2d10 instead of d20? (3.5)

[Seharvepernfan]

If I were to switch all d20 rolls to 2d10, what would the effects be in the game? I mean, the obvious ones being that most rolls are average, and high/low rolls are rare (making crits rarer), but what might I not be seeing?

One of the things that bugs me about skills in particular, is the fact that I can jump 20ft. one time, and only 1ft. the next...which is kinda ridiculous. I know that you can rationalize it by saying that "the random nature of the d20 roll accounts for circumstances outside the character's control", but it still seems like too wide a range.

I never considered rolling two dice for most rolls instead of one, and I think I like the implications, but I want to make sure I'm seeing all of them and understanding them, before I make any changes to my house-rules.

[Thurbane]

Well either effects that on a natural one never occur, or only occur 1 in a 100, depending how you rule.

Effects that trigger on a natural 20 only happen 1 time in 100.

[SiuiS]

By the rules, the only jumping one foot should never occur – that character tripped. They must have, because the default is a ten foot jump.

Shifting to a bell curve makes bonuses weird. A +1 bonus doesn't really do jack all if the DC is like, 13. It is a godsend if the DC is in the 17 range though, because that 5% increase is greater than the difference between pass (17) and fail (16). I don't do it justice, but GURPS makes a solid point for it.

I'm considering toying with 4d6 myself, and just skewing DCs (average is 14 not 10) for just this reason; it makes bonuses from strength or such more meaningful as far as epic deeds while not really differentiating the low level tasks much. That's the theory anyway.

[Seharvepernfan]

Well either effects that on a natural one never occur, or only occur 1 in a 100, depending how you rule.

Effects that trigger on a natural 20 only happen 1 time in 100.

Ah, of course. That's probably going to be the deal-breaker right there.

EDIT: Except on skills. I think I might keep it for skills.


By the rules, the only jumping one foot should never occur – that character tripped. They must have, because the default is a ten foot jump.

I think you might be thinking 3.0.

[GoatBoy]

2d10 isn't actually a bell curve, more like a pyramid.

Any rolling convention that reduces randomness means that encounters with multiple weaker creatures are even easier, since said creatures rely on the very real chance of rolling high in order to provide a threat at all.

Vice-versa for encounters above party level - the threat is much larger because lucky rolls are far rarer.

If randomness bothers you, perhaps expand the circumstances under which characters can take 10 on skill rolls.

[prufock]

One of the things that bugs me about skills in particular, is the fact that I can jump 20ft. one time, and only 1ft. the next

You take 10 and jump 10 feet instead. The thing is, the more you adjust the dice rolls toward the central tendency, the more you're basically just saying "I take 10" all the time.

[JusticeZero]

On any sort of curving scale, the effects of optimizing become more pronounced. As you start pushing the envelope with a flat random, you get to be, say, "frustratingly hard to hit". If there is a moderate dice curve, that guy who was "very tanky" before turns into "You go ahead and rest and regain spells. I'll tank them at the door until you wake up." The ends of the curve start zooming off into the far reaches of absurdly improbable. Mr. Tank Ridiculous has an AC in the high 50's and Fast Healing 1. In d20, he gets hit once every 20 swings, so as long as those swings average more than 20 points of damage per round in total, eventually the enemy can cut him down. 2d10 means they need 100 points of damage per round to match his regen.

[Maginomicon]

Use the official Bell Curve Rolls variant (http://www.d20srd.org/srd/variant/adventuring/bellCurveRolls.htm) instead. It's more thought-out, comprehensive, and realistic than a d20 or 2d10.

However, if you're also using the official Players Roll All The Dice variant (http://www.d20srd.org/srd/variant/adventuring/playersRollAllTheDice.htm), I would very strongly suggest one minor change to the way you present the variant to players:

Present weapons as of different threat "types", much like how the Fatal Frame franchise has different types of film:

"Type-13" for weapons with an old range of 20 (threat size 1) to a new range of 16-18 (threat size 3)
"Type-24" for weapons with an old threat range of 19-20 (threat size 2) to a new threat range of 15-18 (threat size 4)
"Type-35" for weapons with an old threat range of 18-20 (threat size 3) to a new threat range of 14-18 (threat size 5)
"Type-45" for weapons with an old threat range of 17-20 (threat size 4) to a new threat range of 14-18 (threat size 5)
"Type-66" for weapons with an old threat range of 15-20 (threat size 6) to a new threat range of 13-18 (threat size 6)


It's much easier for my players to understand the bell curve with the Players Roll All the Dice variant since the new threat size applies to the high-end PC attack rolls and low-end PC AC rolls.

