I know most of you don't need to read this, but I still can't stand this kind of talk. Francis Salmon advises playing a martingale progession with a base of $10 and a bankroll of $630. For the novice who asked for advice, here are the most simple arguments against this plan.
First, you lose you bankroll if you lose six hands in a row (10 + 20 + 40 + 80 + 160 + 320 = 630). You can lose it fewer than six hands if you double on a hand and lose (not doubling when basic strategy says you should reduces your expected winnings). While losing six hands in a row is unlikely, it happens. With a house advantage of 1%, you will lose your first six hands, ignoring pushes, about 2% of the time.
Moreover, without counting cards, your bets won't be correlated with your advantage (determined more or less by the ratio of high to low cards remaining in the shoe) so you should have a negative expectation of each hand. So while you are unlikely to lose your bankroll in the first six hands, it will tend to decrease overtime. In fact, you are more likely to lose your bankroll than to win $200.
While I won't post the calculations for this, please look through the archives of this board for refutations of similar schemes.
And if you're willing to accept Salmon's ideas on faith or hope, but are unwilling to reason through it, you probably will doubt it when its time to put down the big bet. I've never seen anyone bet $320 at a $10 table, and I've never seen anyone attempt a martingale for more than a few lost hands.
BTW, you might think having a bigger bankroll would fix things. The problem is that you will run against the table limit (usually $500 or $1000 at a $10 table) after the next bet. I suppose you could then move to the high-limit area, and bet up to $50,000, but do you really want to risk that before doing your homework?