First off, I never said anything about PAM/GWM builds. I first referred to PAM, then someone spotted a mistake. They were right, and I corrected myself, but then thought of a new option to achieve something similar, using GWM. Neither build is for extra damage per attack, but to get that BA attack. Also, while personally, I might play Beast for the same reason, other people might choose it simply to get that extra attack I was talking about. Four attacks per round in certain situations already at level 5 is nothing to joke about. Fighters get that at level 20 (well, it's not exactly the same, but even 3 attacks if we ignore the BA one, Fighters get only at level 11).
And again, it really depends on the situation. Let me give you a more specific situation so we can both be thinking about the same thing. Let's assume a party of 4, all level 3 characters. The fight is against 3 Deinonychus. The DDB encounter builder thinks of this as a Hard encounter, which makes sense. From the DMG: "A hard encounter could go badly for the adventurers. Weaker characters might get taken out of the fight, and there’s a slim chance that one or more characters might die."
Okay, now, let's assume your party has... a Barbarian, a Cleric, a Rogue and a Wizard, a somewhat typical adventuring party. You don't want the monsters to reach your backline, because they'll be in trouble. Let's say your DM decides to attack only the Barbarian for whatever reason. Maybe he's standing in a choke point, Idk. Until this point, I see nothing extraordinary about this encounter, correct me if you think otherwise.
Each Deinonychus has 3 attacks per round, and there are 3 of them. Let's assume the Pounce feature doesn't come up because it complicates things, but just saying they might get even more attacks. These are 9 attacks per round. Let's assume your Barbarian has AC 16, though I find that most Barbarians who didn't roll well for stats will have less at this level. Still, let's assume that (in hindsight, I forgot to include shield here, but I already did the calculations so sorry about that. For most Barbarians it will only be +1 to the AC though). The Deinoes have +4 to hit, so they need to roll 12 or higher. Assume you don't use Reckless Attack because of whatever reasons. 40% to hit. 2/5*9=3.6. Three or four hits per round, pretty good on paper. Each hit deals an average of 6.5 damage so you take 19.5-26 damage per round, on average (23.4, exactly). You're a Barbarian, Raging, so half that. 12 damage per round seems reasonable?
Using your tail can, but nothing guarantees that it will, deflect 1 attack per round. Since it's not guaranteed, I think we can agree that we'll just round that 3.6 down to 3, and I think I'm being generous. 3*6.5/2=9.75 damage taken per round, assuming everything is just on average.
Using your bite will affect nothing before you reach 50% HP, but! Your average HP at this level, assuming 16 Con (and again I believe your Con will be lower to match that AC above, unless you rolled well or build for Dex instead of Str), is 12+7*2+3*3=35. 50% of that is 17.5. If the Deinoes attacked before your rage, even if not all of them, you'll already be very close if not already in this zone. If the Deinoes attacked after your rage, you'd lose on average 3.6*6.5/2=11.7 damage per round, so on the second round, you already are at below 50%. And that's assuming you reached the encounter with max HP already, yes? The Deinoes have an AC of 13, which is rather low. You should be able to hit most of the time, even without a Reckless Attack. Assuming you hit, you lower the damage for that round to 9.7 which is already better than the tail.
Now, just for my own curiosity, I'll check a few more levels. - Party of four as mentioned above. - Encounter will be Hard, as many Deinoes as it takes to make it so. - ASI is taken to increase STR, so not taking that into account. - Tail will always round down the number of hits per round. It's far from reality, but I'm not sure how to calculate it easily, so let's assume it's that. It will probably be more generous most of the time this way. - To balance that, I'll assume Barbarian is always >50% because that's not really that hard to get to, and that bite always hits because as levels go up, the likelihood of hitting at least once is going to spike up.
Level 4: - 4 Deinoes. 12 attacks per round. 2/5*12=4.8. - Tail: 4 hits per round. 4*6.5/2=13. - Bite: (4.8*6.5/2)-2=13.6.
Level 5: - 7 Deinoes (to make the encounter Hard). 21 attacks per round. 2/5*21=8.4. - Note: Extra Attack and +1 to Proficiency make Bite more likely to hit at least once while changing nothing for Tail. Also, Bite heals 3 instead of 2 now. - Tail: 8 hits per round. 8*6.5/2=26. - Bite: (8.4*6.5/2)-3=24.3.
Level 6: - 8 Denioes. 24 attacks per round. 2/5*24=9.6. - Note: Bestial Soul, but that makes no difference. - Tail: 9 hits per round. 9*6.5/2=29.25. - Bite: (9.6*6.5/2)-3=28.2.
I'll stop here, since it only gets more complicated, and already have I been oversimplifying many things. My conclusion is that Bite is just better against multiple enemies with multiple attacks. Even if we took many low enemies with a single attack, I believe it's still true. Of course, the Tail has its places to shine, in which it's clearly the top option. So does the Bite. That's why it's good you can change on every rage, and not stick to one option for your entire career, or level, like other subclasses.
Yeah, but that's because I was oversimplifying things. Also, these effects can swing the results in either way. Just think of Bless/Bane. Bless will make attacks more reliable, so Bite will hit more. Bane will make enemies hit less, so Tail will be more useful. Damage is also hard to account for. How much? How often? Which enemy? How do you account for varying initiative orders. It's possible, but too complicated for the purpose of showing the Bite isn't always worse.