-------------------

Also, regardless of the above, I recommend make the following changes to a number of effects (some from 3.0e, when threats were treated differently) that messed with threat ranges:

Disciple of Dispater prestige class:

Rewrite the text of the “Iron Power” class feature to read the following:

“When using an iron or steel weapon, a 4th-level disciple of Dispater gains a +1 insight bonus on attack and damage rolls. Furthermore, his threat range is doubled as if he were using a keen weapon. At 8th level, the insight bonus improves to +2, and you also gain a +4 bonus on rolls to confirm a critical threat. This ability does not stack with the keen weapon quality or the Improved Critical feat.”

Mythic Exemplar prestige class

The “Paragon's Gift, Least” ability at 2nd level for Sunyartra instead gives a +4 to confirm a critical hit.

Psychic Weapon Master prestige class

The “Improved Critical” ability at 7th level exception for when you already have the Improved Critical feat instead gives a +4 to confirm a critical hit.

Streetfighter barbarian ACF

The 7th-level benefit instead gives a +4 to confirm critical hits.

[AmberVael]

I've actually played using a similar dice rolling system, albeit for a different game. Most of my experiences with the probability translates over to other systems though.

My brief dabbling with it has convinced me to never use it, and that it is a pretty terrible idea. I'll admit that there might be a good way to do it, but I've not seen or thought of it, at least.*

It seems good from the outside- reducing randomness a bit more, making things not so chancy. But the problem is that reducing randomness this way plays havoc on usable probabilities. Because there is less randomness, the sweet spot of gaming probability- where something could go either way- shrinks drastically. This is even worse than it sounds, because you're having to keep multiple characters in mind... so what is a decent challenge for one character could be murderously difficult for another and laughably easy for the next just because they have a few points different.

For example.
Say the party comes across an evil wizard who wants to hit them all with a spell with a DC of 20. One character has a not so great save, say 6. Another has an average save of 9. The last has a slightly higher save of 11. Pretty reasonable for a normal party.

Under the normal system, the characters would have a 35%, 50%, and 60% chance of making that save respectively. There's enough difference in probability to be pretty notable, but it isn't out of hand.

Under the 2d10 system, the characters have a 28% chance, a 55% chance, and a 72% chance. Even the slightest difference can make a drastic change for a character's likeliness to succeed... or survive.


One point of bonus is no longer equal to another point, and because of the drastic difference a single point can make, this kind of system favors heavy specialization- you're very unlikely to succeed in something you're poor in, after all, so why bother investing into it? Meanwhile, you need every point you can put into what you want to be good at, to stay on the right end of the curve- if you want to reliably hit that boss monster, you can't afford to be even a couple points behind.


* edit: With a bit more thought, I suppose the Fate systems work with a more bell curve probability, but they have a ton less variation in bonuses AND rolls, and give easy ways for anyone to boost their bonuses as necessary or at least make rerolls...

[Piggy Knowles]

OK, this is a bit crazy, but I've thought about this in the past...

What about doing away with the d20 and replacing it with 10 +/- 1d3? (Roll a d6, on an odd roll add the result/2 rounded up, on an even subtract the result/2). That means you're basically always taking 10, with a chance to do a little better or a little worse. No big swings, just consistency across the board.

Super boring? Yeah, probably, but it eliminates wild swings. Crits would have to either be eliminated or happen by percentile.

[Maginomicon]

It seems good from the outside- reducing randomness a bit more, making things not so chancy. But the problem is that reducing randomness this way plays havoc on usable probabilities. Because there is less randomness, the sweet spot of gaming probability- where something could go either way- shrinks drastically. This is even worse than it sounds, because you're having to keep multiple characters in mind... so what is a decent challenge for one character could be murderously difficult for another and laughably easy for the next just because they have a few points different.I get around this problem in my main campaign using a modified version of the official Variable Modifiers variant (http://www.d20srd.org/srd/variant/adventuring/variableModifiers.htm) that I call "Press Your Luck". At any time you would make a roll that the variable modifiers variant would apply to, the PC can choose whether or not to use the variable modifier. Opting to use the variable modifier is called "Pressing Your Luck" because you're putting more luck into your chance of success (instead of taking a simple minimum roll, you risk a lower minimum roll for a chance at a higher maximum roll). For any situation where the GM would make a roll on the player's behalf, the GM must not Press Your Luck.

[JusticeZero]

I get around this problem in my main campaign using a {house rule}...
Translation:

I use a house rule to make random rolls less random, then I use another house rule to make them more random again to solve the problems created by the intended behavior of the first house rule...

"there was an old lady who swallowed a spider to catch the fly.."

The less layers of rules you have to deal with the better, as a general rule. I'll also note that a lot of the horror stories you see have the general pattern of "First the GM wanted to buff X. Then he decided that the buffed version was too tough, so he added nerf Y, but left buff X. Then we were having trouble because of Y, so he gave us buff Z. Then.."

[Maginomicon]

The less layers of rules you have to deal with the better, as a general rule.There's nothing wrong with fixing a fix. House rules are made all the time in response to the 3.5 update and errata updates. Making a variant behave the way you actually wanted it to behave in the first place is not problematic.