Mathematically speaking, who benefits more from a +2 bump to

>>50521223
Depends mostly on the level, but in general, AC has diminishing returns, because once you hit (enemy attack bonus + 20), you're untouchable. 22 would hit that cap sooner than 18.

File: 1468667840192.jpg (23KB, 500x285px) Image search: [Google]

23KB, 500x285px

18, because at 18 he can still be hit with a roll with no modifier but 22 can't be hit without a modifier

>>50521237
>>50521251
Enemies always have at least +3, typically +6, sometimes +10

The future value of the +2 shouldn't be considered, just this jump in particular.

>>50521223
Depends exclusively on the opponent.
So long as both characters ACs and their ACs+ bonus are within the range of the opponents' attack rolls, the one with the better AC benefits more.

But that is not the only consideration to be taken. What about the efficiency of healing and HP? You may be able to reduce one guy's damage intake by 80% where you could only reduce the other guy's damage by 50%. But that means that most of the damage goes to a single character. Can you keep that character alive and active at the front line to protect the weaker characters?

Depends heavily on the difference between your current AC and the attack bonus of your average opponent. The relative damage you take after the increase in AC is
>(19+d)/(21+d)
where d is their attack bonus minus your current AC. Or alternatively
>(21+d-x)/(21+d)
where x is the change in AC.

This approaches 1 as d gets larger and larger (i.e. they get better at attacking, your armor is weaker), which means that 2 AC does "more" (percent wise) on a person with already high AC. For example, if d = (AB - AC) = -17, a 2 AC bonus halves the damage you receive. But if d is only -8, the damage is only reduced by 15%. Past d = 0, your returns are always less than 10% for a 2 AC bonus.

But as >>50521361 noted, maybe taking 10% less damage on your Wizard is a lot better than the taking 20% less damage on your Barbarian, because he can heal easier, or something.

>>50521223
>he doesnt know about diminishing returns


American education

File: 2lewd abandon thread.gif (3MB, 500x281px) Image search: [Google]

3MB, 500x281px

>>50521223

>>50521223

Depends what level bracket and edition you're talking. 3.5, it's usually better for the guy with 20 AC because 18 is such a terrible score for anything beyond the first few levels that you might as well focus on other forms of damage avoidance. On the other hand, the attack modifier range is 5e is so constrained that a +2 bonus matters more for the guy with lower AC, because you're not going to see anything with +20 to hit.

Since you said mathematically, I won't go into party roles mattering.

>>50521895
5e

There's also advantage and disadvantage to consider

>>50521895
What this guy said.

I also think the lowest should be raised but that's primarily because I'm the type of player who likes a level playing field and wants to focus on covering weak areas.

File: Too lewd (but peeking).jpg (43KB, 380x438px) Image search: [Google]

43KB, 380x438px

>>50521820

>>50521929
Yeah well 5e is fucking awful so I think you should just go with the lower AC so you can get killed and leave the game faster. Though, considering you only have a 1 in 8 chance of dying when you go down anyway, because of pussy-ass death saves, and you can't even die in one hit, you'd probably find it's harder to die in 5e than in 4e.

>>50521948
best girl

File: come-get-the-phb.gif (156KB, 540x281px) Image search: [Google]

156KB, 540x281px

>>50521971
It's okay, anon. Admitting you like shit games is fine, no need to insult good ones to keep up appearances.

File: 1468170748600.png (201KB, 331x378px) Image search: [Google]

201KB, 331x378px

>>50522007
>good
>5e
Yeah man first level d+d is so good I wanna play it for 20 levels???

>>50521971
There is still insta death in 5e.

>>50521223
18
>+1 to hit hits 18 20% of the time, 20 10% of the time, 22 5% of the time, 24 5% of the time
>+2 to hit hits 18 25% of the time, 20 15% of the time, 22 5% of the time, 24 5% of the time
>+3 to hit hits 18 30% of the time, 20 20% of the time, 22 10% of the time, 24 5% of the time
>+4 to hit hits 18 35% of the time, 20 25% of the time, 22 15% of the time, 24 5% of the time

Those areolas are perfect.

Also thus question is silly, the lower AC will always benefit more

>>50521223
a catgirl with leopard spot lingerie? Man I am legitimately shocked that this is the first time I seen such a thing. (cattgirls with leopard spots wearing leopard spots don't count, if anything it ruins any sense of contrast)

File: Vaultboy Thums Up.jpg (4KB, 179x122px) Image search: [Google]

4KB, 179x122px

>>50521223

18, simply because he is less powerful, and the bonus would benefit him more.

Also, OP, your image has inspired my worldbuilding, and its legacy will likely continue long after this thread has been archived.

>>50521971
lol very quick to remove the offending spells and resurrection abilities in your game anon. why so angry?

>>50521223

22, because moar powah is better!

> Also, source please?

