Fortune's Formula: The Untold Story of the Scientific Betting System That Beat the Casinos and Wall Street [Englisch] [Taschenbuch]

This is an excellent book about the discovery of the Kelly formula that is unknown outside gambling. This story has three protagonists. Two of them were scientists working at Bell Labs: Claude Shannon, a genius polymath who developed information theory; and John Kelly, a maverick genius, who is directly responsible for the development of Kelly's formula. The third one is a brilliant MIT mathematician, Ed Thorp.

Ed Thorp tested the Kelly formula in both gambling and investing. Also, he came up with an options formula before Fischer Black and Myron Scholes. His formula missed a risk-free rate component due to the structure of the market at the time. As a result, Ed Thorp remained in obscurity while Black and Scholes became famous.

Ed Thorp succeeded in deriving superior returns in both gambling and investing. But, it was not so much because of Kelly's formula. He developed other tools to achieve superior returns. In gambling, Ed Thorp succeeded at Black Jack by developing the card counting method. He just used intuitively Kelly's formula to increase his bets whenever the odds were in his favor. Later, he ran a hedge fund for 20 years until the late 80s and earned a rate of return of 14% handily beating the market's 8% during the period. Also, his hedge fund hardly lost any value on black Monday in October 1987, when the market crashed by 22%. The volatility of his returns was far lower than the market. He did this by exploiting market inefficiencies using warrants, options, and convertible bonds. The Kelly formula was for him a risk management discipline and not a direct source of excess return.

Ed Thorp's career as a hedge fund manager was temporarily cut short. This was due to his fund being involved in a tax-avoiding securities scheme with Drexel Burnham. Thorp was not guilty; but, the fund had to be liquidated. The author stated many of Milken wrongdoings. One included getting large equity positions attached to the junk bonds he issued. The companies thought they were issuing convertible bonds. However, the equity component went straight into Milken's pocket as he sold the bonds to investors as high yield debt with no equity attached.

Ed Thorp rebounded from this mishap and started a second hedge fund in 1994. Thorp continued reaping above market return. As the author states, Ed Thorp's genius consists in "...his continuous ability to discover new market inefficiencies ... as old ones played out." Ed Thorp closed this second fund in 2002. He is now independently exploring inefficiencies in gambling.

Claude Shannon amassed large wealth by recording one of the best investment records. His performance had little to do with Kelly's formula. Between 1966 and 1986, his record beat even Warren Buffet (28% to 27% respectively). Shannon strategy was similar to Buffet. Both their stock portfolios were concentrated, and held for the long term. Shannon achieved his record by holding mainly three stocks (Teledyne, Motorola, and HP). The difference between the two was that Shannon invested in technology because he understood it well, while Buffet did not.

John Kelly was a chain smoking, gun collecting brilliant physicist. He died young at 41 of an aneurysm. He worked closely with Shannon at Bell Labs. Besides being a charismatic character the author does not write much about his life compared to the other two (Shannon and Thorp).

The Kelly formula is Edge/Odds (as explained on page 72). In investment circles, this formula is not always useful because it is hard to quantify your Edge (value of proprietary information). However, Kelly's formula has intuitive practical implications. It entails you should focus on an investment internal rate of return (IRR) instead of its average yearly return. The IRR is always less. Another implication is that higher risk is not always compensated by higher return. There is an optimal risk level beyond which risk taking becomes destructive. The author mentions the Long Term Capital Management as a case in point.

I recommend other excellent similar books: "Fischer Black and the Revolutionary Idea of Finance" by Perry Mehrling, and "When Genius Failed. The Rise and Fall of Long Term Capital Management" by Roger Lowenstein. Both these books describe luminaries in finance and investment fields who were often in contact with Ed Thorp and Claude Shannon. Another excellent book is Sylvia Nasar's "A Beautiful Mind" about John Nash, the Game Theorist.

