|
10 people marked this as a favorite.
|
Hello everyone.
As my day job is as a computer simulation modeler, when I started reading the PF forums here, I noted a lot of concern about the treadmill, proficiency, the level bonuses, and other things. And I thought, 'Hm. I am going to model some things here.'
So right now, my first foray is to examine the damage per minute (DPM) numbers, by level, of the Fighter and the Barbarian. As always when building a model, it is useful to state your assumptions. These are mine, so far:
1) The only feat selected for the Fighter and Barbarian is Sudden Charge.
2) It is assumed that the Fighter and Barbarian will need to use Sudden Charge on the 1st combat round of the minute, and not otherwise.
3) The Fighter and Barbarian each use Strike as often as possible. This means that the Fighter gets more Strikes overall, due to the Barbarian needed to keep Rage up. The Barbarian does Strike while Fatigued.
4) Both characters are using a d12 weapon, neither of which are Agile.
5) No other weapon properties (such as Sweep or Forceful) are modeled.
6) Both character start with an 18 Strength, increasing to 20 at level 10, 22 at level 13, and 24 at level 20.
7) Ability (Str), Level, Proficiency, and expected Item modifiers are included.
8) Due to monster stats in the playtest so far, it is assumed that the Fighter will need an 11 to hit the opponent. The Barbarian will need a higher number, per the lower proficiency bonus.
9) The Barbarian's totem does not increase the damage beyond normal (e.g., Dragon, Giant).
The design:
The model runs for one minute, and simulates the dice rolling for that minute (i.e., 10 combat rounds) for both Fighter and Barbarian. I then ran that model 10,000 times
The results:
At levels 1 and 2, the Barbarian's average damage is slightly (~2%) higher the Fighter. At level 3, the Fighter does on average ~5% more damage than the Barbarian. After that, at levels where the Barbarian gets a bonus Damage increase (7th, 11th, etc.), the Fighter does on average ~10% more damage than the Barbarian. At other levels, the Fighter deals on average ~13% more damage than the Barbarian.
There is a lot of variability in these numbers, as to be expected from using d12s and a d20. I am currently running a regression, but I am confident that the damage difference is statistically significant. I will report on this when I can.
Another point I would like to make is that the coefficient of variation (CoV) of the attacks rolls over all levels does not change significantly from the CoV of a d20 roll. This would seem to quantitatively validate the comments on these forums regarding the feel of play being dependent on the results of the d20.
Good job, thank you for the hard work
However dpm is a poor measure of effectiveness imho but it' s still good to know
Pathfinder Rulebook Subscriber
I assume you were using flat rage. Out of curiousity can you run with the modifiers for the different totems, such as -1 hit but double rage damage on giant totem or +1 damage from dragon?
I can see lots of Dragon Totem Barbarians now. Dragon Totem gives you more conditional damage, and you chose what type when you create your character.
|
3 people marked this as a favorite.
|
Dark Sephiroth: I disagree. With the action economy as it is, modeling anything less than a minute would not obtain a good overall sense of the damage-dealing capabilities of each class.
Basically, the above could be used to argue that Rage should be a Free Action, or perhaps a Reaction with trigger of 'start of turn, not fatigued or raging', is better for the Barbarian. The Barbarian has worse defenses than a Fighter, and far less ability to make use of follow-up attacks. As such, it should theoretically do more damage than a Fighter.
However, this is not the case at and after 3rd level. The Fighter has more damage than a Barbarian at all levels after 2nd. Why, then, play a Barbarian? I could accept 'same-ish damage, but more variable'. But losing 13% average damage over a minute? Not too useful.
Also note that the Fighter is assumed never to make an AoO the entire minute, which would further bolster their damage.
Rayous: I can certainly rerun the numbers. I don't expect much to change. I certainly feel that the Giant totem is a huge (NPI) trap under the current model. But I will do so and report back.
Edit: I have run the numbers for the Dragon totem, and it does indeed get quite a bit closer to the Fighter. E.g., at level 7 (a Barbarian damage-increase level), the Fighter does 144 damage per minute on average, while the Barbarian does 138. Where, at level 10, the Fighter does 208 damage per minute on average, while the Barbarian does 185. It does not really help that the Barbarian damage increases always occur the level before the next potency rune is expected to show up.
My next iteration is likely to add a Longbow Ranger shooting a Hunted Target.
I think a few of the numbers involved are a bit off, unless some form of magic item was accounting for the Strength increases. By 10th level, the numbers sound close to accurate; but the maximum strength value for both classes without some form of magical assistance caps out at 22 by 20th level.
Now the bigger question with this simulation is how successful enemy strikes against both affect their overall damage output. For the purposes of your test, we assume Fury Totem on the Barbarian for the sake of it having no inherent means of adding extra Rage damage. If you don't want to even calculate the effect of resistance, substitute with Superstition Totem. We know that while raging, the Barbarian will be hit more frequently. The question then becomes, how much does damage resistance affect damage output?
|
1 person marked this as a favorite.
|
Nothing guarantees that the fight will even last more than 3 rounds tho. Also, when the barb is raging and thus dealing more damage, his better dpr might give him one kill above the Fighter thus helping the party relative action eceonomy vs enemeies. This is impossible to express in linear numbers but nonetheless statistically significant. Lastly, there is no reason to assume that always attacking would be the optimal choice; the barb might help in other ways while not raging, optimizing his own action economy (only attack when it's most effective, do something else when it's not)
Dpm does not equal effectiveness. It's useful for crunching numbers but in real gameplay it hardly applies.
Kalvit: I believe at 13th level is when the ability-increasing items come online, and if you already have an 18+ in the stat, just gives you a +2 bonus, no? If I am incorrect, please let me know, so I can adjust.
Looking it up, I see the items show up at 14th level, not 13th. My bad. I will rework the model to reflect that.
Also, I am not simulating incoming damage on the Fighter or Barbarian. If I did, the Fighter would become even better. The Barbarian both lacks access to heavy armor and takes a flat -1 penalty while raging. For the purposes of the model, the Barbarian Strikes while Fatigued from Raging, and thus would be massively defensively weakened. While it is true that the Barbarian gets temporary hit points and also damage resistance while raging at higher levels, I am not sure that it at all compensates for being hit more often. I can make a model of that as well, if you like.