>>50522187
Those're holes in those silky scraps there, Anon, not leopard prints.

>>50522273
Is she too poor to purchase some intact lingerie?

>>50522273
>>50522373

Hmm, perhaps its some sort of mesh, intended in its design. Judging by her ascetic, and dancing motions, I'd say she's some sort of exotic dancer in a vaguely Indian/Middle Eastern culture.

>>50522555
>>50522373
Those holes are intentional. They wouldn't form that way due to wear--typically, holes occur at the points where the garment is under the most strain. For a skirt, holes would be most likely to occur at the snuggest points around the hips. And there's the repeating pattern of larger holes near the points, surrounded by smaller holes.

I've had my share of raggedy clothing through the years, but that's not what's in that picture.

>>50522187
>>50522273
>>50522373
>>50522555
>>50522665
>we got from armor class to catgirls this fast

>>50521223
I suppose it depends on whether the enemy will be attacking at advantage or disadvantage more often.

If they're disadvantaged, the lower ac gets more help, if they're advantaged, the higher ac gets more help.

>>50522267
Not OP but the artist is soraname, she has a tumblr and a DA. Mostly draws FF14 OCs.

>>50522674
There's not much to be said on the topic OP's question.
Now catgirl lingerie AC, that's a different question. I don't think we can answer that without knowing the size of her bush though.

>>50522867
m8, there's been no concensus

>>50522897
Of course not. OP's not giving enough information to answer the question satisfactorily.

>>50522674
You can see her bush, that grabs my attention more than the OP's question.

>>50522909
see >>50521304
>>50521929

File: ke4dzSl.png (48KB, 1344x174px) Image search: [Google]

48KB, 1344x174px

>>50522674
It doesn't count when OP made the classic blunder

Dont want catgirl relatedd discussion in less then 20 posts? Don't post a catgirl in the opening, that simple.

>>50522273
ah you are right my good sir. But upon looking closerr at the lingerie.... for science of course, I noticed that the holes are all leopard spot shaped.

An absolute mad man is playing on my vulnerabilities

>>50522089
You need to be way out of your league for that to happen tho

File: 1443151033418.jpg (16KB, 184x184px) Image search: [Google]

16KB, 184x184px

>>50521223
>pubes
Good taste, and I'll likely be seeing you in 3

>>50523335
Too bad that artists also draws all their characters with huge nipples.

>>50523375
+2-1 is still +1.

>>50522773
It namae not name. You fuck, you had me looking at some brazillians shit art

>>50521223
AC is one of the few stats where you don't get diminishing returns, unless you're already 20 above your enemies' to hit bonuses.

For instance: suppose an enemy has a +5 bonus to hit and does 10 avg damage per attack. For the sake of convenience, let's ignore crits for a moment.

Vs. the 22 starting AC guy:
>with 22 AC, 20% hit chance times 10 avg damage = 2 dpr expected
>with 24 AC, 10% hit chance times 10 avg damage = 1 dpr expected
The +2 AC has reduced his damage taken by 50%.

Vs. the 18 AC guy
>18 AC, 40% hit times 10 damage = 4 dpr expected.
>20 AC, 30% hit times 10 damage = 3 dpr expected.
The +2 AC has reduced incoming damage by only 25%.

Usually this affect isn't as pronounced. But you should recognize that AC bonuses are basically worthless if you have terrible starting AC (negligible decline in average damage taken), and really really powerful if enemies already can barely hit you. But that's just theoretical.

In actual play it should be "whoever gets attacked most" or "whoever most frequently comes close to 0 hp".

>>50521971
fuck i hope you're not suggesting that 4e was good?

>>50523648
shouldn't the calculation be based on how many turns it takes to die?

>>50521223
It is horribly easy
It is convenient to give it to the 18 AC guy as long as your opponents have less than +3 attack.
Afterwards, it is always convenient the 22 AC guy.

>>50521223
What system?

>>50524369
5e

File: AC.jpg (53KB, 528x326px) Image search: [Google]

53KB, 528x326px

>>50524361

Is that a miqo'te?

>>50524412
huh, that's some interesting math going on there. Interesting as in, I am not sure who mathematically benefits the most anymore. At this point it all comes down to the situation I guess.

>>50528229
Correct

>>50524412
>>50524412
would it be possible for you to format this with some kind of visual aid to look across the entire lines?

File: PhoXo4.jpg (100KB, 626x427px) Image search: [Google]

100KB, 626x427px

>>50529309

>>50528229
>Interesting as in, I am not sure who mathematically benefits the most anymore.
For any attack, AC 22 either doesn't benefit from +2 or benefits more than AC18

>>50528229
Mathmaticaly, a +2 to AC helps all AC evenly because of the linear distabution of a single d20. Ignoring 20's always being hits and 1 always missing, then adding 2 to a to hit roll only matters if the attacker has the to hit bounses to have a chance to hit the defender at all. For instance, if an attacker does not have any bounses, adding 2 to a 20 AC would leave the defender impervious to that attack while adding 2 to 18 would subtract 10% from the fighters to hit only veing hit on the 5% chance of a 20. Meanwhile, a fighter with a +4 to hit would have his chances reduced by the same amount (10%) reguardless if the AC was 18 or 20.

