Comprehensive DPR Calculator (v2.0!)

[Kerrec]

EDIT: Never mind. Reading comprehension problem.

[LudicSavant]

Fooling around with your awesome calculator and I ran into a problem with respect to Elven Accuracy.

From what I understand, a normal attack roll becomes an attack roll with advantage when you have Elven Accuracy. An attack roll with advantage becomes Super Advantage (your nomenclature per the documentation). And of course, Disadvantage + Elven Accuracy is a weird thing all it's own...

But your calculator doesn't seem to turn a normal attack into an attack with advantage when calculating the Crit Chance% when Elven Accuracy is checked.





Should have circled the relevant bits before uploading those images. But notice the top doesn't have Elven Accuracy checked, and the crit chance% for a normal attack is 0.15. Then the bottom image does have the Elven Accuracy checked, but the crit chance% is still 0.15.

The calculator is correct.

Elven Accuracy doesn’t turn a normal attack into advantage, it only turns an Advantage roll into triple Advantage.

[Kerrec]

Wow, you replied pretty quick. I figured it out too, but apparently too late.

[Bowbreaker]

@LudicSavant: Completely off topic, but your PM folder is full. I stumbled upon some of your homebrew worldbuilding and wanted to ask you about it.

[LudicSavant]

@LudicSavant: Completely off topic, but your PM folder is full. I stumbled upon some of your homebrew worldbuilding and wanted to ask you about it.

NP. Yeah, my PM box is almost perpetually full. I just cleared some space... for however long that will last ^^;;

(It's a shame they only let you keep 100 messages on here)

[Bowbreaker]

Damn. This forum also doesn't inform me when someone quotes me in a thread. So I was too slow. Do you have any other messaging system you are willing to share? Maybe something not so private, like your username on another forum you like to frequent? I have an account on SpaceBattles, SufficientVelocity, Reddit, Discord, Steam...

[LudicSavant]

Damn. This forum also doesn't inform me when someone quotes me in a thread. So I was too slow. Do you have any other messaging system you are willing to share? Maybe something not so private, like your username on another forum you like to frequent? I have an account on SpaceBattles, SufficientVelocity, Reddit, Discord, Steam...

Best way to reach me on short notice is generally via Discord; anyone who wants to can PM me asking for an address (if they can manage to send one when the PM box isn't already full). Sent you mine.

[Hael]

Any chance this super useful tool might get an update? Several of the new feats (piercer/Crusher/Slasher) changes the computation somewhat. Also a few others like the Gloomstalker feature and precision attack could stand to be included.

I usually write my own script for that sort of thing on any dice, but it gets tedious and it’s easy to make a mistake, especially when you combine them.

[LudicSavant]

Any chance this super useful tool might get an update? Several of the new feats (piercer/Crusher/Slasher) changes the computation somewhat. Also a few others like the Gloomstalker feature and precision attack could stand to be included.

I usually write my own script for that sort of thing on any dice, but it gets tedious and it’s easy to make a mistake, especially when you combine them.

As usual with this sort of thing it's really a question of whether Aureus or I have the time to do it -- and aren't using our free time on other things, as we often are.

One of the obstacles isn't just calculating it, but calculating it in a way that is compatible with everything else on the spreadsheet. Piercer is unusually complex because it cares about the order of your actions, and has an element of player choice: You basically need to weigh the value of rerolling your current hit, vs the probability that you will hit a lower damage roll later on in the turn (thus raising the DPR gain). Calculating the minimum DPR gain (e.g. reroll on first below-average hit) is pretty simple, but calculating the optimal DPR gain is complex.

If you have something on AnyDice for Piercer I'd be curious to see how you implemented it.

[VicenarySolid]

Hey, i wanna say thanks to you, this calc is super good!
I have a question is there a function to add on-hit effects like hex or hunter mark, or am i missing something?

[LudicSavant]

Hey, i wanna say thanks to you, this calc is super good!
I have a question is there a function to add on-hit effects like hex or hunter mark, or am i missing something?

Those are just extra damage dice, enter 'em like any other damage dice.

[Pimplup]

Just wanted to say thanks a bunch for making this—it's super intuitive and powerful, and the fact that you made it and are sharing it for free is awesome. :D

[nel__]

Would it be possible to add the lucky feat to the calculations?

[Amdy_vill]

my only suggestion would be to add reaction attack calculators as there are some feat combos and builds that can consistently get reaction damage out.

[Cass]

my only suggestion would be to add reaction attack calculators as there are some feat combos and builds that can consistently get reaction damage out.

