D&D 5E Point Buy vs Rolling for Stats

[ccs]

And it crumbled when STR came into it, being unable to handle % scores well, not to mention giant strength.

It worked just fine concerning % strength. Don't roll a 19 or 20.
You rolled an 18. Is that </= to your 18%? Yes.... congrats, you oassed.
Oh, you rolled a 19 or 20. Are those #s higher than your 18 whatever? Yes. Congrats, you failed.

I honestly don't recall what we did way back when for stat checks with stats 20+. Probably broke out the d30 - wich is what we do here in the 21st century.
Because we're not concerned with it being "realistic" that somehow a 23 stat has a greater % chance of failure than a 19 stat. We just care for a simple, quick, resolution (assuming it's something we feel needs a dice roll for in the 1st place).

 

[Lanefan]

This was a big one for me. A lot of stuff scales with level. Your hps, attack rolls, and saves, most notably. But also some of 1e's other quixotic sub-systems, like detecting an invisible creature or noticing you were just pick-pocketed.

An argument could be made that both then and now too much stuff scales with level, some of which probably shouldn't...or not for all classes.

It just never made sense for a seasoned adventurer to get tougher and tougher, better and better at fighting (even if he did nothing his whole career but stand in the back and cast) but stay clutzy or naive or oblivious.

Perhaps...but where it can make any sense, let it do so.

That was another bad thing about the roll-under system, it often gave you a much easier roll than some corresponding 'special' ability. Bend bars/lift games is an example. Sneaking past something with a roll under DEX for a 17 DEX low-level thief, for another.

Not the best example, as again this is something that has its own roll - in this case against the Thief's move silently (and-or hide in shadows, as appropriate) skill.

And again, the roll-under could easily be modified by the DM. I can't count the number of times my DM has said something like "That's hard to do. Roll under half your {stat}." Occasionally he'll also disallow those with a low stat from even getting a roll in situations like this.

And, of course, it just added to the perennial problem of TSR era D&D, unnecessarily baroque resolution systems.

d20's standardization on d20 + bonuses vs DC was a huge improvement.

I don't mind different systems for different things provided there's some rationale for it and provided the system in use does the job. Using d% instead of d20 can often (but not always) be rationalized on the grounds of better granularity, for example. Roll-under can be rationalized as making stats relevant (and yes, reducing the relevance of level sometimes) no matter what they are.

The d20 standardization model suffers from the designers coming up with what is a good idea for some situations and then trying to shoehorn far too many other things into it, things that would be better served having or retaining their own unique resolution system. For another example of this, see advantage-disadvantage in 5e. A third example is trying to use the hit point mechanic to cover everything that can ever happen to a character where some effects should bypass hit points entirely.

Back to roll-under for a minute: one place we've found it works really well is consciousness checks. If you're at or below 0 h.p. (in our game death is at -10) you need to roll your Constitution score, modified by your current h.p. total, or less on a d20 in order to remain conscious. So, if you're Con 12 and you've been clobbered to -5 you need to roll (12 - 5 = ) 7 or less in order to stay conscious. 20 always fails.

Lanefan

 

[Tony Vargas]

An argument could be made that both then and now too much stuff scales with level, some of which probably shouldn't...or not for all classes.

I'd only entertain that kind of argument if hps scaling with level were on the table. You think the fighter & cleric shouldn't get stealthier with level or have a better chance of solving a puzzle or having picked up a bit of lore? Fine, the magic-user and thief can make do with 1 HD their whole career.

Ridiculous? Nope, skill-based systems went there. RuneQuest, for instance, you advanced in each skill individually, based on how much you used it, and advancement got harder the better you got - a suggestion, at least, of a 'realistic' learning curve. Hit points? Not s'much.

Perhaps...but where it can make any sense, let it do so.

It makes sense that an experienced adventurer gets better at adventuring, so if a check comes up during an adventure, his chance to make it should scale with level. If a check comes up when he's managing his household between adventures, by all means, screw with him.

Not the best example, as again this is something that has its own roll - in this case against the Thief's move silently (and-or hide in shadows, as appropriate) skill.

Actually, that's why it's an excellent example...

And again, the roll-under could easily be modified by the DM. I can't count the number of times my DM has said something like "That's hard to do. Roll under half your {stat}."

It'd be pretty hard to sneak past that sentry, he's fairly alert. Roll under half your DEX.

DEX 12 whatever (magic-user, hobo, you name it) needs a 6, 30% chance of success.

DEX 18 1st level thief needs to roll his move silently and hide in shodows separately and mak both of them. I'm not going to dig up my PH, but I'm guessing something around 9-20% chance of success, in aggregate. Or, the DM could take pitty and and let him roll under half DEX, like everyone else, a 45% chance.

(And, it's a fool's errand, anyway, since someone in the party is going to fail that roll and blow the whole thing.)

I don't mind different systems for different things provided there's some rationale for it and provided the system in use does the job.

Any d20 check boils down to a flat % chance of success, just one divisible by 5. d%, likewise, just you can roll a 23% instead of choosing between 20 or 25. (yippee)

Roll-under can be rationalized as making stats relevant (and yes, reducing the relevance of level sometimes) no matter what they are.