I think D@rK-SePHiRoTH-'s point is very relevant in that a 1 minute model isn't generally going to create a realistic model of actual gameplay. For example, if I assume 10 consecutive rounds of combat on a standard rage-cycling rotation then the Barbarian is getting 3 activation rounds, 2 fatigue rounds, and only 1 full round of effectiveness on the final activation instead of his usual 2.
If you model it on standard encounter design where a typical encounter lasts for 3 rounds and your 1/day "boss fight" might go up to 5 it changes things. Assuming a total of 3 combat encounters for the day, two lasting for 3 rounds and 1 lasting for 5, the Barbarian's total effectiveness for the day is 4 activation rounds where he's an action behind the fighter due to initiating rage, 6 full rage rounds, and only 1 fatigue round.
So you've got (R=rage activation round, B=full rage-buffed combat round, F= fatigue round)
1-Minute Model
RBBFRBBFRB
Compared to
"Realistic" Model
RBBRBBRBBFR
The realistic model adds 1 round to tracking and adds 1 additional activation round, but drops a fatigue round and gives the barbarian one more fully buffed round, which could have a significant impact on the overall averages. And that's before accounting for variance in action economy and player choices (like maybe not raging on that last round because they know the BBEG must be getting close to death and instead swinging out with all of their available actions, or using totem abilities that increase their damage output, etc.)
|
5 people marked this as a favorite.
|
D@rK-SePHiRoTH- wrote: Nothing guarantees that the fight will even last more than 3 rounds tho. Also, when the barb is raging and thus dealing more damage, his better dpr might give him one kill above the Fighter thus helping the party relative action eceonomy vs enemeies. This is impossible to express in linear numbers but nonetheless statistically significant. Lastly, there is no reason to assume that always attacking would be the optimal choice; the barb might help in other ways while not raging, optimizing his own action economy (only attack when it's most effective, do something else when it's not)
Dpm does not equal effectiveness. It's useful but in real gameplay it hardly applies.
The mistake you are making is believing that the Barbarian does more damage while raging. He _does not_, at higher levels. He does more damage than the Fighter _if he hits_. At every level after 2nd, the Fighter has a +2 to hit relative to the Barbarian, and that bonus is very keenly felt.
If you like, I can build a model that expresses damage per round, or damage dealt per rage cycle, but it will not end in any way other than to say that, against the same creature, the fighter puts out more damage than the Barbarian after level 2. (Maybe after level 3.) Essentially, the Fighter's edge in hits and crits vis-a-vis the Barbarian completely overwhelms the Barbarian's extra damage. Perhaps the Barbarian should receive more bonus damage to compensate.
Also note that the target number to roll for the Fighter was chosen to be 11. That means that both Fighter and Barbarian crit only on a 20. If I had selected a 9, the Fighter would crit twice as often as the Barbarian, barring Keen weapons. That is, the Fighter is better against lower-level foes because he crits more than the Barbarian, and better against higher-level foes because he hits more than the Barbarian. So... yeah.
NemisCassander wrote: D@rK-SePHiRoTH- wrote: Nothing guarantees that the fight will even last more than 3 rounds tho. Also, when the barb is raging and thus dealing more damage, his better dpr might give him one kill above the Fighter thus helping the party relative action eceonomy vs enemeies. This is impossible to express in linear numbers but nonetheless statistically significant. Lastly, there is no reason to assume that always attacking would be the optimal choice; the barb might help in other ways while not raging, optimizing his own action economy (only attack when it's most effective, do something else when it's not)
Dpm does not equal effectiveness. It's useful but in real gameplay it hardly applies.
The mistake you are making is believing that the Barbarian does more damage while raging. He _does not_, at higher levels. He does more damage than the Fighter _if he hits_. At every level after 2nd, the Fighter has a +2 to hit relative to the Barbarian, and that bonus is very keenly felt.
If you like, I can build a model that expresses damage per round, or damage dealt per rage cycle, but it will not end in any way other than to say that, against the same creature, the fighter puts out more damage than the Barbarian after level 2. (Maybe after level 3.) Essentially, the Fighter's edge in hits and crits vis-a-vis the Barbarian completely overwhelms the Barbarian's extra damage. Perhaps the Barbarian should receive more bonus damage to compensate.
Also note that the target number to roll for the Fighter was chosen to be 11. That means that both Fighter and Barbarian crit only on a 20. If I had selected a 9, the Fighter would crit twice as often as the Barbarian, barring Keen weapons. That is, the Fighter is better against lower-level foes because he crits more than the Barbarian, and better against higher-level foes because he hits more than the Barbarian. So... yeah. Thank you for clarifying this point;
If the barb's dpr is not better while raging then my point is moot. I get this much.
Do you believe that the other barbarian's features compensate for the loss of damage?
Ssalarn wrote: I think D@rK-SePHiRoTH-'s point is very relevant in that a 1 minute model isn't generally going to create a realistic model of actual gameplay. For example, if I assume 10 consecutive rounds of combat on a standard rage-cycling rotation then the Barbarian is getting 3 activation rounds, 2 fatigue rounds, and only 1 full round of effectiveness on the final activation instead of his usual 2. Very well. I will shift to discussing damage per rage cycle (including the fatigue round), rather than damage per minute, if you like.
Quote: If you model it on standard encounter design where a typical encounter lasts for 3 rounds Well, frankly, so far, the encounters in the playtest for my group (played one session so far) have been over in less than 3 rounds...
Quote: and your 1/day "boss fight" might go up to 5 it changes things. Assuming a total of 3 combat encounters for the day, two lasting for 3 rounds and 1 lasting for 5, the Barbarian's total effectiveness for the day is 4 activation rounds where he's an action behind the fighter due to initiating rage, 6 full rage rounds, and only 1 fatigue round. Again, thanks for the feedback. I will switch it up and report back.
Quote: So you've got (R=rage activation round, B=full rage-buffed combat round, F= fatigue round)
1-Minute Model
RBBFRBBFRB
Compared to
"Realistic" Model
RBBRBBRBBFR
Should this not be 'RBBFRBBFRBBF'? Just checking...
