D&D 5E The tyranny of small numbers

[NotAYakk]

Small numbers add up.

A) 14 dex, +0 weapon, +0 arrows, +0 fighting style, level 7, against AC 18. Makes 20 attacks for 1d8+2 each; gets 1 crit, 7 other hits, deals 56.5 damage.

B) 20 dex, +1 weapon, +1 arrows, +2 fighting style, level 7, against AC 18. Makes 20 attacks for 1d8+7 each, gets 1 crit, 14 other hits, deals 178 damage.

+3 dex mod, 1 fighting style, 2 item upgrades; call this 6 "small number" boosts. Each granted a 20%ish boost, and compounded to 3x as effective.

The point isn't that 16 vs 14 makes you unplayable. The point is that 16 vs 14 at level 1 is the cheapest and easiest small number boost you'll get. And if you pick "naw I don't need it" when it is cheap and easy, you'll probably do it the next 5 times it is offered.

Because 20% isn't a huge gulf. But 1.2^6 is a huge gulf. And picking the 20% boost 6 times makes the PC in a different category than one who doesn't.

 

[overgeeked]

Nobody claims a character with a 14 in their primary stat is unplayable. What they do claim (correctly) is that it will have a significant impact on their effectiveness at their class role. You make a lot of rolls that are modified by your primary ability over the course of 12 levels, compared to having a 16. A 5% lower chance of success on all of them is a significant impediment. It’s not an interesting difference, and it won’t meaningfully affect how the character plays. It just makes them consistently a bit worse at doing exactly the same things.

I have had players rage quit because they didn't roll super-high stats. A character that was average was quote "utterly unplayable trash."

 

[Helldritch]

The problem is not mathematical. It is simply a matter of thrust.
The 14 stat is usually compensated by something your class would not offer. A dwarven proficiency with armor for example more than justify to have "only" a 14 for a much better armor class. But in the long run, that single +1 is all that is seen.
Again it is a matter of thrust.
Because as a DM what will I do? At some point, that character will find something that will give him/her that +1. And I will manage to give this +1 sooner than if the character had a 16 or 18. Contrary to other editions, magic is not mandatory to keep the pace, but it is a great equalizer! But some people do not thrust their DM to make sure that at some point, that missing +1 will become meaningless. And I will tell you a secret that all DM know. We will help the character that needs it over helping the optimized one. That +X great sword or whatever will be found a lot later when the character is optimized. Forcing the optimized one to rely on the magical spells of the poor unoptimized dwarven wizard. That thiefling strength base eldritch knight will find that +1 sword, that pair of gauntlet of ogre's strength way sooner that the optimized character. But to have these, you have to thrust your DM to do it.

And we all know that once level 12 is reached, the little +1 is no longer relevant as it is at this point that all characters have their 20 in their ability. But a campaing must reach this level. A lot do not. And since a lot do not, a lot of players do not thrust their DM to help.

 

[Maxperson]

Small numbers add up.

A) 14 dex, +0 weapon, +0 arrows, +0 fighting style, level 7, against AC 18. Makes 20 attacks for 1d8+2 each; gets 1 crit, 7 other hits, deals 56.5 damage.

B) 20 dex, +1 weapon, +1 arrows, +2 fighting style, level 7, against AC 18. Makes 20 attacks for 1d8+7 each, gets 1 crit, 14 other hits, deals 178 damage.

+3 dex mod, 1 fighting style, 2 item upgrades; call this 6 "small number" boosts. Each granted a 20%ish boost, and compounded to 3x as effective.

The point isn't that 16 vs 14 makes you unplayable. The point is that 16 vs 14 at level 1 is the cheapest and easiest small number boost you'll get. And if you pick "naw I don't need it" when it is cheap and easy, you'll probably do it the next 5 times it is offered.

Because 20% isn't a huge gulf. But 1.2^6 is a huge gulf. And picking the 20% boost 6 times makes the PC in a different category than one who doesn't.

I strongly disagree with your premise that if you don't want to take an ASI instead of a feat, that you will probably spontaneously ignore magic items and class abilities as a result. That's just.....................................wrong. They are not the same thing at all.

So ignoring the Red Herring extras you've thrown in, you have +3 from dex, which isn't nearly as impressive. Especially once you realize that the 20 attacks you are talking about are going to be spread out over 5-10 combats prior to extra attack and 2.5-5 combats after, and is going to amount to just 3 extra hits over that period of time and a bit of extra damage per hit. And that's with a 14 vs. 20. 14 vs 16 is even smaller and is pretty darn trivial.

