[CORed]

Although the Martingale (this is the common name for the "strategy" that you, and several hundred thousand other mathematically challenged gamblers "thought up") is a horrible strategy for any game, blackjack is an especially bad game to use it on. The probability of winning a given hand is something like 47% IIRC, ignoring pushes. Blackjack has a small house edge because some of your wins are more than your original bet: You get 3:2 for blackjack, and you get a double win on a double down. The low win percentage makes long losing streaks even more likely than in a coin toss. The need to double down and split if you're going to play optimally means that your bet in a losing streak will escalalate even faster than in a game of straight even-money bets. If you lose a double down, or both hands on a split. You will have to triple your bet the next hand, if you are intent of making a 1 unit net win on that bet. If you start with a $10 bet. You will likely soon finding yourself risking tens of thousands of dollars make a net profit of $10.

It seems like in every discussion of the Martingale, a debate comes up on whether it would work with an infinite bankroll. I'm not enough of a mathematical theorist to have an intelligent opinion on this matter, and besides, it really doesn't matter because no one actually has an infinite bankroll.