AdultEssay
The Math Professor Who Beat Blackjack and Wall Street
How mathematician Ed Thorp beat blackjack with card counting, then beat Wall Street with a 19%-a-year hedge fund — and the lesson almost everyone takes backwards.
dailymath · · 8 min read
In 1961, a 29-year-old MIT math instructor walked into a Nevada casino wearing the world's first wearable computer in his pocket, wired to a switch in his shoe and an earpiece hidden under his hair. The machine, built with the help of the man who invented information theory, predicted where a roulette ball would land. It worked, giving him roughly a 44 percent edge over the house. Then, worried about what casinos might do to him, he walked away and published how he had beaten a different game instead.
The man was Edward O. Thorp, and his life is the single best answer to a question every young gambler eventually asks: can you actually beat the house with math? The honest answer is the most useful story in quantitative finance. Yes, Thorp did it, twice, at the blackjack table and then on Wall Street. And the lesson of how he did it is the opposite of what most people take from it.
The man was Edward O. Thorp, and his life is the single best answer to a question every young gambler eventually asks: can you actually beat the house with math? The honest answer is the most useful story in quantitative finance. Yes, Thorp did it, twice, at the blackjack table and then on Wall Street. And the lesson of how he did it is the opposite of what most people take from it.
The professor who solved blackjack
Thorp earned a PhD in mathematics from UCLA in 1958 and took a position at MIT. There he did something that had never been done: he used the university's IBM 704 mainframe, learning to program it himself, to work out mathematically whether blackjack could be beaten.
It could. Blackjack is different from roulette or a slot machine because the cards have memory. Once a card is dealt it is gone from the deck, and that changes the odds of what comes next. When the remaining deck is rich in high cards, the player has the advantage. Thorp's insight was to track that shift and bet more when the odds turned in his favor, the technique now known as card counting. He published it in 1962 in a book called Beat the Dealer, which sold more than 700,000 copies and remains the foundational text of advantage play.
Casinos reacted the way any house does when the math turns against it: they changed the rules. More decks in the shoe, more frequent shuffling, and surveillance trained to spot the tell-tale pattern of a counter raising his bet when the count goes positive. Thorp had proved a game could be beaten, and in doing so he taught the industry how to close the hole.
It could. Blackjack is different from roulette or a slot machine because the cards have memory. Once a card is dealt it is gone from the deck, and that changes the odds of what comes next. When the remaining deck is rich in high cards, the player has the advantage. Thorp's insight was to track that shift and bet more when the odds turned in his favor, the technique now known as card counting. He published it in 1962 in a book called Beat the Dealer, which sold more than 700,000 copies and remains the foundational text of advantage play.
Casinos reacted the way any house does when the math turns against it: they changed the rules. More decks in the shoe, more frequent shuffling, and surveillance trained to spot the tell-tale pattern of a counter raising his bet when the count goes positive. Thorp had proved a game could be beaten, and in doing so he taught the industry how to close the hole.
The Kelly Criterion: how much to bet
Knowing you have an edge is only half the problem. The other half, the half that ruins most people who do have an edge, is how much to bet. Bet too little and you barely profit. Bet too much and a normal losing streak wipes you out before your advantage can play out.
Thorp's answer came from a 1956 paper by John Kelly, a physicist at Bell Labs. The Kelly Criterion gives the fraction of your bankroll to wager to maximize long-run growth:
Thorp's answer came from a 1956 paper by John Kelly, a physicist at Bell Labs. The Kelly Criterion gives the fraction of your bankroll to wager to maximize long-run growth:
Here f-star is the fraction of your money to bet, p is your probability of winning, q is 1 minus p, and b is the net odds. The formula says something profound and deeply unfashionable: even when you have a real edge, the mathematically correct bet is a small fraction of your money, and going all in is almost always wrong. Kelly betting is the discipline that separates people who have an edge and keep it from people who have an edge and blow up. It is the single most important idea to take from Thorp's whole career.
From the casino to Wall Street
Thorp realized the market was just a bigger casino with worse-informed players. In 1967 he co-wrote Beat the Market, showing how to price warrants and hedge them against the underlying stock, arbitrage that locked in an edge the way card counting had. He is widely credited with deriving an options-pricing formula equivalent to Black-Scholes several years before it was published.
In 1969 he put it to work, founding the hedge fund that became Princeton/Newport Partners. Over roughly nineteen years, from 1969 to 1988, it returned about 19 percent a year after fees, versus about 10 percent for the S&P 500 over the same stretch, with no losing year. (Some accounts round it to '20 percent, never a down quarter'; the well-documented figure is the one just given.) The fund closed in 1988 in the fallout of a federal investigation into Michael Milken's junk-bond operation. Thorp and his partners were never charged and were cleared, but the legal damage forced a wind-down. He kept investing on his own, later reporting a 20 percent annualized return over nearly three decades of personal trading.
In 1969 he put it to work, founding the hedge fund that became Princeton/Newport Partners. Over roughly nineteen years, from 1969 to 1988, it returned about 19 percent a year after fees, versus about 10 percent for the S&P 500 over the same stretch, with no losing year. (Some accounts round it to '20 percent, never a down quarter'; the well-documented figure is the one just given.) The fund closed in 1988 in the fallout of a federal investigation into Michael Milken's junk-bond operation. Thorp and his partners were never charged and were cleared, but the legal damage forced a wind-down. He kept investing on his own, later reporting a 20 percent annualized return over nearly three decades of personal trading.
The one-day Madoff catch
In 1991, seventeen years before Bernie Madoff's arrest, Thorp was hired to vet Madoff's returns. In a single day he found the fraud: Madoff reported options trades that could not have happened. On one day he claimed 123 option contracts of a stock when only about 20 had traded on the entire exchange. Thorp quietly pulled his clients out. The math does not lie; you just have to actually do it.
The lesson everyone gets backwards
The wrong lesson from Thorp is 'a smart enough person can beat the casino, so maybe I can too.' You almost certainly cannot, and neither can Thorp today. Modern casinos deal from continuous shuffle machines that reset the deck every hand, deploy AI-assisted surveillance, and simply eject anyone who looks like a counter. The specific hole Thorp found has been mathematically sealed.
The right lesson is what actually made him rich, and it was not blackjack; card counting made him a modest amount and got him barred. The fortune came from applying the same three things to a market that could not ban him: a genuine mathematical edge, disciplined bet-sizing, and the patience to let the odds compound. Those three things are exactly what a quant, an actuary, or a data scientist gets paid to do, legally, at scale, and without anyone drugging your coffee, which Thorp alleged a casino once did to stop him counting. His real edge was never the trick. It was the math, the discipline, and choosing a game he was allowed to keep winning.
The right lesson is what actually made him rich, and it was not blackjack; card counting made him a modest amount and got him barred. The fortune came from applying the same three things to a market that could not ban him: a genuine mathematical edge, disciplined bet-sizing, and the patience to let the odds compound. Those three things are exactly what a quant, an actuary, or a data scientist gets paid to do, legally, at scale, and without anyone drugging your coffee, which Thorp alleged a casino once did to stop him counting. His real edge was never the trick. It was the math, the discipline, and choosing a game he was allowed to keep winning.
Thorp started with a probability calculation on a university mainframe. The skill that carried him from a blackjack table to a hedge fund is the same one the dailymath placement test measures. See where yours stands.
Try dailymath