 

[James Gasik]

Just for kicks, I went through the MM through the letter G looking at CR 1 or lower monsters. This was the breakdown of AC's

15 and under: 25(most of these were 13 or lower, many in the 9-11 range)
16: 2
17: 2
18: 1

I also noted that even at CR 8 and lower a huge number, perhaps even a majority, were at 15 or lower. You don't need to start at 18 or even 16 to hit most of the time in encounters.

A lot of people seemed overly impressed with the "The extra +1 is +351578% more damage!!" And ignore that it amounts to only a few extra points of damage per round, or a slightly less chance for the monster to succeed in the save, and maybe an extra use of an ability that doesn't do much.

Maybe the number of high AC enemies is low, but playing in AL, I watched Tier 1 players struggle against a CR 3 Knight with 18 AC, so even if that's an outlier, if you encounter it and also struggle, I don't see coming away with the idea that you need the best chance to hit possible as being strange.

 

[Asisreo]

Well, both are true. A 10% chance of success is 5% more than a 5% chance of success, and is also twice as likely to occur.

It's a bit iffy how it can be worded.

And it can be hard finding the intuition for it if you're not used to statistical probabilities. The power of a +1 is far more determined by the frequency and difficulty of the challenges you'll face than anything else.

If you're facing DC 5 challenges frequently, a +1 hardly matters. However, if you're frequently facing DC 20 challenges, every bonus counts.

So, as with almost all other things in this game, it's party and DM-dependent. You can't math yourself out of a sociological event, at least not without sufficient datasets, which we do not have.

 

[overgeeked]

"Optimizing the fun out of the game" is not what's going on here.

The principle driving this is a thing called Loss Aversion Bias: People hate a loss about twice as much as they like a win. Meaning you have to win (in this case land a hit) about 67% of the time to feel like you are actually winning. And lo and behold, the 5e devs thought of this, if you dig into the DMG tables, a character with a starting 16 who pumps the main stat with ASIs will hit a generic monster 65% of the time, close enough to "feel right", even more so when you consider situational bonuses.

That's the smoke and mirrors bit that I've talked about a few times. I didn't realize there was an actual name for it though. On-level monsters in 5E keep a fairly precise math progression that hits this range almost perfectly. I wonder if that will pull the veil from some players' eyes.

While you might not think a mere 5% loss in accuracy that stems from starting with a 14 instead of a 16 would matter, it is enough to change that 65% into a 60%, going from skirting the "this feels good" squarely into the "this feels bad" side of the equation.

And yes, obviously not all people are as susceptible to Loss Aversion as others, which is why that 60% feels fine for them.

So to push this theory into practical terms, no matter what else is going on with the character or the sheet or the build, players feel about right when they succeed with a 8+/d20. And the real range will be 55-75% (just for variety's sake), so 6-10+/d20.

Making it even simpler, 3+/1d6. That's the sweet spot right there. You could modify that with +/-1 or bonus/penalty dice.

Anydice tells me that with a +1 that chances 3+/1d6 to 83.33% and with a -1 that chances 3+/1d6 to 50%.

Anydice tells me that with one bonus die that changes a 3+/1d6 to 88.89% and with one penalty die that changes 3+/1d6 to 44.4%.

 

[Maxperson]

Maybe the number of high AC enemies is low, but playing in AL, I watched Tier 1 players struggle against a CR 3 Knight with 18 AC, so even if that's an outlier, if you encounter it and also struggle, I don't see coming away with the idea that you need the best chance to hit possible as being strange.

I'm okay with that. Struggling now and then against a hard to hit enemy is okay. And it switches things up. Suddenly the players are faced with an enemy that they may need new strategies and/or tactics to fight effectively.

Adding a miniscule amount of increase to your chances to hit isn't going to be noticeable. Consider this. If you needed a 13 to hit that AC of 18, you were going to miss one a roll of 1-12 and hit on 13-20. Now if you add +1 by raising that stat, you still miss on 1-11 and you would hit anyway without the extra bonus on 13-20. It literally only helps you on one roll out of 20. If you don't roll that exact number of 12, the bonus meant squat.

Rolling that exact number will happen on average one time out of twenty attacks, which will take 5-10(2-4 round combat length) combats to see. Are you really going to even notice that you hit a single extra time over a 5-10 combat period of game play? I strongly doubt it. And that one hit is unlikely to make any meaningful difference.

 

[overgeeked]

It really depends where a DM places success. Most tasks should be around dc 10 but DMs ramp up DCs to make things ‘interesting’. All this does is force players to boost numbers to succeed or use approaches that favour their apex stat.