I would just add another attack to the computation.

[quindraco]

I was noodling around on the calculator, appreciating its many excellent qualities, but it has a drawback which is fine, except that the relevant tooltip isn't explicit.

It's frankly common and typical for a given situation to have multiple rules entries which are mutually exclusive and so your DM has to decide between them. To give an example the calculator doesn't care about it, I can quote you absolutely canon RAW that states when you throw a longsword you must use your DEX modifier and I can do the same thing for STR modifier. The point of this preface is to emphasize that you cannot assume WOTC has issued errata to address an issue, even when an issue is known.

That applies to the calculator, but it doesn't tell you which of the two RAWs it's using: Great Weapon Fighting.

Here's the RAW rules entry the calculator is using:

When you roll a 1 or 2 on a damage die for an attack you
make with a melee weapon that you are wielding with
two hands, you can reroll the die and must use the new
roll, even if the new roll is a 1 or a 2. The weapon must
have the two-handed or versatile property for you to gain
this benefit. -- PHB p72

Here's another equally RAW rules entry:
The Great Weapon Fighting feature—which is shared by fighters and paladins—is meant to benefit only the damage roll of the weapon used with the feature. For example, if you use a greatsword with the feature, you can reroll any 1 or 2 you roll on the weapon’s 2d6. If you’re a paladin and use Divine Smite with the greatsword, Great Weapon Fighting doesn’t let you reroll a 1 or 2 that you roll for the damage of Divine Smite. -- SAC v2.6, p4 (reminder SAC is a core rules document on par with the PHB, DMG, and MM)

There's no question these two rules contradict each other - the PHB says reroll the dice for the attack, while the SAC says to do it for the weapon. And there's no question the only reasonable choice here for the calculator is the attack, because the calculator doesn't distinguish weapon damage from non-weapon damage - in fact, you can't even give the calculator damage dice for the attack that don't crit. But the GWF tooltip doesn't make it clear that the calculator is using the PHB text, not the SAC text. I made sure of it by taking a look directly at the code for it. Elemental Adept has a similar issue if you're e.g. attacking with a flame tongue sword - it's just how the calculator works. It would be nice having the tooltips clarify that. :)

Speaking of which, while I was at it, I wrote a function for calculating the damage dice coefficients, to make it easier to tell specific dice whether or not to apply elemental adept and great weapon fighting:

function damdie(sides, ea, gwf) {
    var g = 0;
    if (gwf) {
      g = 1-2/sides;
    }
    var e = 0;
    if (ea) {
      e = (1+gwf)/Math.pow(sides,1+gwf);
    }
    return (sides+1)/2 + g + e;
}

That makes averageDamage a much shorter function:

function averageDamage(four, six, eight, ten, twelve, bonus, isGWF, isEA) {
  return damdie(4,isEA,isGWF)*four+damdie(6,isEA,isGWF)*six+damdie(8,isEA,isGWF)*eight+damdie(10,isEA,isGWF)*ten+damdie(12,isEA,isGWF)*twelve+bonus;
}

Tested it for all four situations for every die, no errors.

What I'd really like to do the calculator can't handle at all, unfortunately - calculate the DPR of a ranger with the Tasha's companion murdering someone in melee after the beast of the land has charged, at levels 10 and 11. It can't handle attacks with riders you have to save against that modify later attacks. :'( Still, it's really cool and useful!

[HX2GPX]

Is there any way to specify an auto-crit on hit (such as on someone effected by Hold Person)

[Chaos Jackal]

Is there any way to specify an auto-crit on hit (such as on someone effected by Hold Person)

You could assign 2 (or the minimum number needed to hit) to the crit range, so that any hit is automatically considered a crit.

[HX2GPX]

You could assign 2 (or the minimum number needed to hit) to the crit range, so that any hit is automatically considered a crit.

Thanks, this works great!

[Zrave]

I'm working on a calculator as well, and I've spent several hours now working on the Savage Attacker feat. It was easy enough to produce a table of damage increases relative to the number and type of dice being rolled (from 1d4 to 8d12).

However, things got a lot trickier when considering multiple attacks in one turn since the feat can only be activated once. My solution was to break the range of rolls into deciles to compare the present benefit with the expected benefit from future rolls. I then took the average over all deciles. It's not perfect, but it's more precise than not considering it (the function is concave up, so this method underestimates the true benefit). The interaction between GWF and Savage Attacker is also a bit of a computational nightmare, which I simplified by simply treating 1s and 2s as the average die roll. Are you interested in discussing any of this work? I feel like there must be a better way of going about it, such as using a formula for the frequency of each result.