Quote: The realistic model adds 1 round to tracking and adds 1 additional activation round, but drops a fatigue round and gives the barbarian one more fully buffed round, which could have a significant impact on the overall averages. Not nearly as much as you might think, given the other issues the Barbarian has. I did 1-Round modeling before, and the Fighter still beats the Barbarian at anything other than the lowest levels.
Quote: And that's before accounting for variance in action economy and player choices (like maybe not raging on that last round because they know the BBEG must be getting close to death and instead swinging out with all of their available actions.) I fully admit that I am not trying to account for action-economy variance to reduce the complexity of the model. That way lies madness without a lot more investment of time...
NemisCassander wrote: The mistake you are making is believing that the Barbarian does more damage while raging. He _does not_, at higher levels. He does more damage than the Fighter _if he hits_. At every level after 2nd, the Fighter has a +2 to hit relative to the Barbarian, and that bonus is very keenly felt. Did you do math on this other than that mentioned in the first post?
Because, as others note, the math mentioned in the first post seems predicated on assumptions that are mostly not true in actual play (actually having to deal with the round of fatigue is pretty rare, IME, and having to re-rage again in the same fight even rarer).
And that's a pretty definitive statement to make without having done the math. Which is not to say you're wrong (I suspect that you're right, actually, just from eyeballing it), but this thread was supposed to be about mathematical analysis rather than just eyeballing things, wasn't it?
EDIT: Ninja'd. One round analysis would absolutely do the trick, yeah. What's the percentage difference there?
|
2 people marked this as a favorite.
|
NemisCassander wrote:
Very well. I will shift to discussing damage per rage cycle (including the fatigue round), rather than damage per minute, if you like.
There's actually a problem with that, as I'll explain below.
Quote:
Well, frankly, so far, the encounters in the playtest for my group (played one session so far) have been over in less than 3 rounds...
Which also helps emphasize the issue with the model that I'm about to point out.
NemisCassander wrote:
Ssalarn wrote: So you've got (R=rage activation round, B=full rage-buffed combat round, F= fatigue round)
1-Minute Model
RBBFRBBFRB
Compared to
"Realistic" Model
RBBRBBRBBFR
Should this not be 'RBBFRBBFRBBF'? Just checking...
No, it shouldn't, because you're negatively impacting the barbarian's output by counting fatigue rounds that didn't matter. If the combat ended after 3 rounds, then that 4th round is completely irrelevant to the model. It didn't affect combat at all. So the Barbarian's metrics for the comparison are-
RBB (Combat ends, no further tracking relevant)
RBB (Combat ends, no further tracking relevant)
RBBFR (Combat ends, no further tracking relevant)
That's how I came up with the RBBRBBRBBFR "Realistic" Model. The fatigue round will only be relevant to combat comparisons in rounds where the Barbarian is expected to continue fighting while fatigued. In most encounters that aren't "boss fight" difficulty, that means that the barbarian will never experience a fatigue round where he falls below his baseline and increases the fighter's edge, he'll only be "losing" the rage activation action.
|
1 person marked this as a favorite.
|
D@rK-SePHiRoTH- wrote:
Thank you for clarifying this point;
If the barb's dpr is not better while raging then my...
Personally, I do not feel that the Barbarian's other advantages outweigh the lower damage and lower AC relative to the Fighter. It is why I am writing this in a forum based on feedback.
The Barbarian does get better Fortitude saves, and it is unclear right now whether temporary hit points block Enhancement effects. If they do, then the Barbarian's temporary hit points while raging are more helpful than I currently believe. But, looking at a Barbed Devil (level 10 enemy), the Barbarian's temporary hit points at that level will not even protect him or her completely from a single Strike. It does 2d8 + 12 damage on a hit, and a 10th level Barbarian will get ~14 temporary hit points on Rage. Average damage from the Devil is 19. So... yeah.
Now, I do think that the Dragon and Spirit Totems have more utility than most. The Dragon Totem gets more and non-physical Rage damage, which I do not mean to underestimate, and the Spirit Totem gets Ghost Touch for free. That can be useful.
As for the others, I cannot see the Superstition Totem as very meaningful for PCs and as I have stated before I think that the Giant totem is a trap. Perhaps if you focus on a single Strike per turn and other things in that vein, maybe. But under the auspices of this model/test, I am not sure what the purpose of those Totems are.
If we're talking 10th level, the Barbarian also gets Resistance (and, with Fury Totem, it's actually relevant! I'm...less than thrilled with most other Totems) equal to their Con. -4 damage per attack on top of their Temp HP is relevant. Probably relevant enough to make up for the AC penalty they suffer.
I'm less convinced it makes up for any offensive shortcomings.
Deadmanwalking wrote:
Did you do math on this other than that mentioned in the first post?
I did, actually. My original model was per round, but then I noted that Rage consumes an action, and you likely don't start in contact with the enemy, so I had to decide on a longer timeframe.
Quote: Because, as others note, the math mentioned in the first post seems predicated on assumptions that are mostly not true in actual play (actually having to deal with the round of fatigue is pretty rare, IME, and having to re-rage again in the same fight even rarer). Well, I wanted to capture the fact that the Barbarian's average damage per round cannot simply assume 'you're raging'. If so, that makes the model biased towards the Barbarian. Or, perhaps stated another way, if the Barbarian is only balanced vs. the Fighter while they're raging.... why does Raging have a duration and penalties when you leave it?
Quote: And that's a pretty definitive statement to make without having done the math. Which is not to say you're wrong (I suspect that you're right, actually, just from eyeballing it), but this thread was supposed to be about mathematical analysis rather than just eyeballing things, wasn't it? Yes. I will redo it for, hell, every duration from 1 round to X, where X is a comfortably large number.
Ssalarn wrote:
No, it shouldn't, because you're negatively impacting the barbarian's output by counting fatigue rounds that didn't matter. If the combat ended after 3 rounds, then that 4th round is completely irrelevant to the model. It didn't affect combat at all. So the Barbarian's metrics for the comparison are-
RBB (Combat ends, no further tracking relevant)
RBB (Combat ends, no further tracking relevant)
RBBFR (Combat ends, no further tracking relevant)
Uh, no. You are missing the 4th round of combat. If the Barbarian is Fatigued and the combat isn't over, you can't simply say 'the Barbarian isn't necessary'. If it's a round, and we need to assume that the round actually occurs (you know, without modeling initiative _that again favors the Fighter_), then the Barbarian needs to be Fatigued that round, or am I missing something?
