Sean Boyll
United States Kuna Idaho

Does anyone know of a dice rolling site or app (WinPC or Android) that supports opposed rolls? Preferably with custom valued dice.

Jeremy Lennert
United States California

Assuming you're using the usual concept of an "opposed roll" where you're effectively subtracting one side's roll from the other, you can do all of that with anydice.com
For example, suppose you're making an attack in Betrayal at House on the Hill, which involves an opposed roll of a variable number of special dice numbered from 0 through 2. If the attacker is using 5 dice and the defender is using 3, you can type in:
output 5d{0,1,2}  3d{0,1,2}
Then click the "Roller" button below the code entry box to put it in roller mode, and off you go.

Sean Boyll
United States Kuna Idaho

I'm looking to have the result of individual custom numbered D6es compared individually with other custom D6es so the higher number result from each die rolled wins.
To be more specific; I'm trying to get the odds of multiple types of nontransitive dice rolled in variable numbers against each other.

Jeremy Lennert
United States California

You can probably do that in anydice as well, though it may get complicated depending on exactly how your system works.
Anydice supports comparison operators like < and >, so you can test if one weird die rolls higher than another weird die with something like:
output d{2,4,9} > d{1,6,8}
Comparisons evaluate to "1" for true and "0" for false. If you just want to roll that pair of dice a bunch of times, you can turn the result of that comparison into a new die, and then roll as many copies of it as you like:
output 10d(d{2,4,9} > d{1,6,8})
If you want to combine several different headtohead rolls, you can do math on the results of the comparisons:
output (d{2,4,9} > d{1,6,8}) + (d{1,6,8} > d{3,5,7})
If you want to roll two groups of dice and pair them up based on what they roll (rather than determining the pairings before rolling)for example, to do the combat rolls in Risk where you pair the two highest dice, then the next two highestyou'll probably have to write your own subfunctions. The onsite documentation's pretty good, but you may need some basic programming experience. (Or you can try to talk someone here into writing the program for you, but you'll need to provide a complete description of how your dice system works.)

Sean Boyll
United States Kuna Idaho

Thanks Jeremy. That helped a lot. I am digging through the documentation and functions now.
The example from Risk you use is basically what I want to do so I can test different ways of assigning the dice.
As practice I have done the below based on the Efron Dice found here.
BVM: d{3} > d{6,6,1,1,1,1} output BVM named "Blue vs Magenta" RVB: d{0,0,4,4,4,4} > d{3} output RVB named "Red vs Blue"
I already know the odds for one die verses the other so I know I'm doing that part right as I get:
Blue vs Magenta 0 33.33% 1 66.67% RED vs Blue 0 33.33% 1 66.67%
I tried the head to head you suggested using output BVM + RVB
I think I have the syntax correct as I get a result of: Output 3 0 11.11% 1 44.44% 2 44.44%
But I'm not sure what this tells me exactly.
I also want to see how the odds change when there are, say, more Blue dice rolled against less Magenta dice. I'm sure I just need to find this in the documentation/Functions.

Jeremy Lennert
United States California

SPBTooL wrote: I tried the head to head you suggested using output BVM + RVB
I think I have the syntax correct as I get a result of: Output 3 0 11.11% 1 44.44% 2 44.44%
But I'm not sure what this tells me exactly. That tells you that if you roll BvM and then also make a completely separate and independent roll of RvB, there's a 4/9 chance that you'll win both rolls, a 4/9 chance that you'll win one but lose the other, and a 1/9 chance that you'll lose both.
SPBTooL wrote: I also want to see how the odds change when there are, say, more Blue dice rolled against less Magenta dice. I'm sure I just need to find this in the documentation/Functions. Well, you can't be doing a oneforone comparison if the number of dice is different. You need to start by defining precisely how you are combining all of those dice to produce the final result.
For example, you could add up all the dice from each pool and compare their sums. Or, you could take the single highest die from each pool and just compare those, discarding the other dice. Or you could use the dice to make poker hands (pairs, straights, full houses...) and compare those!
If you want a useful answer, it's critically important to ask the right question.

Sean Boyll
United States Kuna Idaho

Antistone wrote: SPBTooL wrote: I tried the head to head you suggested using output BVM + RVB
I think I have the syntax correct as I get a result of: Output 3 0 11.11% 1 44.44% 2 44.44%
But I'm not sure what this tells me exactly. That tells you that if you roll BvM and then also make a completely separate and independent roll of RvB, there's a 4/9 chance that you'll win both rolls, a 4/9 chance that you'll win one but lose the other, and a 1/9 chance that you'll lose both.
Got it. That will be helpful.


