D&D 5E Feats, class balance and fun

[Barolo]

Oh, for another similar method, I started my gnome battlemaster with the default array, but randomly determined which ability got each number. And randomly choose a race (and subrace), and class (and subclass).

The high Dex/ low Strength immediately said swashbuckler to me. And I went right ahead builing it as a berserker barbarian . . . right up until I noticed the barbarian simply does not work with Dex.

So I built up the gnome swahbuckler concept as a fighter, as an arcane trickster, as a beastmaster ranger, as an abjuration wizard (I have a 14 Intelligence), then merged them all into the battlemaster as well as could be managed.

Wow, this is a really different way for building a character. I have never picked a class randomly or a race. I may have some weird paths to go on my train of thought, though. Last time I was planning to play a sorcerer inspired by the lines "power overwhelming" I started with a somewhat scholarly concept, focused on planar knowledge and ultimate magic understanding, and was trying to use the half-elf to salvage my more-or-less poorly rolled stats, went through the idea of changing the class to abjurer wizard in a point, but ended up playing a front-line sorcerer half-orc devoted to gruumsh (but who would have been considered heretic by any orc shaman), with completely different spell selection, skill selection, stat distribution, subclass, everything.

 

[Satyrn]

Wow, this is a really different way for building a character. I have never picked a class randomly or a race.

I recommend trying it. I doubt I'd have ever chosen forest gnome for a fighter character (maybe as an eldritch knight), but I've since fallen in love with it. The advantage on Int, Wis and Cha saves against magic is great protection against being charmed or dominated - and my table finds there to be nothing more terrifying than being attacked the party's own tank.

And I built his personality around the forest gnome's ability tp speak with critters. He has a pet owl, and rides a dog into battle. Squirrels keep watch when he camps alone. He dubs his guards as woodland squires in gently mockery of the noble party members. The owl is named Lord Skeddadle, the dog is Sir Litmus, both names chosen with a random generator. Great fun. Though I did groan when I realized I was inadvertently referencing Harry Potter by having the owl deliver messages.

 

[Flamestrike]

There is a middle ground between 'feats' and 'no feats'.

Allow feats, but only allow 'half' feats.

This takes the main offenders out of the picture (GWM, PAM, Sharpshooter).

Im working on a redux at the moment that converts all feats into half feats for just this purpose.

 

[Barolo]

There is a middle ground between 'feats' and 'no feats'.

Allow feats, but only allow 'half' feats.

This takes the main offenders out of the picture (GWM, PAM, Sharpshooter).

Im working on a redux at the moment that converts all feats into half feats for just this purpose.

Just a remark, it is not on every table that those "offender" feats wreak havoc. In mine, I don't know if my players that use said feats are downright incompetent or if the other players are tactical geniuses, but I can't discern any problematic spotlight imbalance during combat or even overall performance between those who chose to develop their PCs "combat-heavy" with either SS or PAM+GWM and others that just took a couple of ABIs or the eventual actor feat.

 

[Bacon Bits]

Well, at higher levels they do get to use these feats more often, but surely it is not always on.

On the contrary, that's precisely what I mean.

Take a moderately optimized Fighter at level 10-12 and calculate the average damage you deal on a hit both with and without power attack. Then multiply those values by the percentage chance of hitting the target respective to whether or not you're using power attack. This is your average damage per attack and it accounts for the reduced hit rate. The value for power attack is greater for all AC values from roughly AC 20 and below. That's what "it's mathematically correct to use power attack on any creature with AC 20 or less" means. Yes, you might miss more often, but the amount of damage you gain when you hit makes up for it. You're no longer gambling on rolling well and hoping that luck results in the extra damage taking your target down before your attack rolls regress to the mean. You do enough extra damage that in the average case you deal more overall damage in spite of the misses from using power attack. When you get to these levels, power attacking is the safer bet.

I'm playing a level 15 Fighter with Great Weapon Mastery right now. I haven't not used power attack since about level 11 when I found a +2 maul. In another campaign we have a Fighter with Sharpshooter. He's had the same experience, if not more so because bows deal less damage than heavy weapons and the Archery style offsets the attack penalty from power attack.

You want to know when to use -5/+10? It's easy: (2 x Attack Bonus - Average Damage per Hit + 32) / 2 rounded down is the maximum AC your character should power attack against. By the time you get to level 10, the result of that formula is going to be 18+ for a GWM character, or 21+ for a Sharpshooter. The only way it won't be is if you manage to get a flame tongue greatsword or some similar item that adds to damage without adding to attack bonus.

The closest thing to a defensive reaction that I see monsters and NPCs possessing is "thorns" damage where an attacker takes automatic damage on a hit, such as a Fire Elemental, Black Pudding, or Barbed Devil. That type of defense favors power attack, since you need fewer hits to deal lethal damage. Excepting spellcasters, I'm not aware of any monsters that have reaction abilities beyond opportunity attacks. So I'm not sure what you're talking about here. And, sure, if you have disadvantage, you don't use it, but that is not that common. It comes up during a few rounds in a limited number of encounters. If it's coming up every round or even every combat, then your DM is running an unusual game. In other words, your DM is "fixing" the problem by changing the playing field, which is like "fixing" fireball by making every enemy resistant to fire.

 

[AaronOfBarbaria]

You want to know when to use -5/+10? It's easy: (2 x Attack Bonus - Average Damage per Hit + 32) / 2 rounded down is the maximum AC your character should power attack against.

