D&D 5E Using 3d6 for skill checks

[ProphetSword]

It is interesting, don't really like the automatic success part for some reason. Will think about it. It feels very different to me (breaking the check down into parts) and I don't feel it would work well for our combat, but it is interesting. It could work especially well in situations where partial success makes sense. Rather than having multiple checks, they could be combined to some degree.

This isn't intended for combat. Just skill checks.

 

[Ovinomancer]

I don't think I agree. If we were talking about a game without bonuses, then yes. D&D has bonuses that can stack up pretty quickly, exceeding the standard deviation of the bell curve. If you're already above the hump, it's not a big deal. If you're below, though, you suddenly get even more bang out of each bonus.

I could be wrong. I really don't have the time to run a full statistical analysis, even though I think it'd be fascinating.

What I suspect would happen is that the PCs would hit even more often than they do now -- monsters are generally balanced by hit points, not AC. Meanwhile, monsters would miss more often -- even my casual players have annoyingly high armor classes. The net is to reduce the difficulty of medium to easy foes and (potentially) increase the difficulty of hard and extreme foes, throwing the idea of bounded accuracy out the window.

Also, close to 50% of all possibilities will come up with four results (9, 10, 11, 12) and almost 70% fit into six results (8, 9, 10, 11, 12, 13). For two 1st level characters, one with proficiency and one without, you end up with the sort of "skilled characters are skilled" result you presumably want. By the time you hit 10th level, proficiency ends up being a +35% bonus. That assumes a "fair" challenge is balanced at an average roll succeeding, whether you center it on proficient or non-proficient attempts. If you center it between the two, then you get the full peak of the bell curve and the bonus goes up. You might as well say "You succeed if you're proficient". There's no point in Expertise. If you start including easy modifiers, like bless, enhance ability, magic weapons, etc. things get worse.

In retrospect, what I really meant by "swingy" wasn't so much random in the specific sense. It was more in the sense that any given challenge is likely to be either auto-success or auto-fail. Again, I haven't actually run the numbers. If your play experience is different, that's cool. I just have a very, very bad gut reaction to applying 3d6 to D&D. I think 2d10 might be a shallow enough curve to split the difference, but I'm not sure.

That's a good point, that the 3d6 really does hurt the unskilled and can quickly regulate even mid teens DCs to the near impossible for untrained, 8 stat characters. Similarly, it makes high bonus characters nearly unable to succeed except at high DCs, which it still keeps as pretty unlikely: a 20 stat, expertised 9th level character (total +13) can hit an impossible DC only on a 17+, so a bit more than a 2% chance, meanwhile, he'll rarely miss a DC 20.

So, if that's your definition of swingy, that 3d6 can bin you into auto-success/auto-fail faster than d20 can, then, yep, no arguments. Odd definition of swingy, though.

 

[Horwath]

How about;

If you're untrained in a skill, you roll d20.

If you have, training roll 2d10

if you have, expertise, roll 3d6


basically, more training you have, more consistent your ability is, in addition of bonuses to the die(dice) roll.

As for advantage/disadvantage roll one extra die for 2d10 and two dice for 3d6, dropping one/two of the lowest/highest.

 

[Psikerlord#]

Like many DMs, I use a lot of house rules in my games. However, recently I was a player in a game that used the RAW for everything. Most of my house rules I was happy to live without, but the one house rule of mine that I constantly yearned for in the new game was using 3d6 for skills checks.

3d6 gives the same average (10.5), but the results are clustered much more heavily in the middle range. This means that characters will more reliably succeed on the things they’re good at and fail at the things they’re not. It also means that even small differences in scores are significant. Using 1d20 for skill checks, especially at low levels, often leads to frail wizards successfully out-muscling fighters and dim-witted barbarians out-Arcana-ing wizards. With 3d6, those things can still happen, but they’re much less likely.

When I was playing the RAW game, I found myself choosing not to attempt things my character wasn’t good at, e.g. kicking down a door with 8 Strength, even though my character had a reasonable chance of success and the fighter had tanked a few rolls, because I didn’t want to steal his spotlight by encroaching on his schtick. If we’d used 3d6, my character would have been almost guaranteed to fail and the fighter more likely to succeed, which is how I think it should be.

Have any other folk tried 3d6 for skill checks or heard of anyone else who has? We’ve only played in the low-level zone, so I’d be interested to hear if anyone has tried it at higher levels and had any success or discovered any issues.

I quite like this idea. Another alternative, which we are currently using, is getting rid of DCs altogether, and just making all skill/ability checks roll equal or under, ala OSR. So if you have an 18 Str fighter bashing down a door, he has a very very good chance of doing it, and an 8 Str guy, not so much, but he still could.

Of course all the usual mods on top of that - inc adv/disad - albeit we impose an 18 max cap.

 

[Li Shenron]

Have any other folk tried 3d6 for skill checks or heard of anyone else who has? We’ve only played in the low-level zone, so I’d be interested to hear if anyone has tried it at higher levels and had any success or discovered any issues.

I completely agree with your premise. Swinginess doesn't matter for attack rolls because most of them aren't individually crucial for success/failure on the whole combat, and instead it's rather the combination of many attack rolls that determine if you win or lose. But it does matter for skills because a lot more often one check determines a significant outcome rather than attrition.

As a matter of fact, I like swinginess and randomness, but what I don't like is that because of swinginess, the 'expert' at a task is too often surclassed by someone else.

I haven't house ruled skill checks yet, but if I do, I'd use d8 + d12: less dramatic change in probability distribution than 3d6, you can still get a result up to 20, and more uses for the neglected d8 and d12 dice

How about;

If you're untrained in a skill, you roll d20.