Quote: That's how I came up with the RBBRBBRBBFR "Realistic" Model. The fatigue round will only be relevant to combat comparisons in rounds where the Barbarian is expected to continue fighting while fatigued. If I am to model a 4-round battle, the barbarian must absolutely be fatigued during that round, or the fatigue is meaningless and shouldn't exist.
Deadmanwalking wrote: If we're talking 10th level, the Barbarian also gets Resistance (and, with Fury Totem, it's actually relevant! I'm...less than thrilled with most other Totems) equal to their Con. -4 damage per attack on top of their Temp HP is relevant. Probably relevant enough to make up for the AC penalty they suffer.
I'm less convinced it makes up for any offensive shortcomings.
Correct. I forgot the Resistance, which is important at 10th level. It still doesn't completely prevent the attack. With an average of 19 damage, -4 damage from Con, you'll still take 1 HP of damage and lose all of your temps to one Strike. That's.... not so good, is it?
NemisCassander wrote: I did, actually. My original model was per round, but then I noted that Rage consumes an action, and you likely don't start in contact with the enemy, so I had to decide on a longer timeframe. I noticed that after posting, sadly. :(
I think Ssalarn's suggestion is probably the right measure to use, personally (it matches my own experiences running a Barbarian through the first chapter of Doomsday Dawn almost perfectly).
I suspect they'll still come up a bit short. I'm not sure if that's them being undertuned or Fighter overtuned, but it's an issue worth noting.
NemisCassander wrote: Well, I wanted to capture the fact that the Barbarian's average damage per round cannot simply assume 'you're raging'. If so, that makes the model biased towards the Barbarian. Or, perhaps stated another way, if the Barbarian is only balanced vs. the Fighter while they're raging.... why does Raging have a duration and penalties when you leave it? Oh, agreed. It's not actually a one round in four problem, though. It's an occasional one. One round in ten or twelve sounds aboiut right in actual practice, IMO.
NemisCassander wrote: Yes. I will redo it for, hell, every duration from 1 round to X, where X is a comfortably large number.
Cool. I'm very interested in seeing the numbers. Thanks for doing this kind of analysis, it's always interesting to examine.
Oh, funny thing here, guys and ladies.
Preparing to redo this for differing numbers of rounds, I found that the Fatigue rounds were coded as when the roundNumber % 4 == 4... which never happens.
So all of my numbers basically assume that the Barbarian is never Fatigued, and so he is always Raging... scary.
|
1 person marked this as a favorite.
|
NemisCassander wrote:
Uh, no. You are missing the 4th round of combat.
As you yourself agreed in your previous post, there was no 4th round of combat. It never happened, so you shouldn't be counting it into the model if you want data that's relevant to actual gameplay.
Quote:
If the Barbarian is Fatigued and the combat isn't over, you can't simply say 'the Barbarian isn't necessary'. If it's a round, and we need to assume that the round actually occurs (you know, without modeling initiative _that again favors the Fighter_), then the Barbarian needs to be Fatigued that round, or am I missing something?
If the round never happened, than you are adversely penalizing the Barbarian by counting it against him. So if the average encounter is 3 or fewer rounds (and it is, based on the encounter creation guidelines that establish appropriate challenges by level) then you're actually biasing the results against the barbarian.
NemisCassander wrote:
Ssalarn wrote: That's how I came up with the RBBRBBRBBFR "Realistic" Model. The fatigue round will only be relevant to combat comparisons in rounds where the Barbarian is expected to continue fighting while fatigued. If I am to model a 4-round battle, the barbarian must absolutely be fatigued during that round, or the fatigue is meaningless and shouldn't exist.
You shouldn't be modeling a 4-round battle, except in the third encounter where one actually occurred. This is consistent with standard adventure design and the basic framework created by the level-oriented encounter system. As I noted in my model, the Barbarian's performance in an "average" encounter shouldn't account for a fatigue round, because a fatigue round never happened. Neither he nor the fighter were making attacks in that round, because everything was defeated. However in the "boss fight" encounter where a higher difficulty enemy extended combat an additional two rounds, that fatigue became relevant, both because it gave the barbarian a pronounced "weak" round compared to the fighter, and because it required the barbarian to spend an additional activation round on the following round (which granted, if I was playing the Barbarian I might have foregone raging in that final round to just swing out, but I think at that point we genuinely are affecting the data with unsubstantiated variables.)
Well by 17th level, there's a relative chance that the damage output is about 5 points per hit off instead of higher. That being because that is the level in which a Barbarian never fatigues from leaving rage. There's still the cooldown round, but no further penalties exist during that round. Just an average loss of 5 points of damage per swing.
NemisCassander wrote: Uh, no. You are missing the 4th round of combat. What he's saying is that four round combats are rare and unusual. They exist, which is why the Fatigue round does as well, but they are not something that happens most fights. In fact, he specified an average (one in four fights or thereabout) that sounds about right to me as well.
NemisCassander wrote: Correct. I forgot the Resistance, which is important at 10th level. It still doesn't completely prevent the attack. With an average of 19 damage, -4 damage from Con, you'll still take 1 HP of damage and lose all of your temps to one Strike. That's.... not so good, is it? Well, it's also -4 damage on subsequent hits. Over a three round fight, that's gonna hit a Barbarian three times. Which would be 63 damage. The Barbarian reduces that by 26 to 37.
Of course, a Fighter would have +1 AC over a Barbarian (+2 at 11th level)...but I'm not sure that's enough to make nearly as much of a difference. And any non-Fury Totem Barbarian might easily not get the Resistance (depending on foe).
Looking at it, the defensive buff is bigger than I was thinking (at least for Fury Totem). Hmmmm.
