So in discussion with these guys like Alan and Singer, that claim you can beat -EV short-term (which you can) and then just duplicate that over and over, the example I always use is roulette and how a player can play black and win 3 of 4 spins and quit for the day, but could not do that over and over for 100 days.

So last night I was out at a bar with friends watching the NBA playoffs, when I thought of another great example. The Dallas Mavericks started the game doubling up the Utah Jazz. Scores were 8-4, and 16-8 in the early going. You see this frequently. Occasionally it will last for a whole first quarter with one team on top say 32-16. These are all short term examples. But how often do you see a professional basketball team double up the score on another (even a really bad team) for the entire game? How often do you see a game that ended 120-60? That is a perfect example of short-term vs long-term.

I suppose it has happened, but not very often. The point being with a small sample size or small trial, these results occur. But when you get to a larger sample size, larger trial set, they do not. And that is exactly the way it is with -EV gambling. You can defy the math short-term, small sample size, but when you get to the long run, you can no longer defy the math. The math takes over. This is actually known as the "law of large numbers". Kind of a silly name....I know.

And btw, this also occurs with +EV. It is why those of us that play with an advantage have losing periods. But you get to the long term and the math takes over.