says that if there's nothing listed for failure, then nothing happens on a critical failure either.
Upthread we determined that what you are saying is basically true, in that Power Attack becomes more useful when the -5 from your second attack becomes detrimental to your ability to hit. Given the developers feeding us the tidbits that they intend for crits to happen much mroe often as an assumed part of the game engine, this is most likely not going to be the case, but I'll save the analysis on critical hits when we get more information on AC.
That's kinda why I'm avoiding using crits in my math. Still, bigger weapons are necessary to make the feat work, and you still have to be making essentially full attacks. You can use sudden charge, but then your Power Attack is at -5 instead of full so the whole design around not taking a penalty to hit really isn't going to have much synergy with that tactic, which I also imagine will come up often as a complaint in the play test.
|
You can use sudden charge, but then your Power Attack is at -5 instead of full so the whole design around not taking a penalty to hit really isn't going to have much synergy with that tactic, which I also imagine will come up often as a complaint in the play test.
Wait, what? This sentence makes no sense with the actual rules as I understand them. There are two possible scenarios for Sudden Charge + Power Attack and neither of them result in this.
#1: You can't do this, since both of these things cost two actions and you only have 3.
#2: You can do this (presumably due to wording on Power Attack being that it costs 1 Action extra...I find this unlikely), in which case your one Power Attack would be at no penalty.
I guess you could add in the theoretical scenario of somehow having 4 actions (Haste maybe?), which might result in what you describe, but that's super speculative and it would strongly depend on how both Class Feats are worded how that would work.
|
|
Notes:
20's auto hit and auto crit, so 5% of the time, Power Attack is mathematically pointless.
20s only crit if the attack would be a hit. If your third attack can't beat the AC of the monster, you don't crit. That's why Power Attack is better against high AC enemies.
The existence of Certain Strike guarantees a hit with flat damage and all variables treated as 1's.Certain Strike does not guarantee a hit. It only applies in the Failure condition, not the Critical Failure condition. So against a high AC enemy, your third attack will have <50% chance of doing any damage.
Again, high AC enemies are where Power Attack becomes valuable.
Considering the blog specifically calls out that there is no critical failure result on attack rolls, I'd say we're looking at some bad communication on part of the developers then, because in the blog on critical success/failure it says that if there's nothing listed for failure, then nothing happens on a critical failure either.
Upthread we determined that what you are saying is basically true, in that Power Attack becomes more useful when the -5 from your second attack becomes detrimental to your ability to hit. Given the developers feeding us the tidbits that they intend for crits to happen much mroe often as an assumed part of the game engine, this is most likely not going to be the case, but I'll save the analysis on critical hits when we get more information on AC.
That's kinda why I'm avoiding using crits in my math. Still, bigger weapons are necessary to make the feat work, and you still have to be making essentially full attacks. You can use sudden charge, but then your Power Attack is at -5 instead of full so the whole design around not taking a penalty to hit really isn't going to have much synergy with that tactic, which I also imagine will come up often as a complaint in the play test.
The Strike action doesn’t have a crit fail, but the Certain Strike action does. Check out Mark’s comments about it.
There are many things we don’t know. Doesn’t mean we have to ignore those things. That is why I wanted analysis on the whole range. We can find the inflection points. When does it make sense to Power Attack? Then, when we have a play test bestiary, we will have a better idea of its value.
| 7 people marked this as a favorite. |
I think Power Attack needs to be worth using at all times if my character chooses to take it.
Why the heck would "needs to be worth using at all times" be a desirable situation to be in? A game where there are multiple types of situations that require different actions from the Fighter sounds a heck of a lot more fun than one where you just mash the Power Attack button all day long.