By doubling their impact relative to d20 bonuses, and making them all-important in the 1e version.

The d20 standardization model suffers from the designers coming up with what is a good idea for some situations and then trying to shoehorn far too many other things into it, things that would be better served having or retaining their own unique resolution system.

Any resolution system that gives a random result with a linear distribution can probably be done fine with the d20 core mechanic. That includes roll-under and roll-high-without-going-over and d% resolutions.

It does not include bell-curve distributions, like 3d6, let alone more sophisticated mechanics like dice pools, traits, complications, etc, etc...

For another example of this, see advantage-disadvantage in 5e. A third example is trying to use the hit point mechanic to cover everything that can ever happen to a character where some effects should bypass hit points entirely.

Those are all examples of better game design than we had back in the day, yes. ;P

Back to roll-under for a minute: one place we've found it works really well is consciousness checks. If you're at or below 0 h.p. (in our game death is at -10) you need to roll your Constitution score, modified by your current h.p. total, or less on a d20 in order to remain conscious. So, if you're Con 12 and you've been clobbered to -5 you need to roll (12 - 5 = ) 7 or less in order to stay conscious. 20 always fails.

How 'bout death at negative CON, roll over your current negative hps without rolling over your CON to stay conscious?

 

[Hussar]

/snip

As for undervaluing level: in many cases this is just fine as what's being rolled for should in theory have no link to level at all. A 1st-level PC and a 15th-level PC with the same stats should realistically face exactly the same difficulty balancing on a log to cross a raging river or trying to remember a piece of local historic minutae, for example.
/snip

See, no, I don't think I agree with this. Particularly since skills are no longer class exclusive. You're basically saying that my 1st level Rogue climbs walls just as well as my 15th level Rogue. Or my 15th level wizard, who's been studying grimoires and talking to sages and generally learning all he can knows no more history than my 1st level wizard.

Plus, it winds up that 1st level characters are typically experts in their chosen skills. You quite often will choose skills where you have decent stats - so, essentially 16-18 (ish) out of 20 for that skill.

And, if you actually want to use stats for commoners, it gets REALLY wonky. Why is EVERY smith massively hugely muscled. Sure, smiths usually are big dudes, but, not necessarily the biggest dude in town every time.

 

[Hussar]

Didn't the 2e Psionics rules use the Roll Under But Close and you can Get method? As in a 1 always succeeded, but, very weakly, while rolling your stat gave you a boosted effect. I seem to vaguely remember it working like that.

But, yeah, any time you are breaking out fractions for calculations? No thanks. DM's have a hard enough time calculating odds as it stands. Adding in fractions? Yeah, not my cup of tea.

 

[Yardiff]

Try this rolling method..
Roll 24d6 remove the 6 lowest rolls, then group the remaining 18 die in groups of 3 die.

 

[Maxperson]

Don't remember where it came from, it certainly wasn't anything I saw used at a D&D table, but then these things varied wildly, and I didn't /play/ much in the 90s, just DM'd, so I was wrapped in my own variants.

I don't remember if 2e skills were like this, but 2e psionics was like that. If you had to roll a 17 or under and you hit that 17, it was a critical success. I can't remember if a 1 was a partial success or not, and a 20 might have been a fumble. It's been a long time since I looked at it.

 

[Maxperson]

It worked just fine concerning % strength. Don't roll a 19 or 20.
You rolled an 18. Is that </= to your 18%? Yes.... congrats, you oassed.
Oh, you rolled a 19 or 20. Are those #s higher than your 18 whatever? Yes. Congrats, you failed.

What we did was an 18 was a success, but if you rolled a 19, you got to roll percentile and if you got under your 18/%, you succeeded anyway.

 

[Lanefan]

Try this rolling method..
Roll 24d6 remove the 6 lowest rolls, then group the remaining 18 die in groups of 3 die.

A perfectly average roll on this should in theory generate 4 of each number on the dice, as in 4 1s, 4 2s, etc.

So a dead-on average roll will give for stats:

6-6-6 = 18
6-5-5 = 16
5-5-4 = 14
4-4-4 = 12
3-3-3 = 9
3-2-2 = 7
[2-2-1-1-1-1 - discarded]

Yeah, that's a character I would play every single time - usually as a wizard: Str 9 Int 18 Wis 7 Dex 12 Con 16 Cha 14 - gimme gimme gimme!

Lan-"Int 18 Wis 7 on a wizard-type is probably my favourite combination of stats this game has to offer"-efan

 

[Barolo]

Didn't the 2e Psionics rules use the Roll Under But Close and you can Get method? As in a 1 always succeeded, but, very weakly, while rolling your stat gave you a boosted effect. I seem to vaguely remember it working like that.

Actually, for most powers a 1 meant nothing, but for some, an ill effect would backlash the user, or the power would not work quite as planned. A 20 was not only a failure, it could be pretty dangerous. For instance, if you were trying to activate disintegrate and rolled a 20, the power would instead target yourself and you had to succeed at a save to survive (albeit at a generous +5 to the roll). But if you rolled exactly your "power score", oh, that was awesome!