Can you unpack this a bit? I tried to use your formula and I came out with a result that I know can't be right (+11 attack bonus, 12.35 average damage per hit, came out to maximum AC to power attack against of 11).

 

[Valdier]

From the order of operations... this would be AC 20.825

11x2 = 22 -12.35 = 9.65 +32 = 41.65 / 2 = 20.825

 

[Bacon Bits]

Can you unpack this a bit? I tried to use your formula and I came out with a result that I know can't be right (+11 attack bonus, 12.35 average damage per hit, came out to maximum AC to power attack against of 11).

Absolutely.

First, your math is wrong:

(2 x Attack Bonus - Average Damage per Hit + 32) / 2

Attack Bonus = 11
Average damage per hit = 12.35

(2 * 11 - 12.35 + 32) / 2 = (22 - 12.35 + 32) / 2 = 41.65 / 2 = 20.825 --> AC 20

Don't forget your PEDMAS!



Second, where does the formula come from?

Step 1. Define the chance to hit.

(21 + Attack Bonus - Target AC) / 20

We use 21 instead of 20 because you hit on a tie. If I have a +11 attack bonus and I roll an 11, I hit AC 22, right? And 11-20 is 50% of a d20. So we'd expect this formula to result in 0.50 with those values:

(21 + 11 - 22) / 20 = 0.50

And so it is.


Step 2. Define the average damage per attack prorated for the chance to hit.

Average Damage on Hit * (21 + Attack Bonus - Target AC) / 20

So, if you deal 12.35 damage per hit, have a +11 attack bonus, and are attacking a target with AC 22:

12.35 * (21 + 11 - 22) / 20 = 12.35 * 0.50 = 6.175

Step 3. Define the prorated average damage of power attacking using the terms we have when we don't power attack.

In other words, Attack Bonus is still 11 and Average Damage is still 12.35.

(Average Damage on Hit + 10) * (21 + Attack Bonus - 5 - Target AC) / 20

Or, more simply:

(Average Damage on Hit + 10) * (16 + Attack Bonus - Target AC) / 20

So you were power attacking, you'd do this:

(12.35 + 10) * (16 + 11 - 22) / 20 = 22.35 * 0.25 = 5.5875

Step 4. Put Step 2 and Step 3 together. We want to know when the prorated average damage for power attacking (Step 3) is greater than the prorated average damage normally (Step 2):

(Average Damage on Hit + 10) * (16 + Attack Bonus - Target AC) / 20 > Average Damage on Hit * (21 + Attack Bonus - Target AC) / 20

Step 5. Solve for Target AC.

Now, I don't know about you, but I'm really freaking lazy. I don't want to do algebra myself.

Let x = Target AC
Let a = Attack Bonus
Let d = Average damage on hit

(d + 10) * (16 + a - x) / 20 > d * (21 + a - x) / 20

Now, Wolfram Alpha will solve it for us. Follow that link and scroll down to "Solution," and you will see:

x < 1/2 * (2a - d + 32)

Or:

Target AC < 1/2 * (2 * Attack bonus - Average damage on hit + 32)

So, this value gives us the point at which above formula changes:

1/2 * (2 * Attack bonus - Average damage on hit + 32)

That's likely to be a fractional value, however, so we can round down and be confident that for AC will benefit power attack:

Maximum AC for power attack = Floor [ 1/2 * (2 * Attack bonus - Average damage on hit + 32) ]

 

[AaronOfBarbaria]

Absolutely.

Thanks.

First, your math is wrong:

I laughed out loud when reading this, and said "Well... duh." even though no one was around to hear me. So thanks for that too.

Don't forget your PEDMAS!

My error actually ended up being trying too hard to remember; I was doing all the addition before the subtraction, because I remembered the acronym and what each letter stood for, but not how to actually act upon that information correctly.

Thanks for breaking down the whole process, too. It's kind of cool to see all the steps, and to find out that the advice I'd been giving my players (which was "Probably not going to take your opponent down faster by taking the -5 to hit if you need a roll of 11+ on the die to hit without the -5") wasn't too far off from what math says.

Edit to add: and thanks to you too, [MENTION=4556]Valdier[/MENTION]. Didn't notice you'd also helped me with my math before I originally posted this.

 

[FormerlyHemlock]

It's only a problem when you do the math, and realize that by level 10 or so that it's mathematically correct to use the ability when attacking anything below AC 20. When you combine that with the knowledge that essentially everything has AC 20 or less, you see the issue. Feats that you use every round shouldn't be very powerful. Feats that are very powerful shouldn't come up every round.

Why?

Off the top of my head, here's four things you shouldn't try Sharpshooter against:

(1) Spellcasting dragons w/ Shield spell (AC 24)
(2) Drow Elite Warriors shooting at you from beyond your darkvision range (AC 18 + disadvantage)
(3) Nycaloths inside of Darkness spells (AC 18 + disadvantage)
(4) Almost anything that isn't a mook while you're being constricted by a giant snake/toad/octopus or caught in a Web spell or whatever (typically AC 15-18 + disadvantage)

The decision point for Sharpshooter doesn't always come into play, but sometimes it does (GWM even moreso), and PCs now have an additional reason to jockey for advantage over disadvantage. And that adds interest to the fight (at least for me as a DM!).

Forcing players to make decisions is what make creating scenarios and running scenarios interesting. Teaching them to "solve" the puzzle of how to deal with drow in the dark is fun.