Difficulty Classes - appropriate challenge?

[hamurai]

So, I've been wondering about Difficulty Classes (DC) in 5E ever since our DM announced a check was average (DC 15) and the entire party of 6 failed.

For the uninitiated, the DC is the number you have to beat with a d20 + bonuses (attribute bonus and, for some, the proficiency bonus).

The Player's Handbook lists the following DC's:

Task Difficulty	DC
Very Easy	5
Easy	10
Medium	15
Hard	20
very Hard	25
Nearly Impossible	30

The usual attribute bonuses are around +1 or +2, so the non-proficient folks will mostly roll a d20+1 or d20+2 to beat 15. The proficient character, say they have even great attributes, rolls d20+5 on levels 1-4. 50% chance for an average check.

So, which characters are those DCs scaled to? It seems to me, always the proficient ones with great attributes. 

"Hey wizard, you need a Strength (Athletics) check to jump the chasm before the fire engulfs you. Don't worry, it's an average check." (Evil smile) - Wizard rolls d20+0...

Opinions?

[Finarvyn]

I look at it this way -- If you think about a standard d20 roll the average is 10.5, so let's say a 10. In 5E the minimum proficiency bonus is +2 so that moves our average to 12. A decent stat bonus is +1 or +2, so that moves our average to somewhere around 13-14. This means that a 15 as an average is a tad high, but not unreasonable. In other words, my thinking exactly seems to mirror yours.

I think the secret is to decide in advance how often you want the characters to succeed.

I think that 10=average as a default makes more sense, but I tend to run lower powered games. I suspect that the DC's given in the rulebook are set to assume that the characters are a decent level (higher than in my games) and a lot of times players can finagle Advantage on rolls and somehow they always seem to end up with higher numbers than I expect.

Long-winded way of saying you are spot on with your analysis, IMO.

[waysoftheearth]

I've just been looking at proficiency checks in Five Torches Deep, which works the same way, but it uses DC 11 for almost everything.

TBH, I don't like DC or proficiency checks much. I think it's too easy for them to get overused. E.g.:

Step 1. Reducing (or providing a general mechanism to reduce) almost everything to a target number means...
Step 2. Almost everything can be beaten with a proficiency check, which results in...
Step 3. Proficiency checking being the default path for almost everything.

Lots of checking comes with its accompanying edge cases, including the one hamurai mentioned (everyone fails) or the other classic where the superman fails and then the weakling succeeds.

But for me it's more about the frequency of checks.

To start, I'm not sure what the point of easy and very easy checks is. These could be replaced with the simple 0e logic that: Normals automatically do normal stuff, heroes can automatically do heroic stuff, and superheroes automatically do superheroic stuff. The only time a roll is required is when a player attempts something "over or above" their standard/expected capabilities. In 0e, a throw of 5 or 6 on a d6 would just about do it most of the time (so, 33% likely). In 5e terms, that would probably be nearest to the "hard" DC (20+ on a d20, so notionally 5% chance of success, but realistically more like 35% by the time you put in the typical +2/+3 proficiency and +2/+3/+4 ability adjustment). One could tweak the limit of what's standard or challenging, of course.

But all odds aside, I think the key thing is where, and how frequently the DM chooses to use these checks. So, really, I guess I'm wondering aloud: what if you didn't roll at all unless something was at least challenging and consequential? Everything else could just be assumed.

Judicious use of checks can still add something fun to the game, of course, but overuse can become a bit of a grind. Or at least, that's my 2 c.p. for tonight

[hamurai]

Aug 2, 2021 3:13:53 GMT -6 Finarvyn said:

I think that 10=average as a default makes more sense, but I tend to run lower powered games.

That's what I do, too  (see below)

Aug 2, 2021 4:45:49 GMT -6 waysoftheearth said:

But all odds aside, I think the key thing is where, and how frequently the DM chooses to use these checks. So, really, I guess I'm wondering aloud: what if you didn't roll at all unless something was at least challenging and consequential? Everything else could just be assumed.

Yeah, tell that to "those" DMs A lot of them just use a roll to decide anything.


The d20 is far too fickle to use 15 as the average challenge, in my opinion. I tend to see the average character as the "norm", whereas the PCs are usually above the average. At least, none of them are below or far below if you use the standard way of creating characters in 5E.
The average person should succeed an average challenge half the time, that's my reasoning. So 10 is my default "average" challenge. A character of, say, 5th level with a proficiency bonus of +3 and a "maxed" ability bonus (for that level, using the standard creation of 5E) of +4 would have to roll a 3 or better to get that average challenge, so they don't roll in my game. It's because they're specialists in that area if they have a +7 bonus, why roll unless the consequences of a failure are dramatic?
Even a hard challenge would only be a roll of 8+ for such a character, and while vanilla 5E may argue that's too easy, I say "They're very very good at that!"