If you have, training roll 2d10

if you have, expertise, roll 3d6

Not good, I think this means the more training you have, the higher your chances of success if the DC is lower than you average, but at the same time the lower your chances of success if the DC is higher than your average. So for example if you need a result of 18+, an untrained has 15% chance of success, a trained has 6% chance, an expert has 1/216 = less than 0.5% chance!

 

[EzekielRaiden]

Not good, I think this means the more training you have, the higher your chances of success if the DC is lower than you average, but at the same time the lower your chances of success if the DC is higher than your average. So for example if you need a result of 18+, an untrained has 15% chance of success, a trained has 6% chance, an expert has 1/216 = less than 0.5% chance!

Correct. For those wishing to see a comparison, check out this Anydice program. You become substantially worse at the top end of skills, even if you do 3d6+2 (so that it can still hit 20)--in fact, with 3d6+2, you have a lower chance of rolling anywhere in 18, 19, or 20 than you have of just critting on d20. It's a bad trade-off.

Possible way to fix it: Make it "best 2 of 3d10" for Trained and "best 3 of 5d6 + 2" for Experts. You still take a minor hit to the highest heights--the untrained can occasionally best the master through having no preconceived notions--but that is rare, and it's a lot more likely that the untrained will do more poorly than an expert ever could (Experts can't roll less than 5, meaning a full 20% of the Untrained's distribution doesn't exist on the Expert's.) Incidentally, these also equate almost perfectly to +3 for Trained, +5 for Expert (this includes the Trained bonus). The averages are 10.5 (1d20), 13.48 (best 3 of 2d10), and 15.43 (best 3 of 5d6+2).

It also means people will be able to feel the difference between their skills. You roll 1 die for untrained skills, 3 dice for trained skills, and 5 dice for expert skills--that's a clear, palpable difference.

 

[CapnZapp]

The reasons against 3d6 have already been given.

Basically, in D&D EVERYONE is a hero. There's no such thing as some heroes "realistically" failing at tasks. All heroes have a decent shot at success.

The basic mechanism that ensures this is the d20, or rather, the use of dice with a decent chance of rolling really well.

Lots of die rolls that doesn't really upset expectations isn't D&D's thing.

I think the drive behind 3d6 can better be summed up as "don't roll when you already have a result you want".

That is, any time it doesn't matter (the stakes aren't high, the drama isn't peaking) then... don't use 3d6 but instead just narrate the results: "the strong fighter succeeds while the frail wizard fails"

But in those "rare" circumstances where a hero really could make a difference, you roll, and you use a d20.

 

[Li Shenron]

Lots of die rolls that doesn't really upset expectations isn't D&D's thing.

Ah but here is the crux of the matter

The problem which I (and the OP) have is with skills, and has a lot to do with the fact that "anyone can try anything". The swinginess of d20 in skill checks is good when it makes you unexpectedly ("heroically") succeed, but it also works the other way around. You might think that it is fun, for the hero to occasionally "heroically" fail. The problems are:

- excessive swinginess can make it happen too often
- even when it happens, another PC can pick up the task, and negate the failure (this depends on the skill, of course)

Dialing on the DC doesn't completely fix the problem, because if you make it easier for the trained character, you'll also make it easier for the untrained, and viceversa. It's like pulling the blanket either left or right, but the other side gets cold.

Just think of this: you wrote "Lots of die rolls that doesn't really upset expectations isn't D&D's thing.", but the problem is that whenever the expert fails, everybody else can try (or even the same person can try again) and practically every time you do get lots of die rolls which in fact make sure the expectations aren't upset.

Using different dice doesn't fix this problem for good, but at least it makes the proficiency bonus matter a bit more, so this is already an improvement to merely adjusting the DC.

5e already has a couple of mechanics to help: expertise and the Rogue's ability to default to a minimum skill check result, plus group skill checks.

At the same time, an inherent problem is in the skill system covering wildly different things, most notably both stuff that everybody must check for (e.g. the whole party rolling checks to avoid falling or to sneak) and stuff that require a single person's success to work (e.g. knowledge, search, lockpicking), which work fundamentally different.

 

[Horwath]

Not good, I think this means the more training you have, the higher your chances of success if the DC is lower than you average, but at the same time the lower your chances of success if the DC is higher than your average. So for example if you need a result of 18+, an untrained has 15% chance of success, a trained has 6% chance, an expert has 1/216 = less than 0.5% chance!

Not really as untrained has 1d20, trained has 2d10+2 and expertise has 3d6+4. that are minimum bonuses.

 

[CapnZapp]

Ah but here is the crux of the matter

The problem which I (and the OP) have is with skills, and has a lot to do with the fact that "anyone can try anything". The swinginess of d20 in skill checks is good when it makes you unexpectedly ("heroically") succeed, but it also works the other way around. You might think that it is fun, for the hero to occasionally "heroically" fail.

If you don't find it fun or realistic for the hero to fail, why roll at all? Clearly he's doing something he's supposed to handle, så simply narrate the success and move on.

Look, I understand - I've been playing skill based rpgs my entire life.

But D&D isn't about the skills. It's a poor engine to base the kind of everyday verisimilitude you're after.

That's why I'm coming up with the alternatives. Basically every case of "ordinary" challenge (the ones where we would have an instinctual opinion on success, where the mundane strength difference between a burly big man and a weak scholar would matter)... basically isn't the focus of D&D.

Take 10 or roll or don't roll, it's just not important, so make an off the cuff decision and move on.

Then, when it comes to the cinematic larger than life things, where all heroes get a shot at success simply because they're protagonists (and heroic ones at that) THAT is when you pick up the dice.

And that die happens to be a d20, to ensure the die roll truly matters: that every hero can fail and every hero can succeed! Sure, the fighter might succeed more often, but it's not like in real life where one person would realistically always win and another person always lose.

Cheers and good luck with your game!