I noticed you stayed away from the more cumbersome calculation with once per turn effects (Essentially the same that I used for savage attacker: comparing the benefit now with the expected benefit over the remaining attacks, rather than always taking the benefit now). The difference is potentially significant, especially with advantage or a wide critical range. I'm currently working on that section of my calculator, I'll post later with an estimate of how much difference it makes.

[Zrave]

As promised, I calculated the difference between using a "once per round" effect on the first hit, versus considering the expected value of the further attacks before making that decision.

Let's assume a level 20 fighter, dual wielding longswords with 5 attacks per turn, each with a 60% chance to hit and 5% chance to crit. They have one Battlemaster maneuver remaining and want to optimize it.

As expected, the effect without advantage is very small. If the first attack is a hit, it's a tiny gain of 0.33 DPR to hold off instead of using it right away. The effect is neglible past that, so they should use it on a hit starting from the second attack.

However, with advantage it's a different story. It's now a 1.31 DPR gain to hold off on the first hit, and if the second attack is a hit waiting is still a 0.89 DPR gain.

Under these parameters, it takes a minimum of 3 attacks with advantage for waiting to be profitable, so a rogue is unlikely to benefit from this strategy.

On the other hand, it's probably really useful for an elven rogue with 5 levels of champion and elven accuracy.

[LordShade]

As an aside, for anyone getting the #NAME error, try making a copy in Google Docs as opposed to downloading as an Excel file. The special functions that Ludic/Aureus have programmed into the sheet are apparently not recognized or don't exist in Excel.

[Person_Man]

Question to those who have played with this a lot - do any options stand out as being particularly efficient at increasing damage?

Honestly, I think the 5E devs did a great job on this and there are no clearly “broken” damage options I’ve seen yet, though there is some splat that grants moderately better AC or slightly easier access to Advantage that might bother some DMs. The most powerful options seem to still be high level Summons, Dominate, and battlefield control. Which is fine with me. Though it would be nice if non-full casters got a bigger damage bump at high levels to compensate for the lack of these options.

Also, more kudos for the OP who developed this. You guys are great and math rules!

[LordShade]

My takeaways--advantage is huge, especially for the power attack feats. I intuitively knew this already but the DPR calculator reinforces just how important it is. If you have a -5/10 feat, you should be selling out on just about everything else to get advantage any way you can.

Conversely, it shows how bad other options are. I already knew GWF was terrible, but this DPR not only 1) shows mathematically how little damage it adds and 2) shows you what you could be gaining if the fighting style helped you get advantage (like Blindfighting).

I was also surprised about Elven Accuracy. It's certainly good, but it's not as good as I thought it was. I think a lot of the value of the feat is the +1 ASI. If it was triple advantage only, without a stat increase, it would probably be a pass in most cases.

[MaxWilson]

Question to those who have played with this a lot - do any options stand out as being particularly efficient at increasing damage?

No-concentration summons are a standout, especially if you can give them advantage somehow such as with a Web spell. Doubling your damage is as easy as casting twice much Animate Dead/Tiny Servant. (You can also do things like give your skeletons dual shortswords for TWF to almost double it again.)

This stands in stark contrast to character-intrinsic options like feats and abilities cantrip damage, which scale slowly: it is very, very difficult for a Sharpshooter archer with Extra Attack already to double his damage (and the easiest way is for him to Conjure Animals before shooting).

Minionmancy is playing 5E on Easy mode.

[LudicSavant]

Question to those who have played with this a lot - do any options stand out as being particularly efficient at increasing damage?

Honestly, I think the 5E devs did a great job on this and there are no clearly “broken” damage options I’ve seen yet, though there is some splat that grants moderately better AC or slightly easier access to Advantage that might bother some DMs. The most powerful options seem to still be high level Summons, Dominate, and battlefield control. Which is fine with me. Though it would be nice if non-full casters got a bigger damage bump at high levels to compensate for the lack of these options.

Also, more kudos for the OP who developed this. You guys are great and math rules!

My takeaways--advantage is huge, especially for the power attack feats. I intuitively knew this already but the DPR calculator reinforces just how important it is. If you have a -5/10 feat, you should be selling out on just about everything else to get advantage any way you can.

Conversely, it shows how bad other options are. I already knew GWF was terrible, but this DPR not only 1) shows mathematically how little damage it adds and 2) shows you what you could be gaining if the fighting style helped you get advantage (like Blindfighting).