Once in a blue moon? Depends on the party, depends on the DM. I find myself in similar (outnumbered) circumstances much more often than I find myself in a big boss encounters. Regarding the Dodge action, it might be better, but most people don't play like that.
But you assume three things that are not necessarily true: 1. No critical hits, which I also tried to ignore because they complicate things but since they cannot be blocked by tail and drive the Barbarian below 50% even faster, they will help the Bite more. With 9 attacks per round, they shouldn't be that rare either. 2. The Barbarian starts with full HP, but that's not always the case. Another fight before, some trap, Wizard Fireballed you... you don't always start with full HP. Even if you heal during a short rest, you might be missing a few points, especially in this level where you don't have many hit dice. 3. Barbarian raged before any attack was made against them. Using RAW initiative, this won't always be the case. Even if only one enemy attacks you before you get to rage, that's a lot of lost HP. These are exactly the things that make such calculations so difficult to make. Too many variables.
You're wrong because your Tail isn't a permanent boost to AC. For a single attack - assuming you didn't use a reaction for something else, but tbf in this scenario you probably didn't - which you choose, you raise your AC by 1d8. Timing, as well as luck, might mean you simply failed to block any in a given round, but you can never block more than once. On average, you will not be blocking an attack each round, far from it. Only in the case one of them rolled exactly your AC you can guarantee to block. In any other case, it's a gamble, and maybe you gambled on the wrong attack. Maybe the next ones will all roll exactly your AC. Unlikely, but I'm exaggerating to make a point. Timing is also a factor here.
Similarly, Bite doesn't always hit. That's true. You're far more likely to hit in this scenario the higher your level goes up, but in the end it's true you won't always hit. You can use your free hands to Grapple and Shove them until you reach below 50% though, to be able to attack with advantage without being attacked with advantage, but I suppose this also applies disadvantage to their attacks, so that starts to become too complicated. Since I only calculated the effect per round, I decided to count that as a hit. Maybe I should have included that in my calculation, though. 60% to hit would mean 1.2 instead of 2. Although this would make Tail more useful at level 3, that will quickly change at higher levels. Also, since we don't know exactly how impactful Tail is on average, it would be wrong to apply this penalty to the example I've given, without looking further into the math of Tail. Maybe at this level it isn't enough, though. Maybe.
Regarding AC, 17 with shield is reasonable, but as I mentioned, I forgot to consider it before claculating. Not all would use a shield, though, and for those this works better. Not all would use a Scale Mail either, since it means disadvantage to Stealth, and Breastplate or Half-Plate at level 3 isn't normal. Considering I forgot shield, 16 is a fair AC. Of course a shield helps, but I was lazy and this conversation isn't so important to start correcting such a mistake. Think of that what you will.
Even taking all this into consideration, while it might flip the numbers at level 3, I believe levels 5 and 6 would still be in favour of Bite, even if at a smaller difference. It is clear Bite isn't perfect, but I still think it is not always inferior to the other options.
Two attacks at level 5 with +6 (+7 usually, but you know what, let's assume you put that ASI into CHA for no reason) against AC 13 means you need to roll 7 or higher to hit. 65% to hit (should be 70 but whatever) for each attack. Your chances of missing both is 12.25%. If you did put ASI into STR, you have only a 10% of missing both. Okay, so 90% of 3 HP you get 2.7 per round on average. At level 5, of course.
Tail doesn't work on every attack until it works, and that's the thing. Maybe you used it to block a 19 and rolled a 1, so they still hit, and on the next attack they roll a total of 18 which you'd 100% block, but you already used your reaction. That's the issue with timing. Also, if you have AC of 18, they have +4 to hit, so they don't need 19 or higher, they need 14 or higher. Also, 18 AC is a vary specific build. Not every Barbarian chooses to use a shield. Especially with Bite, which allows you to use your second attack for any weapon of you choice (so can tail, of course, but Bite really has no reason to be used more than once per round, while Tail has reach so it is useful as sole weapon). Of course you can also just bite again. And while Tail can use a greatsword too, of course, lowering your AC by 2 makes it less useful (though still useful, not like it's a bad decision).
But GWM or PAM isn't an edge case. Rather, it's much more common. Neither really works with a shield (I mean spear does, but eh, THAT is your edge case).
Because at the base of it, the Bite is more simple to calculate. If you wanted to, you could have calculated otherwise, but you don't want to do that either. I DMed for a Beast Barbarian once and he was really disappointed from the tail, exactly because of this reason. He didn't like the Bite either, true, but what I'm trying to say is I base my assumptions on experience. And yes, I do stack the deck, because this is a situational ability. I'm trying to show that there's a situation where it shines brighter, and that the situation isn't so absurd and impossible. Not that it always does.
Wear medium armor and your AC is 16, or 17 once you can pick up half plate.
The claws allows you to have 3 attacks and still carry a shield, making your AC 18 or 19 plus a lot of extra damage.
The tail defense gives you an average of +4.5 to your AC. Carry a great sword and it bumps your AC to an average of 20.5 or 21.5 against a single attack. Sword and shield and your average AC is 22.5 or 23.5 against that attack. In my actual games the tail causes a lot of misses. When they miss it's the functional equivalent of "healing" for all the damage it would have dealt, and it does not matter if you're at quarter health or full health.
At levels 5-8 (3 proficiency bonus), causing just one modest 9 damage attack to miss is the equivalent of landing 3 healing bites. A 15 damage attack caused to miss is equal to 5 healing bites, which would take at least 5 turns.