Of course, when you're looking at it from the endgame perspective, that specialist will end up with a +11 bonus (+5 from ability bonus, +6 proficiency bonus), but that's probably what the PHB does, because that character would need a 19 or 20 to succeed at a "Nearly Impossible" task. 

Anyway, I'm not running a high-level game and I'm not playing in one, so when capping the game at level 10ish, the DC Table should look more like this:

Task Difficulty	DC
Very Easy	0
Easy	5
Medium	10
Hard	15
Very Hard	20
Nearly Impossible	25

Yes, that means a Very Easy task will not be rolled for, and even the Easy ones won't get too many rolls unless with disadvantage. Less rolling, more role-playing!

[waysoftheearth]

Aug 2, 2021 14:21:29 GMT -6 hamurai said:

Aug 2, 2021 3:13:53 GMT -6 Finarvyn said:

I think that 10=average as a default makes more sense, but I tend to run lower powered games.

That's what I do, too  (see below)

So... 10+ on a d20 gives you a 55% chance. Maybe you want 11+ for a 50% chance (which is exactly what Five Torches Deep does with it)?


My issue with this approach is that 50/50 odds isn't "gameable" insofar as there is no advantage or disadvantage to the player either way. Which means: success is down to dumb luck as opposed to player strategy. This's why I prefer OD&D's 4 in 6 or 2 in 6 chances. Then the player knows they are either odds on to succeed or fail, and this can be an input into player strategy. Basically, to "win" players should avoid throws they are unlikely to win, and push for throws they can win. With 50/50 odds this doesn't work

[sixdemonbag]

Counterpoint: 50/50 is where all the fun lies.

If a player is odds-on to succeed or fail, then just let them succeed or fail--no reason to roll. Let the ref make the call.

And is the assumption that the player knows their odds for every action? Why would they ever choose any action that's 2 in 6 to succeed unless forced to by the ref? Every roll would be 4 in 6 if the input is always binary. Besides, what's more exciting than willingly choosing a 50/50 action? That's where all the fun and drama lives.

Ref: The horde chases you to a dead-end overlooking a chasm 40-ft deep and 8-ft across.
Player: I want to jump the chasm.
Ref: The floor is slippery. That's 2 in 6 to succeed.
Player: Nevermind--I'll try to scale down the chasm assuming my chances of not falling improve.

Player: I want to jump the chasm.
Ref: Plenty of traction. That's 4 in 6 to succeed.
Player: I jump!

Player: I want to jump the chasm.
Ref: That's 3 in 6 to succeed.
Player: Bloody hell...

[sixdemonbag]

That said, I'm with waysoftheearth preferring OD&D's X in 6 chances. Plenty granular for most things and easier to judge on-the-fly. I just wouldn't throw out 3 in 6.

[waysoftheearth]

Good points Six

I agree the D&D player doesn't always have the level of visibility or control I implied, and of course, the stuff you described happens.

I guess I was just talking at a super-abstract level about gaming in general. I.e., if you're a player in a game, and you want to win, and you can win or lose by your choices, you'll probably prefer the better odds choices.

D&D is waaaay more complex than this, but at some super abstract level, I think it's still relevant.

[sixdemonbag]

Agreed. From a strategic standpoint, you’d certainly want to rig the odds in your favor whenever possible.

The nice thing about OD&D, to your point about abstraction, is that each ref can tailor the game to suit the level of strategy vs. drama and fiddle with the levers of abstraction vs. realism as desired. Not to mention, how much of the game that is to be hidden from players is easier to manage too. OD&D makes all of that very easy to do which is one of its biggest strengths.

Also, and this is building off your original point, I’ve long dreamed of running a D&D game—pretty much any edition—where the players only roll damage while every hit, save, and action is decided by a simple hidden DM d6 roll. They might think I’m rolling a d20 with all sorts of mods and table lookups, but little do they know that they (the Player Characters) will succeed 4 in 6 for everything. Monsters succeed 2 in 6 for everything.

I’d be genuinely curious if the players would ever notice this. Rolling 5 in 6 seems pointless. Just let it succeed. Likewise rolling 1 in 6 feels unnecessary. Just say it fails (or will fail if attempted.) NPCs will succeed 3 in 6 just for fun. They are unpredictable let’s say.

In a theoretical campaign, leveling gives you more HD and an extra attack or damage die and that’s it. For a fighter, that’s every three levels, etc.

[hamurai]

Players in 5E will also avoid rolling with bad odds - our DM, for example, also tells them "Jumping the chasm will be a medium challenge, TN 15". So we know what the odds are for each of our characters. Those with a +0 bonus will likely look for another way, or see if they can get some advantage due to equipment. Also, other characters will probably help the weaker ones.

Oh, and true, 10+ means 55% of success, I just use it because it seems easier to remember as 5/10/15/20...

