(FORTUNE Magazine) – Dear Morningliner: On a recent visit to Las Vegas, I felt the need for some inspirational literature, like The Power of Positive Thinking , by the late Norman Vincent Peale, but the Gamblers Book Shop on 11th Street was not stocking it that day. This turned out to be my lucky break, since the volume I came away with is hard to beat for uplift. Casino Gambling for the Winner, by the also late Lyle Stuart, is already changing my life. * Stuart's book, originally published in 1978, serves up numerous nuggets of wisdom for players -- my own personal favorite is, don't waste money tipping the dealers -- but his big strategic insight is this: Quit while you're ahead. ''In almost every game you play,'' he writes, ''there will be a time when you are ahead. Put aside a part of that, the money you're ahead and your original stake, so that you MUST LEAVE THE TABLE A WINNER.'' Stuart tells how his approach to gambling made him a winner ten nights in a row, for a total profit of $166,505. Acting on his advice, I too was a winner on every remaining night of my trip, and my only problem was figuring out exactly what a fellow is supposed to do in Nevada after walking away from a blackjack table at 8 P.M. But my real question is, how have the casinos remained in business since Stuart disclosed the secret of taking away their money? -- CAN'T THINK ABOUT ANYTHING ELSE Dear Can't Think: Your letter raises a number of questions, the first of which is how a book built around any such transcendently dumb idea ever got published. The answer to this one is fairly clear, however. The book was published by Lyle Stuart Inc. How likely is it that players will get ahead at some point in the evening? In some games, very likely. Imagine a totally fair game of craps -- a game in which the rules have been altered so that the typical house edge (perhaps 1%) is eliminated and the player and the house each have an equal chance of winning $10 in a single play. There would be a 50% chance of our man being ahead after the first play. There would be a 12 1/2% chance of his being ahead for the first time after the third play, and an additional 6 1/4% chance after the fifth play. The sum of all the probabilities through the 1,000,001st play and beyond would asymptotically approach 1, meaning that infinite play is guaranteed to eventually get him ahead. Mr. Statistics recently performed a computer simulation of 1,000 casino evenings spent at this game, and discovered that, sure enough, the player always did get to be ahead at some point. Unfortunately, some of the ''evenings'' were rather prolonged. In the sixth evening, it took 20,349 plays before our hero got ahead, and on evening No. 744 it took him 2,250,885 plays to do it. So in the game we are assuming, the guy needs an iron constitution and more than $20 million to make the strategy foolproof. As numerous statistical brethren have pointed out, the player's disadvantages in the casino include not only the house's assorted edges in various games but the house's superior resources as well. The house edges do, however, present another problem for Lyle Stuart's strategy. In the real casino world, edges are inescapable, and this means you are not guaranteed to get ahead at some point, even if you have resources for a billion games. The way to view the strategy is to think of it as one of many in which the gambler gets himself a high probability of a small win while accepting a low probability of a disastrous wipeout. To see the strategy in action, the computer was also instructed to imagine a sequence of 1,000 evenings of roulette, a game in which American casinos have an edge of 5.26%. The instructions specified that the player would start each evening with $10,000 and play until he either got ahead or lost the $10,000. Each time the wheel was spun, he put $10 on 32, his lucky number. (His daughter was born on March 2.) The computer run was most depressing. The player had modest wins on 911 evenings and got clobbered for $10,000 on 89 evenings. For the whole 1,000 nights, he ended up with a loss of $729,880. And had no inclination to tip the dealer.