I was also surprised about Elven Accuracy. It's certainly good, but it's not as good as I thought it was. I think a lot of the value of the feat is the +1 ASI. If it was triple advantage only, without a stat increase, it would probably be a pass in most cases.

No-concentration summons are a standout, especially if you can give them advantage somehow such as with a Web spell. Doubling your damage is as easy as casting twice much Animate Dead/Tiny Servant. (You can also do things like give your skeletons dual shortswords for TWF to almost double it again.)

This stands in stark contrast to character-intrinsic options like feats and abilities cantrip damage, which scale slowly: it is very, very difficult for a Sharpshooter archer with Extra Attack already to double his damage (and the easiest way is for him to Conjure Animals before shooting).

Minionmancy is playing 5E on Easy mode.

I must say, I'm glad to see people taking the topic in this direction, with people sharing their discoveries! Empowering people to learn about the game and see for themselves is what I made this tool for, after all

Feel free to use this thread to share such things!

[Dalinar]

Question to those who have played with this a lot - do any options stand out as being particularly efficient at increasing damage?

It also falls under the minionmancy umbrella, but Simulacrum = two concentrations, which means potentially two Animate Objects or Tasha summons; also twice as many regular spells, not counting your 7th-level slot. There's a reason both the casting time and the material cost of it are very high. I haven't actually run the numbers, but I've seen math elsewhere showing, for instance, 8th-level Summon Celestial on its own out-DPRing most martial builds. (I think that was Ludic, actually!)

[quindraco]

It also falls under the minionmancy umbrella, but Simulacrum = two concentrations, which means potentially two Animate Objects or Tasha summons; also twice as many regular spells, not counting your 7th-level slot. There's a reason both the casting time and the material cost of it are very high. I haven't actually run the numbers, but I've seen math elsewhere showing, for instance, 8th-level Summon Celestial on its own out-DPRing most martial builds. (I think that was Ludic, actually!)

Yeah, infinite Simulacra makes all of the math break down. There isn't a whole lot of point comparing numbers at level 17+ unless you put constraints down.

I would be interested in someone inverting the standard narrative of making a monobuild and analyzing its output, instead explicitly listing just how good a legal glyph of warding spam chamber they can make. I've never actually sat down and tried it.

[Hael]

I was also surprised about Elven Accuracy. It's certainly good, but it's not as good as I thought it was. I think a lot of the value of the feat is the +1 ASI. If it was triple advantage only, without a stat increase, it would probably be a pass in most cases.

It adds the usual +1asi, which is good, but how good it is is build specific. It’s ok in general, but where it really shines is with hexblade or dex champions bc the extra crit range starts actually being quite effective for dpr increases, particularly against high AC foes.

[Person_Man]

No-concentration summons are a standout...
Minionmancy is playing 5E on Easy mode.

Strongly agreed.

Early on in development, a number of play testers (myself included) pointed out that the most important balancing factor in the game was not damage output per se, but the Action Economy.

WotC took this seriously, and went out of their way to make sure that most combat benefits required an Action, Bonus Action, Reaction, or Concentration. This also has the effect of making the passive bonuses they do allow a bigger deal, because they can be stacked.

But then they ran into a worldbuilding problem. Undead are a thing in D&D. Where do they come from? If its magic, why can’t players access it? If Animate Dead has a high cost attached to discourage players, why would anyone in the game world make them? If they disappear when Concentration is broken, why are there still weak undead roaming around in the game world for the players to encounter? Playing a Necromancer or other summons focused build has almost always been a thing in D&D, and the devs were super concerned with making sure everyone felt like “their version” of a class could be played.

The net result is that Animate and summons spells were somewhat nerfed, but are still a thing that breaks the Action Economy.

On the plus side, I think this math “problem” is easily fixed by competent DMs. Yes, the Necromancer can have his squad of undead minions. And if thats what they want to do, let them have fun with their toys. Let them play some battles on easy mode, because they clearly want to. But bringing minions into any populated area is going to be problematic, mobs of weak creatures can be wiped out by area of effect builds, its hard to maneuver your mob of minions in a narrow dungeon, and having a bunch of undead will probably draw the attention of local Paladins/Clerics/etc to come and stop you.

So its probably a better idea to just animate/summon a small-ish group of minions, don’t let them stand near each other, give them shortbows and two shortswords, keep them out of town, and have them hang back and use them as needed, instead of investing significant class resources into them.