https://mathjokes4mathyfolks.wordpress.com/2015/01/15/making-heads-or-tails-of-randomness/

If only dice were less random...

Cool article - I figured it was either B or C, but I look at a lot of plots & data for my job.

I would love to have some "rubberband luck" option for games.  If your dice luck got bad enough the computer would cheat on your behalf.

here's an example algorithm, that considers your last 100 rolls to adjust your current roll.

L100  = luck over last 100 turns.

if player loses battle, and L100 is negative, there is a L100 chance of a reroll of the battle.

Specifially If you had -20 luck over the last 100 rolls, and  you lost your next roll, you would have a 20% chance of getting a free (maybe hidden) reroll.

Might have to tweak those #s a bit, maybe double the % chance?  I dunno.  For smaller boards, maybe instead of 100 it should be 50 or 20.  Anyway, it would be nice to have less games decided on the basis of lucky dice.

My guess was C.  I didn't like B because it looked too clustered in some places and I didn't like A and D because they were too uniform in their distribution to be random.  C looked suitably random in that it wasn't uniform nor particularly clustered anywhere.  I guess I was wrong :-(

I got B and C right, but switched A and D.

Thanks to WG I chose B :P