I doubt anyone still cares about the tail vs bite debate but I got some free time and decided to write a mess of Python to simulate the Deinonychus scenario.
For level 3 I assume a standard array/ point buy with a 16 or 17 Strength, raging, using reckless attacks, already below half health, and an AC of 16. I simulated this round of 3 Deinonychuss and the barbarian attacking a million times and healed a total of 1,755,032 hp for an average of approximately 1.755 per round. This is very close to the expected value which is just your chance to hit multiplied by your proficiency bonus. In this case with a +5 to hit against a target with 13 ac with advantage we have an 87.75% hit chance, or a 0.8775 probability of hitting. at level three our proficiency bonus is 2 so multiplied by our hit chance we get an expected healing of exactly 1.755 per round.
So, how did the tail do? Well across the million simulated rounds it blocked a total of 3,622,911 points of damage. Now we are raging so half of that would have been resisted anyways, so the amount we effectively healed is 1,811,455.5 hp. Unlike the bite I don't have a simple formula to cross check this amount, this is why I went with writing a simulation. However from this we can see we averaged blocking approximately 3.623 damage per round. We know the average damage per attack of a Deinonychus is 6.5 so together we can calculate a successful deflection rate for the tail in this scenario. 3.623 divided by 6.5 gives us a deflection probability of 0.55737 or 55.737%.
Things change however at level 5. Keeping everything else the same the proficiency bonus increases to 3 and I assume the barbarians Strength goes up to 18 or 19 from the ASI/feat. Unsurprisingly the amount of damage prevented by the tail doesn't change because the enemies are the same and we still effectively heal ~2.5hp per round. However for the bite our attack bonus has gone from +5 to +7 and we now have two attacks instead of just one. This makes our new chance to hit at least once across both attacks ~99.61% so now our expected healing from bite becomes 2.988 hp per round.
Now this does assume some things that bias the comparison to favor the bite. The biggest is that we will always be below 50% health. Another is that the monster we are using have a lot of low damage attacks. Also we picked which attack to attempt to deflect with the tail at random. With so many attacks coming in from the same kinds of creatures a player may have an opportunity to pick attacks to try to deflect that are more likely to succeed. One factor working in favor for the tail is that I had the enemies make straight rolls instead of at advantage and the more accurate the enemy the harder it is to deflect their attack. At level 3 giving the enemy advantage brought the effective healing of the tail down to ~1.685 hp per round.
Also I am surprised no one else has mentioned this but the bite has a different action economy cost. So if you want to save your reactions for something like Sentinel AoO lockdown then bite may be the correct choice to increase your survival. And spending your rage to change your natural weapon is something you can do, though it is expensive to do so.
Finally I don't know if this will work but here is my spaghetti code for anyone that wants to play around with it or validate for themselves what I am saying here.
j = 0 # Loop for the number of attacks that target the barbarian # to generate the list of possible blockable hits while j < 9: attackRoll = random.randint(1, 20) hitTotal = attackRoll + toHit if attackRoll != 20 and hitTotal >= barbAc and hitTotal < barbAc + 8: blockableHits.append(attackRoll) j += 1
# Check to make sure there actually is a blockable hit if len(blockableHits) > 0: # Pick one of the blockable hits at random tailBlock = random.randint(1, len(blockableHits)) - 1 # Determine if the hit is blocked if attackDeflected(blockableHits[tailBlock], toHit, barbAc): # Calculate the damage of the blocked hit and add it to the total blocked damPrevented += random.randint(1, damDie) + damBonus
# Make Bite attack barbAttack = random.randint(1, 20) + barbToHit # Rolling with advantage due to Reckless Attacks barbAttack = max(barbAttack, random.randint(1, 20) + barbToHit) # Check if we hit if(barbAttack >= ac): # Add proficiency bonus to the total healed damHealed += barbProf
i += 1
blockedByTail = "Damage blocked by tail: {}" healedByBite = "Damage healed by bite: {}" print(blockedByTail.format(damPrevented / 2)) print(healedByBite.format(damHealed))
Fengeye, that's really good data. I would say though that the AC and average enemy damage you chose heavily favors the bite, at least, based on my games (I can't speak for others). In my games the enemies tend to do way more damage per hit (favoring the tail), and have higher AC (favoring the tail again) especially at level 5+
Also worth mentioning the bite only starts to work once you're below half hit points, so all the work the tail is doing between 100% and 50% health (when the bite is doing nothing) must be considered. If you immediately get hit by something huge and it drops you to, say, 25% hit points, and you never get back to half health, it's a non issue. But the reality is it will be very common that the Barbarian never drops below half, and in those cases the bite is literally useless. Or if you drop to 40% toward the end of fight, it's only "almost" useless.
I'm sure we can theory craft a situation of a very large quantity of low AC enemies who's individual strikes do little damage, and you get ambushed before you can heal and you start combat at 30% health, but when you start having to theory craft... the ability just frankly sucks.
Fengeye, that's really good data. I would say though that the AC and average enemy damage you chose heavily favors the bite, at least, based on my games (I can't speak for others). In my games the enemies tend to do way more damage per hit (favoring the tail), and have higher AC (favoring the tail again) especially at level 5+
Also worth mentioning the bite only starts to work once you're below half hit points, so all the work the tail is doing between 100% and >50% health (when the bite is doing nothing) must be considered. If you immediately get hit by something huge and it drops you to, say, 25% hit points, and you never get back to half health, it's a non issue. But the reality is it will be very common that the Barbarian never drops below half, and in those cases the bite is literally useless. Or if you drop to 40% toward the end of fight, it's only "almost" useless.