[jeffb]

Aug 2, 2021 4:45:49 GMT -6 waysoftheearth said:

But all odds aside, I think the key thing is where, and how frequently the DM chooses to use these checks. So, really, I guess I'm wondering aloud: what if you didn't roll at all unless something was at least challenging and consequential? Everything else could just be assumed.

Which is the mentality of 13th Age, Dungeon World and Castles & Crusades*. Don't roll unless it is something meaningful.

In DW for example, if someone is unaware of you and you stab them in the back of the neck with a short sword- no roll needed. Done, dead.



One of the hardest things I think to learn as a DM, at least for me, was finding that fine line between

"I have to be an impartial Judge/Referee and besides, random is the way"

and

"This is tedious BS that is not contributing to the fun of the game for anyone"


Some game systems end up encouraging the latter, whether intentionally or not. I think that DMs are just not reminded of this enough in the books and instead newer editions devote way too much time advising DMs about things like "getting encounter balance just right" and "weaving PC backgrounds into your Adventures".





*None of which have a codified skill system- only a loose ability check mechanic.

[Deleted]

The last handful of times I ran 5e, I used the alternate DC spread from the DMG. I don't remember the page or even the section but I believe it was the part that suggested alternatives to core game mechanics, like "proficiency dice" instead of a set proficiency rating. Anyway, that one was slightly more forgiving, and 20 was considered a very difficult one, if I recall. I also seem to recall there was a significantly wider margin of success for a "moderate" roll. I could have this all very wrong, though. It's been at least a couple of years.

[jeffb]

Aug 3, 2021 15:44:01 GMT -6 @ampleframework said:

The last handful of times I ran 5e, I used the alternate DC spread from the DMG. I don't remember the page or even the section but I believe it was the part that suggested alternatives to core game mechanics, like "proficiency dice" instead of a set proficiency rating. Anyway, that one was slightly more forgiving, and 20 was considered a very difficult one, if I recall. I also seem to recall there was a significantly wider margin of success for a "moderate" roll. I could have this all very wrong, though. It's been at least a couple of years.

I'm not seeing such a table in that section or the section in running the game that talks about assigning DCs (there is a table there, but it's the standard DC table). There is the option to have auto successes on ability checks (ONLY) if your ability score is 5 over the DC of the check (And the example of a strong fighter who rolls poorly to bash down a door vs. a wimpy MU who makes the roll). This is actually how the original NEXT playtests started out- If the Door had a DC12 and you have a 12 Strength, you automatically can force it open- I wish they had kept that model and adjusted math to suit.

Maybe you were thinking of a different book? Xanathars or something? No big deal, I'm just curious.

[Deleted]

Aug 3, 2021 17:36:12 GMT -6 jeffb said:

Aug 3, 2021 15:44:01 GMT -6 @ampleframework said:

The last handful of times I ran 5e, I used the alternate DC spread from the DMG. I don't remember the page or even the section but I believe it was the part that suggested alternatives to core game mechanics, like "proficiency dice" instead of a set proficiency rating. Anyway, that one was slightly more forgiving, and 20 was considered a very difficult one, if I recall. I also seem to recall there was a significantly wider margin of success for a "moderate" roll. I could have this all very wrong, though. It's been at least a couple of years.

I'm not seeing such a table in that section or the section in running the game that talks about assigning DCs (there is a table there, but it's the standard DC table). There is the option to have auto successes on ability checks (ONLY) if your ability score is 5 over the DC of the check (And the example of a strong fighter who rolls poorly to bash down a door vs. a wimpy MU who makes the roll). This is actually how the original NEXT playtests started out- If the Door had a DC12 and you have a 12 Strength, you automatically can force it open- I wish they had kept that model and adjusted math to suit.

Maybe you were thinking of a different book? Xanathars or something? No big deal, I'm just curious.

It could have been from one of the two DM screens. Doesn't really matter now, anyway. I decided about that time that I don't much like 5e beyond level 8 or so. 

[waysoftheearth]

Ah right. Not *exactly* difficulty per se, but somewhat related... another background annoyance about the "core mechanic" (d20 + proficiency adj. + ability adj. for virtually everything) thing for me is... why make me add three numbers together virtually every time I roll? Sure, the math is trivial, but it's virtually *every time* I roll. That tiny little toll adds up over time. Plus there's the decision time to figure out whether or not I'm proficient, and which ability adj. to use. It takes more than zero time.

Suppose the ratio of with:without proficiency on checks is something like 90:10 across all 5e play (just guessing). Surely then a bunch of time/effort could be saved by applying a penalty to 10% of checks, rather than a bonus to 90% of checks? (Yes, I know our brains like addition better, so maybe it could be done by adding to the difficulty when you're non-proficient?).

