D&D 5E Ability checks not using modifier

[Sadrik]

I would like to see a system were it is looking at the success rate and work backwards from that. If you are trained in something you might have a 70% chance of success +/- your stat. If your stat modifier is big and swingy this really limits your ability to expect anything but random results. Random in this case = the stat of the doer.

 

[keterys]

Well, slight differences from 2e. But, yes, a modernized version thereof. DCs of 10 - 40 range should cover things well too.

 

[keterys]

Sadrik, it's a d20 system. The way you alter randomness is by changing the dice used, the bonuses to skills or ability scores doesn't change that math too much.

Otherwise, yes, d20 + (anything) will always have quite the random curve.

 

[Szatany]

1) You are splitting the mechanics of ability checks off from the mechanics of attacks and saves. To have the ability check modifiers (Score - 10) and attacks/save modifiers (Score - 10 / 2) is to create a disparity between mechanics to solve a problem that could probably be solved in another way. Right now... it seems like the reason for something like this is all to just solve the "+1 to ability score from even to odd grants no benefit" idea. Which to my mind isn't as big of a deal as it seems others feel it is. Do you "get anything" by going from 14 to 15 in a score? Yes. You are getting one step closer to your modifier going up next time.

There one another reason I forgot to mention. I'm talking about difference between characters with low and high scores. I dislike the fact that a character with str 10 has rather large % chance of defeating str 20 character in a contest of strenght. Interestingly, some people want the exact opposite I guess for game balance? But to me that sacrifices just too much realism from the game. Ability scores is the one place I wouldn't mind as much realism injected as possible, because they are the basis of every character and creature. I'm not saying such contest should be unwinable by a low score character, but I'd reduce his chances. Using full ability score instead of modifier for checks allows me to acomplish that.

 

[Szatany]

2) Using (Score - 10) is not really the best idea when you're stuck with the concept of a 3-18 span for ability scores, because you get so wide of a modifier span from ability score alone.

I played various games, WFRP for example which had even wider spans. Never ran into any issues with them. Lets not kid ourselves, 3-18 is assuming extreme circumstances. Most of time its really 8-20.

Which is basically 2e

All the better.

 

[Szatany]

I would like to see a system were it is looking at the success rate and work backwards from that. If you are trained in something you might have a 70% chance of success +/- your stat. If your stat modifier is big and swingy this really limits your ability to expect anything but random results. Random in this case = the stat of the doer.

My proposal doesn't include training or skill subsystem, but lets assume high score = trained/expert.
If you set DCs like this:

Easy - DC 15
moderate - DC 20
hard - DC 25
extreme - DC 30

then character will have the following % chance of success
average joe (ability =10) easy 80%, moderate 55%, hard 30%, extreme 5%
expert joe (ability =15) easy 100%, moderate 80%, hard 55%, extreme 30%
g.i.joe (ability =20) easy 100%, moderate 100%, hard 80%, extreme 55%

This doesn't include skills or other possible benefits.
Skills under this system would have to do something else than grant numeric bonus to checks, and that's easy to do.
Other bonuses should be kept to minimum, and be only on ocasion.

 

[Jeff Carlsen]

There one another reason I forgot to mention. I'm talking about difference between characters with low and high scores. I dislike the fact that a character with str 10 has rather large % chance of defeating str 20 character in a contest of strenght. Interestingly, some people want the exact opposite I guess for game balance? But to me that sacrifices just too much realism from the game. Ability scores is the one place I wouldn't mind as much realism injected as possible, because they are the basis of every character and creature. I'm not saying such contest should be unwinable by a low score character, but I'd reduce his chances. Using full ability score instead of modifier for checks allows me to acomplish that.

Contests of strength rarely have a random element. In these situations, the character with the higher strength simply wins every time.

 

[Falling Icicle]

I think giving a +1 bonus for each point above 10 is worth considering. So far, the game has felt very swingy to me, with the d20 roll mattering far, far more than a character's ability and skill. Even a 20 ability score, the pinnacle of mortal human(oid) ability, only grants a +5 bonus, which is worth a lot less than the d20 is. Even an average roll of 10.5 is worth twice as much as that modifier. This doesn't feel right to me at all.

With the DCs given in the playtest packet, even characters with a 20 ability score are very likely to fail at easy actions. A character with a +5 bonus has a 20% chance to fail at an Easy (DC 10) action. That may not sound that bad at first, but that's one in every five attempts, and that's the lowest DC you'll typically ever have on an action (most DM's aren't going to make you roll if the difficulty is "trivial"). The chance of failure increases to 45% for a "Moderate" difficulty task, to 70% for a "Hard" task, and to 95% for a "Very Hard" task! A "Formidable task" is literally impossible, even if you roll a 20, as is a "Nearly Impossible" task. Think about that for a minute. This is a person with a 20 ability score. For an average person (ability score of 10), add +25% to all of the failure chances I listed above. An average joe has a 45% chance to fail at an Easy action! This is totally ridiculous.

Skills help, of course, but those are going to be completely optional, and they even suggest increasing the already too high DCs if you use skills in your game. For the basic game, players are only going to have their ability score bonus and maybe a bonus from magic items or spells. Either characters need to get more bonuses, or the DCs need to be reduced across the board, and by a lot. For the basic game, a 25 should be "nearly impossible", since it is. Only a person with a 20+ ability score can even attempt such an action, and even then they only have a 5% chance of pulling it off!

Back to the idea of using +1 per point above 10, this does solve quite a few problems. First, it solves the "odd number ability scores don't do anything" problem. Completely. Second, it increases the difference between average people and those with amazing scores. Should an average person (Str 10), really have such a significant chance to beat a 20 Str character in a contest of strength? Right now, the 20 Str character only has a +25% advantage over the Str 10 guy. That's far too little, IMO. Third, it makes the system much smoother and allows you to use an ability score for a lot of things, like AC, passive checks or static saving throws, since one's ability score is effectively the same thing as taking 10.

As for "bounded accuracy", there's nothing about this idea that violates that. There's nothing that says you have to have tiny numbers for the system to be bounded. "Bounded accuracy" is about the numbers not inflating greatly with level, and that's still true here. Besides, even a +10 bonus is far less than the upper ends of the bonuses characters could add to their rolls in 3rd or 4th edition!

 

[Falling Icicle]

Assuming they keep the existing +1 per even number above 10 scheme, I think the DCs need to be adjusted by -5 across the board. Yes, that means that the only way to fail a Trivial action is if you have a negative ability score bonus, or if other penalties are in play. Even then, failure is so unlikely that the DM should only call for a roll in the most extraordinary circumstances. That is, after all, what "trivial" means. Moderate tasks would be DC 10, which means an average person succeeds about half of the time, and a legendary person succeeds about 3/4 of the time. Sounds a lot better to me. By reducing the DCs by 5, the descriptions actually match the math!

 

[DonAdam]

A nice feature of the mod = score -10 system is that your score is your "take 10" result, so it's easy to figure out which tasks you can auto-succeed at when you're not under pressure.

At also let's you bring back static defenses a la 4e: instead of a Wisdom save, you hit someone's Wisdom score with an attack roll. More powerful effects could have penalties--e.g., attempts to dominate someone could have a -5, whereas just distracting them has no penalty.

They would have to completely revamp damage dice and hit points, though, so I don't see this coming to pass. Too many sacred cows on the bar-b.