I'm sure we can theory craft a situation of a very large quantity of low AC enemies who's individual strikes do little damage, and you get ambushed before you can heal and you start combat at 30% health, but when you start having to theory craft... the ability just frankly sucks.
Oh I totally agree. I didn't think bite would perform as well as it did even in this tilted scenario. I just felt there was a better argument to be made for the bite.
Also I think I figured out a formula for the deflection chance for the tail.
Let 'X' be the set of blockable attack rolls, 'x' be a given roll from that set, and 'p(x)' the probability of rolling x.
Sum from N=8 to N=(8 - size of X): F(N) = p(x)*(N/8)
This starts with the probability of the attack roll exactly matching your AC multiplied by one, and then on to one higher than your AC multiplied by 7/8, and so on. For Advantage and Disadvantage the probability of rolling a given x is depends on what number on the d20 it is. In these cases you have to know which numbers on the d20 correspond to blockable hits and do the summation by hand.
When it is just a straight roll it gets a lot simpler. It is just (1/8)*4.5 or 9/16 when there are the maximum 8 blockable rolls. If you calculate the case for 7 blockable rolls you get 5/8, or 10/16. For 6 blockable rolls it is 11/16, and this pattern continues.
The Deinonychus example above hits on a 12 or higher, so rolls 12-19 are blockable for a total of 8 possible blockable rolls. In my previous post I estimated the chance to block an attack to be 55.737% which doesn't equal 9/16, or 56.25%. This is because in some rounds all of the attacks rolled will be unlockable. This can be due to missing outright, critting, or simply rolling to high above your AC. In my example there is a 55% chance of the enemy missing and a 5% chance of critting. So across 9 attacks a round there is a 0.6^9 probability that all of the attacks are unlockable. When we factor in the chance that at least one unlockable attack is rolled then the theoretical block rate becomes 55.683%. I believe the rest of the difference is just noise in the simulation.
Anyways, thank you for tolerating my completely unnecessary over analysis.
Is it a situational ability? Absolutely. Can you choose this when you rage to match the situation? Yes, this isn't Totem or the one with the auras whose name eludes my mind. There's no point in pointing out the fact that the situation is very specific because that was the idea. Read the title: "Whats the point of Path of the Beast Bite?" This is the point - situations like this one.
There's a difference between the claim "ability x is always better" and my claim which is "ability x isn't necessarily always worse".
Regarding the python code, although I do believe you, I wanted to see for myself for my curiosity. It didn't work though (with the online compiler at least) and I don't know python so I didn't know how to fix it.
Is it a situational ability? Absolutely. Can you choose this when you rage to match the situation? Yes, this isn't Totem or the one with the auras whose name eludes my mind. There's no point in pointing out the fact that the situation is very specific because that was the idea. Read the title: "Whats the point of Path of the Beast Bite?" This is the point - situations like this one.
There's a difference between the claim "ability x is always better" and my claim which is "ability x isn't necessarily always worse".
Regarding the python code, although I do believe you, I wanted to see for myself for my curiosity. It didn't work though (with the online compiler at least) and I don't know python so I didn't know how to fix it.
Exactly, this is why I wanted to better understand the math behind the bite and tail options. The flexibility in the class features is a major strength of the Beast Barbarian.
And thanks for giving the code a try! At least one reason it didn't work is that what I had posted had a bug in it. When I copied it into DnDBeyond all the indents were stripped out and I didn't put them back in for the attackDeflected function. I fixed this and also reduced the round loop to 10. With these changes it now runs when I paste it into a "Try It Yourself" window for the Python Random Number page on W3 Schools here: https://www.w3schools.com/python/gloss_python_random_number.asp
You won't be able to simulate a large number of rounds running it there because they will time out your request. But you will be able to see how things change when you change the different parameters.
Thanks, it seems to be working now. I managed to run even up to 100,000 rounds with no problem. Tail seems to be better at this level, and with these many repetitions, the numbers get pretty close to the math. I suppose that in my first example, the base assumption that Bite always hits was a more significant mistake than I thought it would be since that's really what makes all the difference here.
Given that, I assume that at higher levels as Bite becomes more reliable for pretty much any Barbarian and heals more too, the numbers flip (and probably get closer to my calculations than at level 3). The numbers do flip at level 5 by just increasing barbProf and barbToHit to 3 and 7 respectively (and they do so even if you only change barbProf without touching the to hit at all) though there the similarity between python and C++ ended and I couldn't code for a second attack while making sure not to allow double healing.
I think the conclusion of this long conversation is that Bite's main problem is the 50% health thing. From the 5th level on, the Bite's healing is very consistent. I still think reaching 50% can happen faster than one would first imagine, but the truth is that as we level up that actually becomes harder to achieve. Right now I think level 5 is the Bite's peak, and from there on it only declines.
While it's outside the scope of the discussion, my last thoughts are whether there are combos that make it better. While I cannot recall any, if there are effects that boost healing, or have any other effects, whenever someone regains health (sort of like Life Cleric's ability or the Shephard Druid's, except not locked only for healing done by that specific character). Another option is for any feature that rewards you in any way for harming yourself. There's [Tooltip Not Found], but that's a spell and we're raging so it's not relevant. Another option is Radiant Consumption from the Aasimar race (or the Scourage Aasimar from the legacy races) which also deal damage to you. The last option that comes to mind is either Crown or Redemption Paladins who use their reaction to take damage from someone else on the team (which also means you won't have a reaction for Tail). Any of these features will quicken your descent to below 50%, and the healing from the Bite will allow you to use such features more freely.
