Rules Junkies: 2d10 vs d20

[LightPhoenix]

Why would each of the p(x) being equal (i.e. 0.05) not allow for a 'meaningful' mean ??

It's not that it's not meaningful, it's that it's very misleading. Yes, the mean of the results of a d20 is 10.5, which most people then conclude, incorrectly, that you're more likely to roll in the middle ranges, which is simply not true.

 

[Nifft]

Hijack:

I like the idea of open-ended rolling. The "extra d20 on a 20" rule in the ELH looks like it'll come up too infrequently to really influence play.

However, I was thinking that 3d6, plus a rule granting an extra d6 on a 6, might make the whole system work better.

Thus, no "critical failures", nor the need for a critical success -- you can instead grade success or failure by how much you succeed or fail on a roll.

Why I like the idea:
- Action Points can buy you an extra d6, which counts towards critical success.
- You have a 42% chance of earning at least one re-roll, which will lift the high end of the bell curve.
- More interesting critical hit scenarios. The obvious one is three natural sixes, but that's too rare (0.5% or so). Another idea would be at least one six, and the rest fives or sixes, but that's still fewer than now (1.85% vs. 5%). Still, the granularity means that it's going to be possible to come up with a rule to allow very flexible critical hits, even if it's not as obvious as "natural 20" was. Perhaps the best thing to do would be to divorce the idea of unconditional success from critical hits, and allow that ANY roll will succeed on a "natural 18", while a critical success would require that you beat your target DC or AC by +10, +15 or even +20.

-- N

 

[fuindordm]

I agree with 'elven', it's a very useful thing to do for opposed rolls, not so much for
fixed DCs.

For combat, 2d10 actually maps pretty nicely onto the existing threat ranges:

20 (5%) becomes 18-20 (6%)
19-20 (10%) becomes 17-20 (10%)
18-20 (15%) becomes 16-20 (15%)
17-20 (20%) becomes 15-20 (21%)
15-20 (30%) becomes 14-20 (28%)

But criticals will become rarer overall because the confirming hit is often
more difficult. If I were doing this I would say that 19+ is an automatic
hit (3% of the time is still a reasonable rate), 3- is an automatic miss,
20 is an automatic crit provided you didn't need a 20 to hit, and 2 is a
fumble.

Ben

 

[Vrecknidj]

Rolling 2d10 on saves, instead of 1d20, will result in a lot more successes on middle and low save DCs, and a lot more failures on higher save DCs. This could be a real problem for balance.

A save DC of 11, with a +3 modifier, only needs an 8. Well, an 8 or better on d20 will happen 65% of the time. An 8 or better on 2d10 will happen 79% of the time.

A save DC of 14, with the same modifier, needs an 11. An 11 or better on d20 happens 50% of the time. An 11 or better on 2d10 happens 55% of the time.

A save DC of 19, with a +4 modifier, needs a 15. A 15 or better on d20 occurs with a frequency of 30%. A 15 or better on 2d10 happens only 21% of the time.

Dave

 

[Tonguez]

It's not that it's not meaningful, it's that it's very misleading. Yes, the mean of the results of a d20 is 10.5, which most people then conclude, incorrectly, that you're more likely to roll in the middle ranges, which is simply not true.

Yes what LightPheonix said - I could have explained further but 'meaningful mean' had a more pleasant poetry to it

 

[Impeesa]

Note that doing this makes anything that expands critical threat ranges a lot more valuable.

--Impeesa--

 

[Malin Genie]

... which most people then conclude, incorrectly, that you're more likely to roll in the middle ranges ...

Seriously?

Malin "this is doing nothing to reassure me" Genie

 

[LightPhoenix]

Seriously?

I'm not sure if you're being sarcastic or not... so I'll just assume not and maybe make an idiot out of myself.

Look at some of the treasure or encounter tables from the original Gygax games. All of the rare stuff is at 1 or 100. Of course, those are often weighted as well (frex, an entry might be 46-48), so as to alter the probabilities that way. In general though, most of the random charts produced through the history of D&D tend to follow this pattern, especially the charts that have an equal number of outcomes for each treasure/encounter/what have you.

 

[AeroDm]

Ignore me if I am not understanding what you are saying, but it seems that you are implying your 20 sided die has a higher chance of landing on an 11 than on a 12, a 12 than a 13, and so on. Simple rational thought shows that this cannot be. Every number _should_ come up 5% of the time.

A d100 has an equal chance of producing any number between 1 and 100. The fact that significant treasure items are placed at the ends is simply because the ends are more significant than the middle. Contrived, eh?

2d10 skews the results closer to 11 because of the total set of outcomes it can have. While each die is still equally likely to come up with any number 1-10, of the 100 possible combinations only one equals 20 (two 10's) while ten combinations equal 11 (1,10..2,9..3,8..4,7..5,6..and their opposites). Hence you are 10 times more likely to get a result of 11 than you are of 20.

 

[Bagpuss]

Weapons with threat ranges greater than 20 become stronger, as do feats and spells that increase the threat range, since it is no longer a 5% increase with each step.

A weapon with a 18-20 threat range in a d20 system is likely to threaten 3 in 20 times. Three times more often than a threat range of 20.

The same weapon in 2d10 system will 6 threaten in 100 times, compared with 1 in 100 for a threat range of 20. So it will threaten six times more often.

Improving the threat range by 1, is a regular increase in probability in a D20 system, but in 2d10 each increase gives a larger step in the chance of the event occuring.

The same also becomes true for things that alter dice rolls for example any spell to improve armour class, for each +1 to AC you can make yourself much less likely to be hit. While in D20 it just removes one chance of being hit, in 2d10 it could remove any number of possibilities of being hit, and is impossible to predict since it depends on where it is on the bell curve.