Also, examining monsters, I think you might be underselling PC accuracy. Taking the Barbed Devil as an example, it has an AC of 27. A 10th level Barbarian will likely have +17 to hit easily (+10 level, +5 Str, +2 Weapon)...so the Barbarian hits them on a 10 (while a Fighter hits them on an 8), rather than the Fighter on an 11 and the Barbarian on a 13 as your initial post indicates. That seems likely to make a pretty big difference (whether in the Barbarian's favor I'm less certain), and is fairly typical, based on the other monster ACs I've examined (which has been a fair number).
|
4 people marked this as a favorite.
|
Note: This is with a Dragon Barbarian.
Okay, so a 3-round combat, where the Fighter goes
Sudden Charge -> Strike
Strike -> Strike -> Strike
Strike -> Strike -> Strike
and the Barbarian goes
Rage -> Sudden Charge
Strike -> Strike -> Strike
Strike -> Strike -> Strike
gets the following average damage over the combat, by level:
1)
Ftr: 26.79
Bbn: 26.98
2)
Ftr: 26.76
Bbn: 27.06
(This likely is the same result, as nothing really changes numbers-wise.)
3)
Ftr: 26.78
Bbn: 25.40
(Note that the Barbarian's damage goes down because his to-hit got worse compared to the Fighter.)
4)
Ftr: 43.32
Bbn: 36.75
(Extra die of damage.... starting to get noticeable)
5)
Ftr: 43.37
Bbn: 36.83
6)
Ftr: 43.43
Bbn: 36.74
7)
Ftr: 43.46
Bbn: 38.63
8)
Ftr: 59.89
Bbn: 49.89
(At this point, the Fighter is doing on average 20% more than the Barbarian.)
I think I've made my point at this point. Also to note for level 8 is that, while the Fighter has more damage, the Barbarian has a higher variance. So not only does the Fighter do more damage, it's more consistent damage than the Barbarian. And note, again, that this is with a Dragon Barbarian. A non-Dragon Barbarian would be worse, and a Giant one is likely the worst of all, under this model.
I fully admit that the model is very simple, but I would think that the Barbarian would be better than it is...
Deadmanwalking wrote:
What he's saying is that four round combats are rare and unusual. They exist, which is why the Fatigue round does as well, but they are not something that happens most fights. In fact, he specified an average (one in four fights or thereabout) that sounds about right to me as well.
Ah, my apologies. I was reading the sequence as a single encounter, which made it look like it was totally skipping the first Fatigue Round.
Again, my apologies for not understanding... I may be a bit defensive of my models at times...
Quote: Well, it's also -4 damage on subsequent hits. Over a three round fight, that's gonna hit a Barbarian three times. Which would be 63 damage. The Barbarian reduces that by 26 to 37. This is true. But, as you note, it's a crapshoot whether the resistance applies.
Quote: Of course, a Fighter would have +1 AC over a Barbarian (+2 at 11th level)...but I'm not sure that's enough to make nearly as much of a difference. And any non-Fury Totem Barbarian might easily not get the Resistance (depending on foe). Again, true, but this analysis is showing the relative amount of damage that a +2 to hit makes... I am willing to concede that defensively it might be pretty moot...
Quote: Also, examining monsters, I think you might be underselling PC accuracy. Taking the Barbed Decil as an example, it has an Ac of 27. A 10th level Barbarian will likely have +17 to hit easily (+10 level, +5 Str, +2 Weapon)...so the Barbarian hits them on a 10 (while a Fighter hits them on an 8), rather than the Fighter on an 11 and the Barbarian on a 13 as your initial post indicates. That seems likely to make a pretty big difference, and is fairly typical, based on the other monster ACs I've examined (which has been a fair number). Sure. I will model that next. (A TN to roll of 9.) I can tell you that this favors the Fighter, though, because a TN to roll of 9 means that the Fighter will crit on a 19 as well as a 20 (on his first Strike each turn), while the Barbarian does not. But I will model it, nonetheless.
NemisCassander wrote: Ah, my apologies. I was reading the sequence as a single encounter, which made it look like it was totally skipping the first Fatigue Round.
Again, my apologies for not understanding... I may be a bit defensive of my models at times...
No worries on my account. Just clarifying things as best I can. :)
NemisCassander wrote: This is true. But, as you note, it's a crapshoot whether the resistance applies. Yeah. The Totem Variance is a huge issue. Fury is just the best by verging on an order of magnitude, especially at 9th level plus.
NemisCassander wrote: Again, true, but this analysis is showing the relative amount of damage that a +2 to hit makes... I am willing to concede that defensively it might be pretty moot... The Barbarian also has more HP than the Fighter, which seems like a relevant factor as well. Looking at it, the Barbarian's defenses, taken in total, may even be better than the Fighter's. I'm skeptical that it's enough to make up for a big offensive shortfall, though.
NemisCassander wrote: Sure. I will model that next. (A TN to roll of 9.) I can tell you that this favors the Fighter, though, because a TN to roll of 9 means that the Fighter will crit on a 19 as well as a 20 (on his first Strike each turn), while the Barbarian does not. But I will model it, nonetheless. That's actually my expectation. I'm just pedantic enough to want all the baseline numbers more-or-less correct. :)
I suspect the Barbarian needs some offensive fine-tuning, along with several Totems getting complete overhauls.
So, modeling the same as my previous data post, but the base TN to roll is a 9 rather than an 11...
8)
Ftr: 84.54
Bbn: 64.13
Compared to previous data,
Difference:
Ftr: 84.54 - 59.89 = 24.65 damage increase
Bbn: 64.13 - 49.89 = 14.24 damage increase
So, at this point, the Fighter improves when it starts being able to crit more often. (Also, I have assumed that the Barbarian's extra damage doubles on a crit.) Now, just to be thorough, let's assume a TN to roll of 7, which will allow more crits to the Barbarian:
8)
Ftr: 110.38
Bbn: 91.62
When compared to the original numbers:
Difference:
Ftr: 110.38 - 59.89 = 50.49
Bbn: 91.62 - 49.89 = 41.73
So even when the Barbarian starts critting more often, the Fighter is improving faster in the damage department than the Barbarian with lower TNs.
What about higher TNs? Say, 13?