>>50523298
Two words: Mind Flayers
If you don't already have a plan for getting your comrade out of grapple, a careless party can easily lose a member.

>>50523375
How huge are we talking about?

>>50531016
See
>>50521223

>>50530958
>Ignoring 20's always being hits and 1 always missing
But that's how it works, so your premise is flawed, and will lead you to a flawed conclusion.

> while adding 2 to 18 would subtract 10% from the fighters to hit only veing hit on the 5% chance of a 20. Meanwhile, a fighter with a +4 to hit would have his chances reduced by the same amount (10%) reguardless if the AC was 18 or 20

That ten percentage points, not ten percent, and so how big a part of the chance to hit that is varies.

+0 against AC18: 15% to hit.
+0 against AC20: 5% to hit.

So reducing the chance to hit by 10 percentage points drops it by two thirds, ~67%.

+4 against AC18: 35%
+4 against AC20: 25%

Here instead we reduce the chance to hit by less than a third, ~29%, despite subtracting the same ten percentage points.

>>50531140
But those are perfect for sucky-sycky

>>50530958
>>50530872
>>50531404
Guys, it's math. It's not like personal opinions are involved. This >>50529714 answers OP

>>50521223

It helps the guy with an 18 more, because it means the GM is less likely to throw an enemy who is easily capable of hitting an AC22 at the party (and thus who will EASILY hit the remainder of the party) for the purpose of challenging the character who's optimized for AC.

The DM can always throw higher attack bonuses at you. If the party all keeps their defenses around roughly the same values, they won't throw a challenge out there to target the skewed-high character. Keeping low-defense characters competitive with everyone else helps everybody.

>math doesn't matter
>the meta matters

>>50531404
While i did state that i would ignore nat20's and 1's, i did this to simplify things. If we did include nat20's and 1' then there is always a 5% chance that an attack will miss as well as a 5% chance that it will hit. This is fine, but it makes things weird when going to AC's above 20. In the case of an attacker without any bounuses, then anything at 20 or above has 5% reguardless of it's acual number, making the 18 the smarter choice. It get weird when including the constant 5% chance of failure as well. But it does not change the chamces when you start to shift the range of the attack by adding bounses.

>So reducing the chance to hit by 10 percentage points drops it down by two thirds, ~67%

>Here instead we reduce the chance to hit by less than a third, ~29%, despite subtracting the same ten percentage points.

You are not wrong, the percentage to hit do drop by those percentage points, porportionaly. But, a d20 has a discrete uniform probablity distribution. Every single side has a 1/20 chance to be rolled each time (assuming a fair die). Then the chance that a range would be rolled would be the sum of all of the probabilities in the range. So the range from [18,20] would be 15%, which we agree. When we change the 18 to a 20 this makes the porbability of rolling to hit just [20] making the probabality 5%. While the percant chance to has been reduced by 67%, that 67% is comparing the ratio of the 15% to the new 5%. Same with the plus 4. The rangre changes from the orginal [1,20] to [5,24] but each number still only has a 5% chance of landing. So the [18,24] would have a probabality of 35% and [20,24] would have a 25% chance. Again we agree on this point. But while the portiotion has dropped (10 is a smaller portion of 35 then 15) the bacic probablity of rolling to hit, landing the die in the range above the AC, is only reduced by 10% either way.

>>50531767
that is only for older editions of dnd

5e requires the math for advantaged rolls and disadvantaged rolls as well

>>50532114
>This is fine, but it makes things weird when going to AC's above 20

Not really, it just means you get >>50529714

>the bacic probablity of rolling to hit, landing the die in the range above the AC, is only reduced by 10% either way.

No, that's dead wrong. It's reduced by ten percentage POINTS. Those are very different things. The probability curve of the dice used has nothing whatsoever to do with it. If you can't keep the two apart, start expressing probabilities as fractions instead of percent, and never ever get yourself a mortgage.

>>50532003
A DM stupid enough to think along those lines would also be too stupid to be adjudicate monsters by factors beyond listed CR.

>>50533163

So cheating's OK with you? Glad you'll never be at my table.

>>50522139
at very high +to hits 20 gets a bigger benefit.

>+15 to hit hits 18 90% of the time, 20 80% of the time, 22 70% of the time
>+16 to hit hits 18 95% of the time, 20 85% of the time, 22 75% of the time
>+17 to hit hits 18 95% of the time, 20 90% of the time, 22 80% of the time
>+18 to hit hits 18 95% of the time, 20 95% of the time, 22 85% of the time