long story but it seems to me if you chart a number of games - not hands, say 100 hands each, the number of wins (or losses) should approach a normal distribution and IF you play the same rules this should be at near the probability level of a win (or loss). Similar to if you flip a coin 10 times each for 100 replicates you should get a mean number of heads at about 5 and have less times when you get 4, or 3 or 1 "heads." So what if you get a strongly bimodal distibution - e.g. peaks of 2 or 3 "heads" AND another peak in terms of number of trials, where you get 7 or 8 "heads."
You have a better chance of having more wins or losses than expected by simple probability than you do of hitting the expected level. Sometimes they let you win - sometimes they make you lose (sometimes you're the windshield and sometimes you're the bug) and this can be easily done in programming by stacking the virtual deck with high or low cards.
observation 2: losing streaks of 8-12 or more losing hands that far exceeds probability expectations.
IDK - just wondering if anyone else thought this made sense?