8)
Ftr: 45.89
Bbn: 35.78
So, again, the difference to the baseline of 11:
Difference:
Ftr: 45.89 - 59.89 = -14
Bbn: 35.78 - 49.89 = -14.11
Going up, there appears to be no difference in the gap between the damage numbers. At very high--that is, unreasonably due to PF guidelines--I am sure that they will converge, but there it is. For now.
|
1 person marked this as a favorite.
|
NemisCassander wrote: At levels 1 and 2, the Barbarian's average damage is slightly (~2%) higher the Fighter. At level 3, the Fighter does on average ~5% more damage than the Barbarian. After that, at levels where the Barbarian gets a bonus Damage increase (7th, 11th, etc.), the Fighter does on average ~10% more damage than the Barbarian. At other levels, the Fighter deals on average ~13% more damage than the Barbarian. This is what I thought.
Barbarian should really have +1 hit when raging. And really, Barbarian should be better than just a weapon's expert at some point.
This doesn't even factor in many of the significant feats that a fighter has or AoO, which is huge.
|
1 person marked this as a favorite.
|
Jason S wrote:
This is what I thought.
Barbarian should really have +1 hit when raging. And really, Barbarian should be better than just a weapon's expert at some point.
This doesn't even factor in many of the significant feats that a fighter has or AoO, which is huge.
Ironically, a +1 to hit while raging might make Barbarian an even bigger trap. At 1st/2nd level, the Barbarian would be clearly better than the Fighter offensively. However, I am not sure that a +1 (to go to a relative -1 at 3rd level) is enough to have your damage bonus offset the other bonus.
Really, though, it should be noted that the Barbarian is FAR better than the Fighter at Fortitude and Will saves. Reflex... nasomuch, but Barbarians get Legendary Fortitude and Master Will saves (and other ancillary boosts, like Resistances) with no feat investment. The Fighter does get those floating feats, and that class has options to get at least Master Will saves, but that would, at best, increase them to just par.
All in all, the Barbarian is likely better than the Fighter, defensively, when the spells/suckstuff start flying. The Fighter, though, DOES get Attack of Opportunity, which when the class is next to a caster, that caster feels it.
|
3 people marked this as a favorite.
|
My playtest experience has been most fights lasting 4 or more rounds as PC's struggle to hit if their dice aren't hot.
|
3 people marked this as a favorite.
|
Don't forget the Fighter also gets access to feats that increase his DPR when making full round attacks such as Furious Focus at lower levels and Certain Strikes at higher levels.
The Barbarian doesn't have access to such feats, and thus, his Strike-Strike-Strike sequence is even less effective vs. a Fighter's Strike-Certain Strikes-Certain Strikes sequence.
The +2 Attack bonus from Proficiency the Fighter gets is HUGE. Accuracy matters a lot in PF2, and without it, our friend the Barbarian isn't going to keep up vs. the Fighter.
So if the Barb can't deal as much damage even while raging and takes more damage from having lower defenses, the only advantage it has over the Fighter is the better saves and the ability to shrug off smaller hits via damage resistance. The temp hp it gains from raging is minuscule.
Against a single target, the Barbarian can't compete. It can only hope to use a reach weapon and attack multiple weaker enemies to increase the damage it deals.
|
2 people marked this as a favorite.
|
You know, something I hadn't really noticed until going back to look at the Barbarian (I've had a couple players trying barbarian and played alongside them but barb has never really been my favorite class so I haven't put in play time with one myself) is that the class actually isn't trying to compete in single-target damage. It's got tons of mobility and area effects, so it actually kind of seem like it became the martial controller.
Between abilities like No Escape, Swipe, Cleave, Dragon Totem Breath, Great Cleave, Knockback, Terrifying Howl, etc. they actually seem to be filling a completely different niche.
|
1 person marked this as a favorite.
|
Ssalarn wrote: You know, something I hadn't really noticed until going back to look at the Barbarian (I've had a couple players trying barbarian and played alongside them but barb has never really been my favorite class so I haven't put in play time with one myself) is that the class actually isn't trying to compete in single-target damage. It's got tons of mobility and area effects, so it actually kind of seem like it became the martial controller.
Between abilities like No Escape, Swipe, Cleave, Dragon Totem Breath, Great Cleave, Knockback, Terrifying Howl, etc. they actually seem to be filling a completely different niche.
Be that as it may, the Barbarian still needs a little bit of help to shore up its weak accuracy and defenses. Being a melee controller means it needs to survive vs multiple foes, but the mitigating features it gets are not quite enough.
- Temporary Hp: getting 5-25 hp when you rage works out to not be enough, since it doesn’t enable you to take even 1 hit from equal level enemies (even factoring in DR at higher levels). If fights don’t generally last more then 3-4 rounds, then the Barb is gaining this temp hp only once per encounter, which is not enough. I’d like to see the Barb either get more hp or more frequent hp. Either increase the temps to 2xLevel + Con mod or have it refresh every round the Barb is raging should be ok.
- Damage Resistance: if a Barb starts with 16 Con, he’ll get DR 5 to 2 things by level 15. By that level, that amount is next to nothing, especially when enemies are rolling 3-5 dice worth of damage before static bonuses. In Unity RPG, characters can easily get up to 8-10 physical resistance, and that game’s damage is a lot less than PF2 (only rolling 1-2 dice for damage, max hp is around 100). I like the flavour of Barbs getting resistance based on their Totem, so perhaps they could get a smaller amount of universal Resistance, and more Resistance to the types of damage favoured by their totem.
An example of what I’m talking about: At level 9, Barbarians would get 5+Con mod Resistance while raging, and double that resistance for the two types of damage they choose. Numbers could be tweaked, but I definitely feel the current resistance numbers are too low.
Finally, the level 16 Barb feat Dragon Transformation needs to explicitly say that the altered form uses the Barb’s attack, AC and class DC, otherwise it’s completely useless, since the stats in the spell are much lower than a level 16 PC’s.
|
1 person marked this as a favorite.
|
Pathfinder Rulebook Subscriber
Ssalarn wrote: You know, something I hadn't really noticed until going back to look at the Barbarian (I've had a couple players trying barbarian and played alongside them but barb has never really been my favorite class so I haven't put in play time with one myself) is that the class actually isn't trying to compete in single-target damage. It's got tons of mobility and area effects, so it actually kind of seem like it became the martial controller.