I'd say it more depends on how often the referee requires you to roll. If they auto-succeed a lot of things, it doesn't matter as much, but...importantly...the players seem to not notice those as much. Rolling dice raises the tension. The possibility of failure raises tension. So a lot of referees make players roll way too much for incredibly petty things. It's a terrible habit. But, because it's such a common habit...over time the numbers keep going up on character sheets and the numbers behind the screen keep going down. To me, unless there's something to be gained or lost, there's no point in rolling. If there are no real stakes, or no interesting stakes, don't roll. If the DC is crazy high or crazy low, there's no point. Just auto-fail or auto-succeed the check.

DMs forget that an average commoner has a 9 or 10 in all their stats. A 14 is an exceptional attribute. A 20 is the peak of mortal ability. It’s weird to have all these people with 20s in stats looking down their noses at people with ‘mere’ 16s.

I'd say it's the players who forget this. To a lot of players 14 is too low. Players, as described in this thread and as I've had at my table, treat anything less than the absolute max they can possibly have in a stat or a skill or a check as trash. It's a weird binary of "perfect or suck."

 

[James Gasik]

I'm okay with that. Struggling now and then against a hard to hit enemy is okay. And it switches things up. Suddenly the players are faced with an enemy that they may need new strategies and/or tactics to fight effectively.

Adding a miniscule amount of increase to your chances to hit isn't going to be noticeable. Consider this. If you needed a 13 to hit that AC of 18, you were going to miss one a roll of 1-12 and hit on 13-20. Now if you add +1 by raising that stat, you still miss on 1-11 and you would hit anyway without the extra bonus on 13-20. It literally only helps you on one roll out of 20. If you don't roll that exact number of 12, the bonus meant squat.

Rolling that exact number will happen on average one time out of twenty attacks, which will take 5-10(2-4 round combat length) combats to see. Are you really going to even notice that you hit a single extra time over a 5-10 combat period of game play? I strongly doubt it. And that one hit is unlikely to make any meaningful difference.

I know the numbers are small, but I think you also have to look at what all they are affecting. If I'm a Rogue, and I only ever get one attack a turn (without two weapon fighting) and everything rides on that attack, AND I get so many benefits from Dexterity, what would be better than raising it anyways?

I have had players rage quit because they didn't roll super-high stats. A character that was average was quote "utterly unplayable trash."

I had this happen in a 2e game. I started running for new players, from another group. One player rolled a stat line with not one, but two 17's, stared at it, and said "I can't do anything with this".

I was blown away. Especially since he claimed to want to play a Specialty Priest of Waukeen, so I couldn't imagine what else he needed!

Different ways of playing the game, I guess.

 

[overgeeked]

I have felt for a while, that ability scores do not make sense if they're viewed like most of this board does.

If every character of given class is expected to have the same main ability score or constitution score at a given level, why is it a choice in the first place? If the class effectively dictates your ability scores, why are we pretending they're a choice, why are we pretending they represent individual variation? In such a paradigm they can't. It would be far more logical to get rid of ability scores then, and just have the classes provide the expected level appropriate bonuses.

It is not exactly what I would want, but it would be far more honest than pretending there is a choice and that they represent different characters being different, if in practice your class effectively dictates the score placement.

So seriously stop to think: what is the purpose of ability scores and what do they represent?

I definitely see your point. It would be easier, and more honest, to ditch ability scores and skills (because they're basically a non-choice as well), and simply give classes advantage in the things they're expected to do. It's the equivalent to a +5 so it covers a lot of ground. Keep abilities as a thing and give each class two they get advantage in. That covers ability scores, skills, and saves. Fighters are proficient in STR or DEX and CON, go.

 

[Maxperson]

I know the numbers are small, but I think you also have to look at what all they are affecting. If I'm a Rogue, and I only ever get one attack a turn (without two weapon fighting) and everything rides on that attack, AND I get so many benefits from Dexterity, what would be better than raising it anyways?

To me?

Alert
Athelete
Actor
Dual Wielder
Dungeon Delver
Keen Mind
Lucky
Mage Slayer
Magic Initiate
Martial Adept
Mobile
Observant
Sentinel
Tavern Brawler
Skilled
Sharpshooter
Skulker

Those are just from the PHB and just for rogue characters. I'd be able to list several more from other books, and would have slightly different picks if picking for other classes.

 

[Charlaquin]

I have had players rage quit because they didn't roll super-high stats. A character that was average was quote "utterly unplayable trash."

Yeah, I mean if that’s the players’ attitudes, probably best to do point buy.

 

[AbdulAlhazred]

I have felt for a while, that ability scores do not make sense if they're viewed like most of this board does.

If every character of given class is expected to have the same main ability score or constitution score at a given level, why is it a choice in the first place? If the class effectively dictates your ability scores, why are we pretending they're a choice, why are we pretending they represent individual variation? In such a paradigm they can't. It would be far more logical to get rid of ability scores then, and just have the classes provide the expected level appropriate bonuses.

