I asked this question on Reddit a few minutes ago, but maybe one of the three people that reads this blog is good at math/dice stuff:

I want to use two rolls to randomize what is basically a d100 roll. First I roll a d10 to figure out what kind of dice to use for the second roll: 1-2 use a d4 for the second roll, 3-4 use a d6, 5-6 use a d8, 7-8 use a d10, 9-0 use a d12.

For the second roll, instead of just using 2 d10s to make a d100 roll, I’m going to use 2 of the dice generated by the previous roll, which will give some weird results. Like if my first roll was a [2], then I’m using 2 d4s like percentile dice, which gives a pretty limited range: 11-14, 21-24, 31-34, and 41-44. Reading a percentage is the same as if I was using d10s: first die is the tens column, second die is the ones column. So if I rolled 2 d8s and got [6] and [2] that’s a 62. d12 results are treated slightly differently: in the tens column, an [11] is 110 and a [12] is 120; in the one’s column, an [11] adds 11 (which increases the tens column by 1), and a [12] adds 12, so a roll of [12] and [11] would be 131.

So the actual range is 01-132 (right?), but with very weird distributions. Getting 01-09 is pretty hard, actually getting anything that ends in a ‘9’ is hard, but maybe 31 or 33 or 42 is the most common result? I’m hoping that’s where you guys come in – can someone figure out the probability for each result between 1-132? Or is that like some insanely difficult math?

This is for the OD&D supplement CARCOSA, which uses funky dice mechanics, but I’m interested in taking it a little further for some random tables that need some wildly variable results. *edit: The range doesn’t go all the way to 132… there are only a few results you can get above 100. I think…*

My end goal is to create a mutation chart with some wildly variable results, and I want the unique CARCOSA dice mechanics to be a part of it. I don’t know enough math to even know if this is a hugely difficult question.