First off, I never said anything about PAM/GWM builds. I first referred to PAM, then someone spotted a mistake. They were right, and I corrected myself, but then thought of a new option to achieve something similar, using GWM. Neither build is for extra damage per attack, but to get that BA attack. Also, while personally, I might play Beast for the same reason, other people might choose it simply to get that extra attack I was talking about. Four attacks per round in certain situations already at level 5 is nothing to joke about. Fighters get that at level 20 (well, it's not exactly the same, but even 3 attacks if we ignore the BA one, Fighters get only at level 11).
And again, it really depends on the situation. Let me give you a more specific situation so we can both be thinking about the same thing. Let's assume a party of 4, all level 3 characters. The fight is against 3 Deinonychus. The DDB encounter builder thinks of this as a Hard encounter, which makes sense. From the DMG: "A hard encounter could go badly for the adventurers. Weaker characters might get taken out of the fight, and there’s a slim chance that one or more characters might die."
Okay, now, let's assume your party has... a Barbarian, a Cleric, a Rogue and a Wizard, a somewhat typical adventuring party. You don't want the monsters to reach your backline, because they'll be in trouble. Let's say your DM decides to attack only the Barbarian for whatever reason. Maybe he's standing in a choke point, Idk. Until this point, I see nothing extraordinary about this encounter, correct me if you think otherwise.
Each Deinonychus has 3 attacks per round, and there are 3 of them. Let's assume the Pounce feature doesn't come up because it complicates things, but just saying they might get even more attacks. These are 9 attacks per round. Let's assume your Barbarian has AC 16, though I find that most Barbarians who didn't roll well for stats will have less at this level. Still, let's assume that (in hindsight, I forgot to include shield here, but I already did the calculations so sorry about that. For most Barbarians it will only be +1 to the AC though). The Deinoes have +4 to hit, so they need to roll 12 or higher. Assume you don't use Reckless Attack because of whatever reasons. 40% to hit. 2/5*9=3.6. Three or four hits per round, pretty good on paper. Each hit deals an average of 6.5 damage so you take 19.5-26 damage per round, on average (23.4, exactly). You're a Barbarian, Raging, so half that. 12 damage per round seems reasonable?
Using your tail can, but nothing guarantees that it will, deflect 1 attack per round. Since it's not guaranteed, I think we can agree that we'll just round that 3.6 down to 3, and I think I'm being generous. 3*6.5/2=9.75 damage taken per round, assuming everything is just on average.
Using your bite will affect nothing before you reach 50% HP, but! Your average HP at this level, assuming 16 Con (and again I believe your Con will be lower to match that AC above, unless you rolled well or build for Dex instead of Str), is 12+7*2+3*3=35. 50% of that is 17.5. If the Deinoes attacked before your rage, even if not all of them, you'll already be very close if not already in this zone. If the Deinoes attacked after your rage, you'd lose on average 3.6*6.5/2=11.7 damage per round, so on the second round, you already are at below 50%. And that's assuming you reached the encounter with max HP already, yes? The Deinoes have an AC of 13, which is rather low. You should be able to hit most of the time, even without a Reckless Attack. Assuming you hit, you lower the damage for that round to 9.7 which is already better than the tail.
Now, just for my own curiosity, I'll check a few more levels.
- Party of four as mentioned above.
- Encounter will be Hard, as many Deinoes as it takes to make it so.
- ASI is taken to increase STR, so not taking that into account.
- Tail will always round down the number of hits per round. It's far from reality, but I'm not sure how to calculate it easily, so let's assume it's that. It will probably be more generous most of the time this way.
- To balance that, I'll assume Barbarian is always >50% because that's not really that hard to get to, and that bite always hits because as levels go up, the likelihood of hitting at least once is going to spike up.
Level 4:
- 4 Deinoes. 12 attacks per round. 2/5*12=4.8.
- Tail: 4 hits per round. 4*6.5/2=13.
- Bite: (4.8*6.5/2)-2=13.6.
Level 5:
- 7 Deinoes (to make the encounter Hard). 21 attacks per round. 2/5*21=8.4.
- Note: Extra Attack and +1 to Proficiency make Bite more likely to hit at least once while changing nothing for Tail. Also, Bite heals 3 instead of 2 now.
- Tail: 8 hits per round. 8*6.5/2=26.
- Bite: (8.4*6.5/2)-3=24.3.
Level 6:
- 8 Denioes. 24 attacks per round. 2/5*24=9.6.
- Note: Bestial Soul, but that makes no difference.
- Tail: 9 hits per round. 9*6.5/2=29.25.
- Bite: (9.6*6.5/2)-3=28.2.
I'll stop here, since it only gets more complicated, and already have I been oversimplifying many things. My conclusion is that Bite is just better against multiple enemies with multiple attacks. Even if we took many low enemies with a single attack, I believe it's still true. Of course, the Tail has its places to shine, in which it's clearly the top option. So does the Bite. That's why it's good you can change on every rage, and not stick to one option for your entire career, or level, like other subclasses.