Between abilities like No Escape, Swipe, Cleave, Dragon Totem Breath, Great Cleave, Knockback, Terrifying Howl, etc. they actually seem to be filling a completely different niche.
I think there's definitely something to this. I feel like to really embody martial control it might need to be able to get AoO though. There are cool things it can do with the giant totem's huge reach but that reach doesn't accomplish as much without a reaction based strike.
|
1 person marked this as a favorite.
|
In Pathfinder 1st a +1 equates to 5% chance to hit which is 5% more average weapon damage. In 2e this holds true as well except when you cross the 10-to-hit threshold at which a +1 becomes 10% more average weapon damage because the only number you are really adding is an additional crit value. There is a reddit post where someone calculated the average AC of monsters and it more or less worked out 1-6 Trained hit on 10, 7-12 Trained hit on ~11, and 13-20 Trained hit on 12.
With that in mind a fighter can expect to a hefty percentage more than a trained individual (barbarian below level 13)
For a Full 3 Strike round
at level 1
.10+.05+.05 for +.2 Average weapon damage
at level 3-6
.2+.1+.1 for +.4 Average weapon damage
7-12
.15+.1+.1 for +.35 Average weapon Damage over trained
13-20
Same verus Barbarian as the last as both stepped up by +1 and Ac of monsters by +1
vs trained though
.2+.15+.15 for +.5 average weapon damage
|
2 people marked this as a favorite.
|
Ssalarn wrote: You know, something I hadn't really noticed until going back to look at the Barbarian (I've had a couple players trying barbarian and played alongside them but barb has never really been my favorite class so I haven't put in play time with one myself) is that the class actually isn't trying to compete in single-target damage. It's got tons of mobility and area effects, so it actually kind of seem like it became the martial controller.
Between abilities like No Escape, Swipe, Cleave, Dragon Totem Breath, Great Cleave, Knockback, Terrifying Howl, etc. they actually seem to be filling a completely different niche.
Well, this may very well be true. However, without a clear designation of focus by Paizo, I think it is very murky exactly what the class is trying to be.
I might also point out that most of the abilities you list above require you to hit, as others have noted. If you can't hit, you can't use the abilities, and that's exactly the concern people in this thread (including myself) have.
And, again as others have pointed out, the Barbarian does not get AoO, one of the best controlling mechanisms in the game. It is also quite odd that a Martial Controller does not get _reliably_ better defenses than the single-target alternative (the Fighter). Yes, they get temporary hit points. As noted by myself and others, that is a single hit mitigated in 3 rounds. They also get DR at higher levels, but it is an absolute crapshoot whether the DR even applies.
Now, something you did not mention but I think should be in support of your claim is that the Barbarian gets the critical specialization effect of EVERY weapon while raging while the Fighter only gets it for his 'chosen weapon group' until very high levels, IIRC. Since I have assumed that the Fighter was using a such a chosen weapon, it may not matter, but several of the crit effects are more suitable for control.
I hope to have the Ranger comparison up shortly. By my reading of Hunt Target, it would seem that the Ranger is meant to be the single-target DPS of the three, with the Fighter being in the middle and the Barbarian the multitarget. We shall see.
|
1 person marked this as a favorite.
|
All right. I have the Ranger numbers up now. For the Ranger, I assumed the following rotation:
Hunt Target -> Strike -> Strike
Strike -> Strike -> Strike
Strike -> Strike -> Strike
At level 1, the Ranger uses a Longbow, and all other levels the Ranger uses a Composite Longbow. No penalties for range, cover, screening, or volley range are assumed. The bonus damage for Deadly is given. Ranger Strength modifier is assumed to be half of the Dex modifier.
Here are the comparative damage numbers:
Level 1:
Ftr: 26.80
Bbn: 27.14
Ran: 13.85
Level 2:
Ftr: 26.73
Bbn: 27.00
Ran: 16.46
(increase due to gaining Propulsive)
Level 3:
Ftr: 26.75
Bbn: 25.39
Ran: 16.51
Level 4:
Ftr: 43.39
Bbn: 36.67
Ran: 28.23
Level 5:
Ftr: 43.31
Bbn: 36.81
Ran: 28.19
Level 6:
Ftr: 43.41
Bbn: 36.89
Ran: 28.22
Level 7:
Ftr: 43.23
Bbn: 38.38
Ran: 30.42
Level 8:
Ftr: 59.60
Bbn: 49.80
Ran: 42.16
Level 9:
Ftr: 59.90
Bbn: 49.78
Ran: 42.11
Level 10:
Ftr: 62.60
Bbn: 51.55
Ran: 41.98
Level 11:
Ftr: 62.47
Bbn: 53.16
Ran: 42.19
Level 12:
Ftr: 78.83
Bbn: 64.82
Ran: 55.97
I will be putting this in the OP and editing the assumptions to reflect the current working assumptions....
Edit: Or not. Apparently, you're not allowed to edit posts after a certain amount of time? How odd.
Ranged weapons are not very good, damage-wise, in the Playtest. Comparing them to two-handed melee weapons is always gonna be more weapon-based than Class based.
If you want to compare Rangers, you probably want to go with a greatsword. No reason they can't use it and the Hunt Target bonus does indeed apply.
You could also do TWF, but since Fighters can do that as well and it involves Featr support, this may not be the best for class comparisons either.
Deadmanwalking wrote: Ranged weapons are not very good, damage-wise, in the Playtest. Comparing them to two-handed melee weapons is always gonna be more weapon-based than Class based.
If you want to compare Rangers, you probably want to go with a greatsword. No reason they can't use it and the Hunt Target bonus does indeed apply.
You could also do TWF, but since Fighters can do that as well and it involves Featr support, this may not be the best for class comparisons either.
I could do Fighter vs. Ranger with TWF, I guess. The action economy is much simpler with a Bow, though.
At least with my numbers so far I can show pretty nice quantitative data as to how bad bows (and, really, all ranged weapons) are so far.
I dunno. The biggest issue is that they are not getting damage bonuses from attributes, really. That's where the damage disparity is coming from, I am fairly sure.
Yeah, not getting stat to damage hurts. In practice, Volley also makes longbows nearly worthless in many games, meaning you're using a d6 instead of a d12 (or a d8 even with a longbow, as in your example).
