Evaluate if the dice rollers are "fair" about tts HOT 3 CLOSED

I ran a couple simulations of the dice (just a script that spawns and rolls the dice a bunch, then outputs the data). Results are attached in the images. One of them is with bounciness=0, which is how the mod currently works. The other is changed to bounciness=1.

I had some help from Screwtape to do the actual analysis. The chi-sq test on the B=0 is in the realm of probably ok, but worthy of attention. B=1 is significantly better, as I think the extra bouncing and spinning introduces more randomness and prevents those odd situations where the dice just spin in the air parallel to the table. Not to mention it's more fun because they better reflect physical dice :)

I think it's worth considering making this the default setting on the dice. I'd also like to run simulations with a greater number of trials.

from tts.

Super cool @nashjaee thanks! I agree B=1 seems more fun.

There has been concerns that the blue dice roller is better than the red. Did you have any simulations on that?

from tts.

These numbers were run using the blue roller. I can try the red one as well to compare.

Also note: this was run using a red attack die. Not that it matters I think as all 8 die faces are expected to be equal. I should also do the 6-sided dice for completion's sake.

from tts.