Varielky
Yeah, but that's because I was oversimplifying things. Also, these effects can swing the results in either way. Just think of Bless/Bane. Bless will make attacks more reliable, so Bite will hit more. Bane will make enemies hit less, so Tail will be more useful. Damage is also hard to account for. How much? How often? Which enemy? How do you account for varying initiative orders. It's possible, but too complicated for the purpose of showing the Bite isn't always worse.
Once in a blue moon? Depends on the party, depends on the DM. I find myself in similar (outnumbered) circumstances much more often than I find myself in a big boss encounters. Regarding the Dodge action, it might be better, but most people don't play like that.
But you assume three things that are not necessarily true: 1. No critical hits, which I also tried to ignore because they complicate things but since they cannot be blocked by tail and drive the Barbarian below 50% even faster, they will help the Bite more. With 9 attacks per round, they shouldn't be that rare either. 2. The Barbarian starts with full HP, but that's not always the case. Another fight before, some trap, Wizard Fireballed you... you don't always start with full HP. Even if you heal during a short rest, you might be missing a few points, especially in this level where you don't have many hit dice. 3. Barbarian raged before any attack was made against them. Using RAW initiative, this won't always be the case. Even if only one enemy attacks you before you get to rage, that's a lot of lost HP. These are exactly the things that make such calculations so difficult to make. Too many variables.
You're wrong because your Tail isn't a permanent boost to AC. For a single attack - assuming you didn't use a reaction for something else, but tbf in this scenario you probably didn't - which you choose, you raise your AC by 1d8. Timing, as well as luck, might mean you simply failed to block any in a given round, but you can never block more than once. On average, you will not be blocking an attack each round, far from it. Only in the case one of them rolled exactly your AC you can guarantee to block. In any other case, it's a gamble, and maybe you gambled on the wrong attack. Maybe the next ones will all roll exactly your AC. Unlikely, but I'm exaggerating to make a point. Timing is also a factor here.
Similarly, Bite doesn't always hit. That's true. You're far more likely to hit in this scenario the higher your level goes up, but in the end it's true you won't always hit. You can use your free hands to Grapple and Shove them until you reach below 50% though, to be able to attack with advantage without being attacked with advantage, but I suppose this also applies disadvantage to their attacks, so that starts to become too complicated. Since I only calculated the effect per round, I decided to count that as a hit. Maybe I should have included that in my calculation, though. 60% to hit would mean 1.2 instead of 2. Although this would make Tail more useful at level 3, that will quickly change at higher levels. Also, since we don't know exactly how impactful Tail is on average, it would be wrong to apply this penalty to the example I've given, without looking further into the math of Tail. Maybe at this level it isn't enough, though. Maybe.
Regarding AC, 17 with shield is reasonable, but as I mentioned, I forgot to consider it before claculating. Not all would use a shield, though, and for those this works better. Not all would use a Scale Mail either, since it means disadvantage to Stealth, and Breastplate or Half-Plate at level 3 isn't normal. Considering I forgot shield, 16 is a fair AC. Of course a shield helps, but I was lazy and this conversation isn't so important to start correcting such a mistake. Think of that what you will.
Even taking all this into consideration, while it might flip the numbers at level 3, I believe levels 5 and 6 would still be in favour of Bite, even if at a smaller difference. It is clear Bite isn't perfect, but I still think it is not always inferior to the other options.
Varielky
Two attacks at level 5 with +6 (+7 usually, but you know what, let's assume you put that ASI into CHA for no reason) against AC 13 means you need to roll 7 or higher to hit. 65% to hit (should be 70 but whatever) for each attack. Your chances of missing both is 12.25%. If you did put ASI into STR, you have only a 10% of missing both. Okay, so 90% of 3 HP you get 2.7 per round on average. At level 5, of course.
Tail doesn't work on every attack until it works, and that's the thing. Maybe you used it to block a 19 and rolled a 1, so they still hit, and on the next attack they roll a total of 18 which you'd 100% block, but you already used your reaction. That's the issue with timing. Also, if you have AC of 18, they have +4 to hit, so they don't need 19 or higher, they need 14 or higher. Also, 18 AC is a vary specific build. Not every Barbarian chooses to use a shield. Especially with Bite, which allows you to use your second attack for any weapon of you choice (so can tail, of course, but Bite really has no reason to be used more than once per round, while Tail has reach so it is useful as sole weapon). Of course you can also just bite again. And while Tail can use a greatsword too, of course, lowering your AC by 2 makes it less useful (though still useful, not like it's a bad decision).
But GWM or PAM isn't an edge case. Rather, it's much more common. Neither really works with a shield (I mean spear does, but eh, THAT is your edge case).
Because at the base of it, the Bite is more simple to calculate. If you wanted to, you could have calculated otherwise, but you don't want to do that either. I DMed for a Beast Barbarian once and he was really disappointed from the tail, exactly because of this reason. He didn't like the Bite either, true, but what I'm trying to say is I base my assumptions on experience. And yes, I do stack the deck, because this is a situational ability. I'm trying to show that there's a situation where it shines brighter, and that the situation isn't so absurd and impossible. Not that it always does.
Varielky
Wear medium armor and your AC is 16, or 17 once you can pick up half plate.
The claws allows you to have 3 attacks and still carry a shield, making your AC 18 or 19 plus a lot of extra damage.