Bows doing somewhat lower damage makes a certain amount of sense, but I think they went overboard.
And for a Class comparison, I really would just go Greatsword Ranger. It's a totally valid build and compares only Class.
Deadmanwalking wrote: Yeah, not getting stat to damage hurts. In practice, Volley also makes longbows nearly worthless in many games, meaning you're using a d6 instead of a d12 (or a d8 even with a longbow, as in your example).
Bows doing somewhat lower damage makes a certain amount of sense, but I think they went overboard.
And for a Class comparison, I really would just go Greatsword Ranger. It's a totally valid build and compares only Class.
I guess, yeah. Why Rangers don't get access to PBS... I mean, it's a glaring oversight, but I understand the devs are looking at Ranger bow builds already, so I feel certain that they're on it...
Okay, Greatsword Ranger? Fair enough... feels so weird, though...
The other option I thought of was to model the Animal Companion + Bow build. My group is still level 1, but the Druid's Raptor companion is doing good work...
how does the results change if we assume "hit on 8" instead of 11 (flanking and +1 buff from an ally) ?
Pathfinder Adventure Path, Lost Omens, Rulebook, Starfinder Roleplaying Game Subscriber
NemisCassander wrote: Deadmanwalking wrote: Ranged weapons are not very good, damage-wise, in the Playtest. Comparing them to two-handed melee weapons is always gonna be more weapon-based than Class based.
If you want to compare Rangers, you probably want to go with a greatsword. No reason they can't use it and the Hunt Target bonus does indeed apply.
You could also do TWF, but since Fighters can do that as well and it involves Featr support, this may not be the best for class comparisons either.
I could do Fighter vs. Ranger with TWF, I guess. The action economy is much simpler with a Bow, though.
At least with my numbers so far I can show pretty nice quantitative data as to how bad bows (and, really, all ranged weapons) are so far.
I dunno. The biggest issue is that they are not getting damage bonuses from attributes, really. That's where the damage disparity is coming from, I am fairly sure. I don’t think I understood that last bit. As you noted above, composite longbows do add attribute damage... (Well, +Str/2, anyway.)
D@rK-SePHiRoTH- wrote: how does the results change if we assume "hit on 8" instead of 11 (flanking and +1 buff from an ally) ? I ran 'hit on 7' above, which shows that the Fighter increases his lead on the Barbarian at most levels.
'hit on 8' is actually worse for the Barbarian and the Ranger, because in this event, the Fighter will crit on his first attack with 18-20, while the Barbarian and Ranger will still only crit on a 20. I didn't run this because it might be seen as Fighter-biased. Actually, wait, the Ranger would normally crit on a 19, in that situation... it would be better, certainly. I can run that really quick.
Level 8, hit on 8:
Ftr: 97.529
Bbn: 76.808
Ran: 63.407
Not as bad as others, I think...
Difference:
Ftr: 97.529 - 59.60 = 37.929
Bbn: 76.808 - 49.80 = 27.008
Ran: 64.407 - 42.16 = 22.247
Make of that what you will.
Porridge: Yes, I assumed that the Str mod for the Composite Longbow would be half of the Dex modifier for the Ranger. So 14 Str, 18 Dex, to start with... Which means that the Ranger at level 2 is getting a +1 bonus to damage, when the Fighter and Barbarian get +4. It's very noticeable.
Pathfinder Adventure Path, Lost Omens, Rulebook, Starfinder Roleplaying Game Subscriber
NemisCassander wrote: D@rK-SePHiRoTH- wrote: how does the results change if we assume "hit on 8" instead of 11 (flanking and +1 buff from an ally) ? I ran 'hit on 7' above, which shows that the Fighter increases his lead on the Barbarian at most levels.
'hit on 8' is actually worse for the Barbarian and the Ranger, because in this event, the Fighter will crit on his first attack with 18-20, while the Barbarian and Ranger will still only crit on a 20. I didn't run this because it might be seen as Fighter-biased. Actually, wait, the Ranger would normally crit on a 19, in that situation... it would be better, certainly. I can run that really quick.
Level 8, hit on 8:
Ftr: 97.529
Bbn: 76.808
Ran: 63.407
Not as bad as others, I think...
Difference:
Ftr: 97.529 - 59.60 = 37.929
Bbn: 76.808 - 49.80 = 27.008
Ran: 64.407 - 42.16 = 22.247
Make of that what you will.
Porridge: Yes, I assumed that the Str mod for the Composite Longbow would be half of the Dex modifier for the Ranger. So 14 Str, 18 Dex, to start with... Which means that the Ranger at level 2 is getting a +1 bonus to damage, when the Fighter and Barbarian get +4. It's very noticeable. Yeah, I can see how the MAD of composite bow users will decrease expected damage by forcing them to rely on two different attributes.
I’m actually OK with ranger weapons being generally less effective than melee ones, since they have a number of other benefits (like not needing to spend actions to move on to another target). And, as Deadmanwalking suggested, a melee ranger is probably a better point of comparison. That said, back of the envelope calculations suggest that the results will be roughly the same: fighter will be best, followed by barbarian, with rangers at the back.
(Ps. Greatly appreciate the quantitative analysis. Keep up the good work!)
|
1 person marked this as a favorite.
|
Thanks for the encouragement. It does help. ;)
I am likely to do the Greatsword Ranger, as well as a Double Slice comparison between Fighter and Ranger.
I feel like I should make a new thread regarding 'The difference a +1 makes: a Simulation-based Analysis'...
I likewise appreciate all of this analytical work you've done. I found it very informative. Making a 'clean' thread; with your results consolidated and edited for presentation might be useful for the Developers (and be more likely to catch their attention I think), but there is also a lot of valuable discussion above that would be regretable to lose track of.
Cantriped wrote: I likewise appreciate all of this analytical work you've done. I found it very informative. Making a 'clean' thread; with your results consolidated and edited for presentation might be useful for the Developers (and be more likely to catch their attention I think), but there is also a lot of valuable discussion above that would be regretable to lose track of. Yeah, I do agree. My biggest concern right now, though, is that I cannot edit the original post and the original post has several incorrect statements about my methodology at this point, largely because of feedback from others here.
|
