Phigment said:No. Amazingly enough I actually penned this entire email from my own head.
Then why do you question the facts? It just seems like the standard lines we see in textbooks. If you're saying he didn't get 41 straight losing hands, just call him a liar and let that be the end of it - it has NOTHING to do with any analysis. If you want to say it wasn't blackjack and it was poker, then the same thing applies.
Phigment said:We both know "sample size" is a common statistical phrase. I certainly didn't coin it but I am comfortable using it. How many total hands he played during this BJ session and EVERY session before and after have EVERYTHING to do with this example. We're attempting to determine the long run possibilty of the event occurring, no? The short run is irrelevent. If I hit a jackpot in my first ever pull of a slot machine the casino is not going to run and adjust the payout on the machine. They know the long run is the only accurate number.
I will review the rest of your rebuttal but it's suspect if you're dismissing the need for an appropriate sample size right off the bat.
Sorry to burst your bubble, but it doesn't matter how many hands he played.
If you flip a coin 5 times and get 5 heads, the chance of that happening does NOT depend on whether the coin was flipped BEFORE or AFTER that. It is what we term an "independent event". The probability is not going to be anything different from 50% for flipping a head on single toss whether it was flipped 5 times prior or 434 times prior. The same goes for losing a BJ hand.
If I ask you the probability of rolling a 7 in craps, the probability is 1 in 6 - REGARDLESS of who touched it last, how many times it was rolled, who put a jinx on it, what the temperature and humidity is, how many times you roll it afterwards and so on. There is a 16.6% chance of rolling a 7 in craps - and that is all it is.