The tail defense gives you an average of +4.5 to your AC. Carry a great sword and it bumps your AC to an average of 20.5 or 21.5 against a single attack. Sword and shield and your average AC is 22.5 or 23.5 against that attack. In my actual games the tail causes a lot of misses. When they miss it's the functional equivalent of "healing" for all the damage it would have dealt, and it does not matter if you're at quarter health or full health.
At levels 5-8 (3 proficiency bonus), causing just one modest 9 damage attack to miss is the equivalent of landing 3 healing bites. A 15 damage attack caused to miss is equal to 5 healing bites, which would take at least 5 turns.
I doubt anyone still cares about the tail vs bite debate but I got some free time and decided to write a mess of Python to simulate the Deinonychus scenario.
For level 3 I assume a standard array/ point buy with a 16 or 17 Strength, raging, using reckless attacks, already below half health, and an AC of 16.
I simulated this round of 3 Deinonychuss and the barbarian attacking a million times and healed a total of 1,755,032 hp for an average of approximately 1.755 per round. This is very close to the expected value which is just your chance to hit multiplied by your proficiency bonus. In this case with a +5 to hit against a target with 13 ac with advantage we have an 87.75% hit chance, or a 0.8775 probability of hitting. at level three our proficiency bonus is 2 so multiplied by our hit chance we get an expected healing of exactly 1.755 per round.
So, how did the tail do? Well across the million simulated rounds it blocked a total of 3,622,911 points of damage. Now we are raging so half of that would have been resisted anyways, so the amount we effectively healed is 1,811,455.5 hp. Unlike the bite I don't have a simple formula to cross check this amount, this is why I went with writing a simulation. However from this we can see we averaged blocking approximately 3.623 damage per round. We know the average damage per attack of a Deinonychus is 6.5 so together we can calculate a successful deflection rate for the tail in this scenario. 3.623 divided by 6.5 gives us a deflection probability of 0.55737 or 55.737%.
Things change however at level 5. Keeping everything else the same the proficiency bonus increases to 3 and I assume the barbarians Strength goes up to 18 or 19 from the ASI/feat. Unsurprisingly the amount of damage prevented by the tail doesn't change because the enemies are the same and we still effectively heal ~2.5hp per round. However for the bite our attack bonus has gone from +5 to +7 and we now have two attacks instead of just one. This makes our new chance to hit at least once across both attacks ~99.61% so now our expected healing from bite becomes 2.988 hp per round.
Now this does assume some things that bias the comparison to favor the bite. The biggest is that we will always be below 50% health. Another is that the monster we are using have a lot of low damage attacks. Also we picked which attack to attempt to deflect with the tail at random. With so many attacks coming in from the same kinds of creatures a player may have an opportunity to pick attacks to try to deflect that are more likely to succeed. One factor working in favor for the tail is that I had the enemies make straight rolls instead of at advantage and the more accurate the enemy the harder it is to deflect their attack. At level 3 giving the enemy advantage brought the effective healing of the tail down to ~1.685 hp per round.
Also I am surprised no one else has mentioned this but the bite has a different action economy cost. So if you want to save your reactions for something like Sentinel AoO lockdown then bite may be the correct choice to increase your survival. And spending your rage to change your natural weapon is something you can do, though it is expensive to do so.
Finally I don't know if this will work but here is my spaghetti code for anyone that wants to play around with it or validate for themselves what I am saying here.
import random
def attackDeflected(attackRoll, toHit, barbAc):
acBonus = random.randint(1, 8)
return barbAc + acBonus > attackRoll + toHit
# Store damage healed by the bite
damHealed = 0
# Store the damage prevented by the tail
damPrevented = 0
i = 0
# Loop for the number of rounds you want to simulate
while i < 10:
# List to contain hits that may be blocked by the tail
blockableHits = []
# Enemy stats
damDie = 8
damBonus = 2
toHit = 4
ac = 13
# Barbarian stats
barbProf = 2
barbToHit = 5
barbAc = 16
j = 0
# Loop for the number of attacks that target the barbarian
# to generate the list of possible blockable hits
while j < 9:
attackRoll = random.randint(1, 20)
hitTotal = attackRoll + toHit
if attackRoll != 20 and hitTotal >= barbAc and hitTotal < barbAc + 8:
blockableHits.append(attackRoll)
j += 1
# Check to make sure there actually is a blockable hit
if len(blockableHits) > 0:
# Pick one of the blockable hits at random
tailBlock = random.randint(1, len(blockableHits)) - 1
# Determine if the hit is blocked
if attackDeflected(blockableHits[tailBlock], toHit, barbAc):
# Calculate the damage of the blocked hit and add it to the total blocked
damPrevented += random.randint(1, damDie) + damBonus
# Make Bite attack
barbAttack = random.randint(1, 20) + barbToHit
# Rolling with advantage due to Reckless Attacks
barbAttack = max(barbAttack, random.randint(1, 20) + barbToHit)
# Check if we hit
if(barbAttack >= ac):
# Add proficiency bonus to the total healed
damHealed += barbProf
i += 1
blockedByTail = "Damage blocked by tail: {}"
healedByBite = "Damage healed by bite: {}"
print(blockedByTail.format(damPrevented / 2))
print(healedByBite.format(damHealed))
Edit: Yay! it worked.