"Fortune's Formula" tells the story of the Kelly Criterion -through the experiments, ideas, wins, and losses of those who have espoused it and who have derided it, at race tracks, black jack tables, sports books, and, finally, on Wall Street. The Kelly Criterion is a risk management formula published in 1956 by Bell Labs information theorist John Kelly, Jr. that dictates how much of your bankroll you should bet based on your edge divided by the odds so that you will have zero risk of ruin no matter how bad your luck is, while increasing wealth faster than any other betting system. It does not address what bets you should make, which is another matter entirely. Instead of writing a simple analysis of the Kelly Criterion, author William Poundstone brings this story alive by relating the histories of key figures who have used, promoted, or criticized the Kelly Criterion: information theorists, economists, traders, gamblers, and gangsters. Some readers may find this approach unfocused and unnecessary. But I think the personalities lend "Fortune's Formula" an epic quality and place the Kelly Criterion firmly in the context of real life, with real consequences, as opposed to the realm of abstruse theories that never leave the halls of academe.

The men whom "Fortune's Formula" casts as protagonists are Claude Shannon, the MIT scholar who invented information science and who amassed a small fortune as a buy-and-hold investor, typically making 28% per year on a small portfolio, and Ed Thorpe, author of 1962's gambling classic "Beat the Dealer", 1967's "Beat the Market", co-founder of Princeton-Newport Partners fund (1969-1988) and founder of Ridgeline Partners (1994-2002) quant fund. Ed Thorpe's transformation from MIT egghead to black jack sharp to Wall Street wizard in an ongoing theme, as Thorpe is an immensely successful advocate of the Kelly Criterion -and he is still alive. There is unfortunately little information on John Kelly, because he died in 1965 at the age of 41. The key Kelly challenger is 1970 Nobel Laureate economist Paul Samuelson, who probably overstates his case in calling the Kelly Criterion a "complete swindle" when the point of disagreement seems to be the concept of "utility" in long-tern outcome. Whatever one thinks of Samuelson's outspoken arrogance, he is certainly entertaining. Mobster Manny Kimmel makes an appearance, as do traders Ivan Boesky, Michael Milken, and John Meriwether, as well as numerous information theorists.

William Poundstone obviously has a point of view. He is an advocate of the Kelly Criterion, believing that it produces better results than any other betting system over the long haul while withstanding even the most unlikely confluences of catastrophe that have a way of manifesting themselves periodically. He doesn't support efficient market hypothesis. He thinks people can, and have, made money consistently on the stock market. -But that you need a darned good method of managing risk to do it, a better method than Value at Risk reports. "Fortune's Formula"'s fascination is not only in its history of the Kelly Criterion, but in the realization that a risk management formula has so many seemingly disparate applications. As Poundstone says, "The idea pops up in the strangest places."

I found this to be a very exciting and informative book. Once I started, I was hooked: I couldn't put it down because I wanted to learn more -- not just about the formula but all the intellectual controversy surrounding it and the cast of characters involved. This book tells the story of the 'Kelly criterion' and how certain people used it to beat the casinos and earn consistently above-average returns in the stock market. The cast of characters extend from famous organized crime figures (Bugsy, Longy Zwillman, etc.) to Claude Shannon, father of information theory, to Milken, the junk bond king, and many more.

You can view this book as a study on the history of an idea, as a study of the financial markets, or as a social study of gambling and investing. There are many faces to it. I wouldn't say it is a book on finance, or a book on history, but a bit of everyhting really. It definitely is not a book on 'how to beat the market', don't look for that here. The book does give a good explanation of the Kelly formula, in my opinion, but the whole controversy surrounding it, as well as the story of Ed Thorpe, was much more interesting for me.

If you are interested in finance or financial history, math, or gambling, or how ideas evolve and why certain ideas are better known than others, you should read this book. I think that mathematically sophisticated readers can be a bit disappointed because there is no rigorous treatment of Kelly criterion. But the basic ideas of CAPM, efficient markets hypothesis, the essence of Kelly formula, and random walk, are explained quite succintly and clearly, in my opinion.

The only problem I had with the book was that the author went off in tangents quite a bit, which sometimes distracted from the main story. For instance, right after the description of Kelly formula, there is a discussion of Edgar Hoover's horseracing bets. It seems that the author had a lot of material, may be too much, that he did not want to cut, but also did not know where to put. That detracted from coherence somewhat, which is why I am giving this 4 stars. But at the end of the day, I am feeling better for having read the book -- I learned a lot, and it was an easy and interesting read. Definitely worth the money and time.