Did anyone else find this depressing? I realize the payoff is set up to make you feel cheated (see "crdrost" below), but it had a more existential impact on me until I tried using /dev/urandom parity and after 500 presses it was still within the margins described. I feel like there's a psychology experiment under the hood here... (And also wonder what kind of exciting behavior data unrelated to the "game" that the webdev captured! e.g., average play time, average win during playtime, # of times the B(2,6) showed up, # times Pi parity showed up, etc...)