Fengeye, that's really good data. I would say though that the AC and average enemy damage you chose heavily favors the bite, at least, based on my games (I can't speak for others). In my games the enemies tend to do way more damage per hit (favoring the tail), and have higher AC (favoring the tail again) especially at level 5+
Also worth mentioning the bite only starts to work once you're below half hit points, so all the work the tail is doing between 100% and 50% health (when the bite is doing nothing) must be considered. If you immediately get hit by something huge and it drops you to, say, 25% hit points, and you never get back to half health, it's a non issue. But the reality is it will be very common that the Barbarian never drops below half, and in those cases the bite is literally useless. Or if you drop to 40% toward the end of fight, it's only "almost" useless.
I'm sure we can theory craft a situation of a very large quantity of low AC enemies who's individual strikes do little damage, and you get ambushed before you can heal and you start combat at 30% health, but when you start having to theory craft... the ability just frankly sucks.
Oh I totally agree. I didn't think bite would perform as well as it did even in this tilted scenario. I just felt there was a better argument to be made for the bite.
Also I think I figured out a formula for the deflection chance for the tail.
Let 'X' be the set of blockable attack rolls, 'x' be a given roll from that set, and 'p(x)' the probability of rolling x.
Sum from N=8 to N=(8 - size of X): F(N) = p(x)*(N/8)
This starts with the probability of the attack roll exactly matching your AC multiplied by one, and then on to one higher than your AC multiplied by 7/8, and so on. For Advantage and Disadvantage the probability of rolling a given x is depends on what number on the d20 it is. In these cases you have to know which numbers on the d20 correspond to blockable hits and do the summation by hand.
When it is just a straight roll it gets a lot simpler. It is just (1/8)*4.5 or 9/16 when there are the maximum 8 blockable rolls. If you calculate the case for 7 blockable rolls you get 5/8, or 10/16. For 6 blockable rolls it is 11/16, and this pattern continues.
The Deinonychus example above hits on a 12 or higher, so rolls 12-19 are blockable for a total of 8 possible blockable rolls. In my previous post I estimated the chance to block an attack to be 55.737% which doesn't equal 9/16, or 56.25%. This is because in some rounds all of the attacks rolled will be unlockable. This can be due to missing outright, critting, or simply rolling to high above your AC. In my example there is a 55% chance of the enemy missing and a 5% chance of critting. So across 9 attacks a round there is a 0.6^9 probability that all of the attacks are unlockable. When we factor in the chance that at least one unlockable attack is rolled then the theoretical block rate becomes 55.683%. I believe the rest of the difference is just noise in the simulation.
Anyways, thank you for tolerating my completely unnecessary over analysis.
Is it a situational ability? Absolutely. Can you choose this when you rage to match the situation? Yes, this isn't Totem or the one with the auras whose name eludes my mind. There's no point in pointing out the fact that the situation is very specific because that was the idea. Read the title: "Whats the point of Path of the Beast Bite?" This is the point - situations like this one.
There's a difference between the claim "ability x is always better" and my claim which is "ability x isn't necessarily always worse".
Regarding the python code, although I do believe you, I wanted to see for myself for my curiosity. It didn't work though (with the online compiler at least) and I don't know python so I didn't know how to fix it.
Varielky
Exactly, this is why I wanted to better understand the math behind the bite and tail options. The flexibility in the class features is a major strength of the Beast Barbarian.
And thanks for giving the code a try! At least one reason it didn't work is that what I had posted had a bug in it. When I copied it into DnDBeyond all the indents were stripped out and I didn't put them back in for the attackDeflected function. I fixed this and also reduced the round loop to 10. With these changes it now runs when I paste it into a "Try It Yourself" window for the Python Random Number page on W3 Schools here: https://www.w3schools.com/python/gloss_python_random_number.asp
You won't be able to simulate a large number of rounds running it there because they will time out your request. But you will be able to see how things change when you change the different parameters.
Thanks, it seems to be working now. I managed to run even up to 100,000 rounds with no problem. Tail seems to be better at this level, and with these many repetitions, the numbers get pretty close to the math. I suppose that in my first example, the base assumption that Bite always hits was a more significant mistake than I thought it would be since that's really what makes all the difference here.
Given that, I assume that at higher levels as Bite becomes more reliable for pretty much any Barbarian and heals more too, the numbers flip (and probably get closer to my calculations than at level 3). The numbers do flip at level 5 by just increasing barbProf and barbToHit to 3 and 7 respectively (and they do so even if you only change barbProf without touching the to hit at all) though there the similarity between python and C++ ended and I couldn't code for a second attack while making sure not to allow double healing.
I think the conclusion of this long conversation is that Bite's main problem is the 50% health thing. From the 5th level on, the Bite's healing is very consistent. I still think reaching 50% can happen faster than one would first imagine, but the truth is that as we level up that actually becomes harder to achieve. Right now I think level 5 is the Bite's peak, and from there on it only declines.
While it's outside the scope of the discussion, my last thoughts are whether there are combos that make it better. While I cannot recall any, if there are effects that boost healing, or have any other effects, whenever someone regains health (sort of like Life Cleric's ability or the Shephard Druid's, except not locked only for healing done by that specific character). Another option is for any feature that rewards you in any way for harming yourself. There's [Tooltip Not Found], but that's a spell and we're raging so it's not relevant. Another option is Radiant Consumption from the Aasimar race (or the Scourage Aasimar from the legacy races) which also deal damage to you. The last option that comes to mind is either Crown or Redemption Paladins who use their reaction to take damage from someone else on the team (which also means you won't have a reaction for Tail). Any of these features will quicken your descent to below 50%, and the healing from the Bite will allow you to use such features more freely.
Varielky