It is not exactly what I would want, but it would be far more honest than pretending there is a choice and that they represent different characters being different, if in practice your class effectively dictates the score placement.

So seriously stop to think: what is the purpose of ability scores and what do they represent?

5e has rules for rolling stats using dice, right? In that case the player rolls their dice, and then decides what to make of the resulting character. If you want to really get the feel of "hey, your ability scores determine your fate" then require rolling them in order! IME most players are OK with that, at least some of the time. Usually once they've played with a system for a while though, they will want to try out specific builds and whatnot, and then this process starts to get in the way of the fun. There can be a kind of lingering sense of frustration as well when your character overall has sub-par scores other players are forever wielding an advantage due to a single toss of dice that happened 9 months ago.

In the end you're unlikely to end up with lots of fighters with mediocre physical attributes and such anyway. Notice a few little-mentioned but key points in the design of the original D&D game: There are 'disqualifier' numbers, like if your STR is 5 or less your character MUST be a Magic User in 1e, 5 INT also limits you to Fighter, etc. These, and in OD&D the 'trade points' rule, insures that there are some PCs with things like prime requisites of 12 and such (OD&D's 3d6 in order also kinda guarantees that as well, though people usually toss really pointless ability score sets). Those sorts of things won't come up in 5e, unless you intro some really draconian chargen rules.

When I wrote my own game, I got rid of the notion of 'ability score' and there are just bonuses. While they are not technically tied to specific classes, it usually makes sense to construct certain classic styles of build. OTOH you COULD construct a fighter that pretty much exclusively uses INT and CHA. It will just be a weird character, and requires some unusual build choices (also there isn't really an INT melee weapon, lol, though you could invent an INT based fighting style which is going to basically do the same thing).

 

[Asisreo]

I have had players rage quit because they didn't roll super-high stats. A character that was average was quote "utterly unplayable trash

I'd have them removed from the game anyways. If you think on it, that's not only how they'd view their own characters, but that's how they'd view others' characters on their team.

So the guy playing an average fighter character would probably be "utter garbage" to that person and that's not the type of attitude I'd want in my games.

 

[overgeeked]

Yeah, I mean if that’s the player’s’ attitudes, probably best to do point buy.

Funnily enough it worked as intended. It's one of the filters I use to screen out certain kinds of players.

I'd have them removed from the game anyways. If you think on it, that's not only how they'd view their own characters, but that's how they'd view others' characters on their team.

So the guy playing an average fighter character would probably be "utter garbage" to that person and that's not the type of attitude I'd want in my games.

Exactly. I insist on rolled stats to find and filter out players with terrible attitudes.

 

[MoonSong]

Exactly. I insist on rolled stats to find and filter out players with terrible attitudes.

It depends, can I choose to lower a score that I find too high?

 

[tetrasodium]

I have felt for a while, that ability scores do not make sense if they're viewed like most of this board does.

If every character of given class is expected to have the same main ability score or constitution score at a given level, why is it a choice in the first place? If the class effectively dictates your ability scores, why are we pretending they're a choice, why are we pretending they represent individual variation? In such a paradigm they can't. It would be far more logical to get rid of ability scores then, and just have the classes provide the expected level appropriate bonuses.

It is not exactly what I would want, but it would be far more honest than pretending there is a choice and that they represent different characters being different, if in practice your class effectively dictates the score placement.

So seriously stop to think: what is the purpose of ability scores and what do they represent?

I don't think the problem is trhe fact that there is a choice so much that the choices are meaningless. Take the 3d6 or 4d6 distribution on pg33 of the tome of variance as an example, suddenly you have serious changes being made & the choices are going to result in viscerally different results rather than just the same results but differently arranged. Try a few minmaxed SAD vrs jack of all but master of none type builds & there are serious choices even into things like "how do I want to handle dump stat(s)"

 

[overgeeked]

It depends, can I choose to lower a score that I find too high?

I dunno. It’s literally never happened in almost 40 years of playing D&D or D&D-like games.

 

[Warpiglet-7]

I assume you're referring to me. I certainly take statements like:
"Are we telling people they must have a +1 sword or the character is doomed?"
and
"There are lots of other ways to build one…if you can tolerate a slightly smaller bonus here or there."

as being pretty hyperbolic. The former statement openly implies that even a single deviation results in doom and gloom proclamations from anyone interested in optimization. That is both unfair and inaccurate in most cases. The second likewise implies that optimizers find it intolerable to ever have "a slightly smaller bonus here or there." That is likewise both unfair and inaccurate in most cases.

Do you deny that these paint essentially all optimizers with a broad brush of unfair and inaccurate assertions?

I stand by what I said as it applies to the people I am referencing. If there are people that are doing exactly what I said, I think they are giving bad guidance in most instances.