Do you think a rule mechanic of 2d12 + is too swingy with lower skill ratings when used in a "versus" (i.e. a contested) roll?
For example, with a 2d12 +1 roll against a 2d12 +2 roll, there seems to be too much weight given to the 2d12 variables. In other words, that +1, +2, +3 modifier doesn't seem to make much of a difference.
As the characters improve, they'll have skill ratings up toward +16, but at the lower end (specifically from around the +1 to +4 range), it feels too swingy.
It could be argued that that's how it should be at lower skill ratings; that you are chancing more or less to luck to get you through the day. I'm not convinced though.
Thoughts?
You may need to run some numbers on this.
2d12 vs. 2d12 has I think the same distribution as 4d12. An even contest would be around TN (6.5*4 = 26) on 4d12. (Not exactly but close; an even contest should be 50/50 whereas that TN seems to give about 53%).
Fiddling around with anydice (http://anydice.com/ ) it looks like you need only about a +5 differential to change success chance from about a 50% win to a 75% win (equivalent to a TN of 21 on 4d12). A +10 difference would give about a 90% chance.
Quote from: Scutter;833993Do you think a rule mechanic of 2d12 + is too swingy with lower skill ratings when used in a "versus" (i.e. a contested) roll?
For example, with a 2d12 +1 roll against a 2d12 +2 roll, there seems to be too much weight given to the 2d12 variables. In other words, that +1, +2, +3 modifier doesn't seem to make much of a difference.
As the characters improve, they'll have skill ratings up toward +16, but at the lower end (specifically from around the +1 to +4 range), it feels too swingy.
It could be argued that that's how it should be at lower skill ratings; that you are chancing more or less to luck to get you through the day. I'm not convinced though.
Thoughts?
The bigger the dice range, the less a +1 is worth. That shouldn't be news for you, and likely isn't, but I felt like mentioning it:).
Since it's an opposed roll of essentially (2d12-2d12)+1 for the higher skill, you get a possible variation of up to +/- 22 points. The odds of rolling a result where the difference of +1 from higher skill actually matters, is slim. As in, you can go all session without such a difference making any difference;).
The difference will be felt much more strongly if both characters had to clear a difficult obstacle, say, rated at a TN of 18 or 20 (admittedly, I'm eyeballing it here). Then the difference would likely be noticeable on several consecutive attempts.
Still, it's a swingy dice mechanic. If you're making your own game, my recommendation would be to make skills giving you a +3 bonus per level. In that case one level of difference would be pretty noticeable no matter where you stand in the skill range, IMO. And yes, 3 or more levels of difference would make it almost a guaranteed win, as it should be.
Quote from: AsenRG;834010The bigger the dice range, the less a +1 is worth. That shouldn't be news for you, and likely isn't, but I felt like mentioning it:).
Since it's an opposed roll of essentially (2d12-2d12)+1 for the higher skill, you get a possible variation of up to +/- 22 points. The odds of rolling a result where the difference of +1 from higher skill actually matters, is slim. As in, you can go all session without such a difference making any difference;).
The difference will be felt much more strongly if both characters had to clear a difficult obstacle, say, rated at a TN of 18 or 20 (admittedly, I'm eyeballing it here). Then the difference would likely be noticeable on several consecutive attempts.
Still, it's a swingy dice mechanic. If you're making your own game, my recommendation would be to make skills giving you a +3 bonus per level. In that case one level of difference would be pretty noticeable no matter where you stand in the skill range, IMO. And yes, 3 or more levels of difference would make it almost a guaranteed win, as it should be.
This is exactly what I was looking for
Thanks all
Quote from: Scutter;833993Do you think a rule mechanic of 2d12 + is too swingy with lower skill ratings when used in a "versus" (i.e. a contested) roll?
Not really - a +2 means that the underdog will only win a third of the time. That feels like a pretty meaningful shift in the odds.
Quote from: AsenRG;834010The bigger the dice range, the less a +1 is worth. That shouldn't be news for you, and likely isn't, but I felt like mentioning it:).
Its not just the
size of the range, its also the
shape of the range.
More dice give a v-curve or bell-curve distribution where an initial +1 gives a relatively larger benefit and +1s later give less.
I figured out where the discrepancy in my numbers is. An even contest isn't exactly 50/50 since there's also a (small) chance of a tie, so the roll has to be over. Assuming that defender wins ties, the probability for the attacker winning would be equivalent to a TN 27 on 4d12 (47.21%) not 50%.
Even only a +2 to the defender will drop this to 36.34%, just over a third success rate.
Quote from: Bloody Stupid Johnson;834106Its not just the size of the range, its also the shape of the range.
More dice give a v-curve or bell-curve distribution where an initial +1 gives a relatively larger benefit and +1s later give less.
I figured out where the discrepancy in my numbers is. An even contest isn't exactly 50/50 since there's also a (small) chance of a tie, so the roll has to be over. Assuming that defender wins ties, the probability for the attacker winning would be equivalent to a TN 27 on 4d12 (47.21%) not 50%.
Even only a +2 to the defender will drop this to 36.34%, just over a third success rate.
Well, 2d12-2d12, as in an opposed roll, gives a nice curvy distribution:D!
So, let's say defender wins ties, but attacker has 2 points higher skill, as in your example. In that case, anything that's over 0, which means equal results, is the more skilled winning.
In this case, you get a 58,30%, slightly less than a 6 out of 10 rolls. A +3 makes it a 63,66%, which is your example. (BTW, that's still not odds that make you feel calm about the outcome, just respectable ones).
OTOH, if you have (according to my suggestion) 2 levels of skill over the opposition for a +6, it's a small increase in odds. You win in just 77,99% of the rolls.
Still not guaranteed, though nearly so.
And finally, if you get a +9, you have 88,62 odds of winning. Now, that's what I called "virtually guaranteed".
So, we mostly agree on the odds, but I wasn't talking about odds in my previous post.
I was talking about
the feeling the rolls create. And the feeling on the table is not about the odds, unless the players have Anydice plugged into their heads (which we don't expect to be the case for another couple of years:p).
Therefore, if you roll a result where you get 25 due to your skill, it feels like that skill guaranteed winning. With a +2, the odds for that are 2,08%. With +6, it's 14,58%, making it much more obvious.
Furthermore, if you roll well, say 18 or more, but only have a +2, there are 10,42% chance your enemy's roll would equal or beat it through sheer, dumb luck. That leads the players to statements like "+2 isn't worth anything in this game", IME:).
Thus, my advice to the OP was to make a game where skill boosts give you a more obvious result;).
I haven't looked specifically at 2d12, but generally with a roll-off approach we're looking at what seems to me like a pretty steep slope, even going from 2d6 to 1d20.
However, if you need a 1-level difference to go straight to 2:1 or 5:1 or whatever, and for some reason don't want to use a dice-toss scheme that does it with a 1-pip difference, there's a simple solution:
Try a table.
Use the difference in ability factors as your vector into it, instead of arbitrarily shackling yourself to using them directly as arithmetic.
Is there any specific good reason to use the 2d12 mechanic? Or are you doing it just to be different?
I recommend the 3d8 mechanic. Much better. More novel. Confuses players better as well.