> It's not like I could play the Powerball lottery enough times to prove jackpots pay out at the advertised rates. You don't have to play the lottery yourself

> It's not like I could play the Powerball lottery enough times to prove jackpots pay out at the advertised rates. You don't have to play the lottery yourself to analyze the results. (Not that anyone participating in a state lottery is paying much attention to the odds... if they cared about that they'd take their money to an actual casino where the house takes a much smaller cut.) > they could fiddle the draw so jackpots over half a billion dollars all went to the CEO, and it'd be undetectable It might be undetectable once. Even twice might be written off as coincidence by some, though many would rightfully be convinced it was rigged and stop playing. Sure it's not impossible for that to happen by chance, but the probability of the CEO winning multiple large jackpots is remote. Anyway, most lotteries and the like bar their own employees and close associates from participating; their business rests entirely on their reputation for fair play and they can't afford that kind of conflict of interest. There are algorithms which can be used to generate provably random results. For example: Pick a random 256-bit seed and announce the hash of that seed to the public. Wait for six blocks to be mined on the Bitcoin blockchain and then announce the seed. The random result is determined apply a predetermined formula to the hash of the seed value plus the blocks mined during that interval. The lottery operator has to choose the seed before they know which blocks will be mined, and the miners don't know what the seed value is until after the fact, so no one has the ability to influence the result.