Ability score adjustments are a whole other slippery slope. Personally, don't see something that applies almost all the time (e.g., dex adj. to AC, or str adj. to hit) as an "adjustment". It's *standard*. And anything that's standard should ideally be baked in without requiring any further additions. E.g., (just making this up on the spot now) Given we have more funny dice these days it's not impossible that a stronger PC could throw a d22 or d24 "to hit" while weaklings throw d16 or d18 "to hit", with no adds required. I haven't thought out what that would specifically mean to the probabilities, and I'm sure there are other ways to do it. My point is that simplicity/practicality at the table wins out for me over more elaborate systems.

Apologies. Rant over.

[hamurai]

Aug 3, 2021 17:36:12 GMT -6 jeffb said:

This is actually how the original NEXT playtests started out- If the Door had a DC12 and you have a 12 Strength, you automatically can force it open- I wish they had kept that model and adjusted math to suit.

That's an interesting bit I didn't know. I had never really bothered with NEXT at that time.

Maybe some sort of "Take 10" would make sense for 5E. I believe 3.XE had it? Unless you were in a stressful situation, you could always use 10 + Skill mod to calculate your result without rolling. In 5E it would probably just be a Passive [Skill Name], as in Passive Perception. With an Acrobatics skill of +5, your Passive Acrobatics is 15. No roll required unless the DC is higher or unless you're under pressure. Especially good characters could "borrow" unneeded points from that passive score to a single weaker characters, meaning they help them out. Passive Acrobatics 15 for a DC 10 roll means you have 5 points left to give to that clumsy wizard with a -1 Acrobatics score, who would be at passive Acrobatics 14 then.

I'm probably re-inventing the wheel here, I'm sure there's an optional rule like this in the DMG or some other book already. If not, feel free to quote me

Edit: Before using DCC dice in D&D, I'd rather play DCC in the first place It is a good idea to scale the dice, though. That's just one thing I love about DCC.

[jeffb]

Aug 4, 2021 23:33:55 GMT -6 hamurai said:

Aug 3, 2021 17:36:12 GMT -6 jeffb said:

This is actually how the original NEXT playtests started out- If the Door had a DC12 and you have a 12 Strength, you automatically can force it open- I wish they had kept that model and adjusted math to suit.

That's an interesting bit I didn't know. I had never really bothered with NEXT at that time.

Maybe some sort of "Take 10" would make sense for 5E. I believe 3.XE had it? Unless you were in a stressful situation, you could always use 10 + Skill mod to calculate your result without rolling. In 5E it would probably just be a Passive [Skill Name], as in Passive Perception. With an Acrobatics skill of +5, your Passive Acrobatics is 15. No roll required unless the DC is higher or unless you're under pressure. Especially good characters could "borrow" unneeded points from that passive score to a single weaker characters, meaning they help them out. Passive Acrobatics 15 for a DC 10 roll means you have 5 points left to give to that clumsy wizard with a -1 Acrobatics score, who would be at passive Acrobatics 14 then.

I'm probably re-inventing the wheel here, I'm sure there's an optional rule like this in the DMG or some other book already. If not, feel free to quote me

There were so many revisions of NEXT- I actually liked many of the earliest versions of the playtest rules MUCH better than what we got in 5E. I got fed up about 2/3 the way through with the way things were going and stopped playtesting. I was running it with my adult group (I use the term "adult" lightly), and did some testing with my Kids group as well. I ended up just running C&C with the adults, yoinking Advantage from the NEXT playtest and replacing the PRIME12/18, and don't think I ever bothered playtesting again until the game was released.

As I mentioned above there is the optional rule in the DMG so that if your Ability score is 5 above the DC, no roll needed. Which kind of functions like a passive check. This way most PCS will be able to beat easy tasks (DC10) that are related to their "prime attribute". e.g. Most Fighters will probably have a 15 STR or better and so can handle many doors in a test of strength without needing a roll.

[Finarvyn]

Aug 3, 2021 17:58:10 GMT -6 @ampleframework said:

I don't much like 5e beyond level 8 or so. 

This is somewhat tangential to the thread at large, but I agree 100% with your statement and wish WotC would publish a "Basic Player's Handbook" with all of the material for levels 10+ scrubbed out. Only put in the powers and options and spells that could be used in the first ten levels of the game. That would become my default 5E book.

[jeffb]

Aug 5, 2021 7:51:40 GMT -6 Finarvyn said:

Aug 3, 2021 17:58:10 GMT -6 @ampleframework said:

I don't much like 5e beyond level 8 or so. 

This is somewhat tangential to the thread at large, but I agree 100% with your statement and wish WotC would publish a "Basic Player's Handbook" with all of the material for levels 10+ scrubbed out. Only put in the powers and options and spells that could be used in the first ten levels of the game. That would become my default 5E book.

Have you ever heard of S&W WHitebox? It sounds perfect for you!