The basic premise of this book is to long stock and delta hedge w/ out of the money warrants. This was great when the tax incentives encouraged corporations to sell bonds w/ detachable warrants rather than selling convertible bonds. In addition, valuing the warrants was tricky prior to the use of the Black-Scholes model. This strategy - delta neutral covered calls - is profitable when the market has peaked out but you can get your handed to you if you employ it in a scenario such as '99-'00. Further their reocmmendations that you short more as your warrants fall is very dangerous - shorting a lot of gamma. The warrant game played itself out and the authors made a a lot of money. This is a very interesting book written by a very profitable hedge fund manager, but I would not recommend attempting to replicate this strategy w/ LEAPS. Pricing is much tighter now so your margin of error has dramatically decreased
*The strategies in this wonderful book are only employable in Asia now, and in very limited ways (often insurance and bank stocks and bonds) but you take a lot of liquidity risk, so I'm not sure if you are still paid alpha over an expected return.
Still, a marvellous read. Pre-dates the Black-Scholes by five years, but in a replicating portfolio no-arbitrage method (which implies a lognormally distributed expected equity return) which Thorp then correctly pointed out was arbitrageable.
This book also serves as a curious filter rule. Those who read this and understand the old world and Thorp's method most likely can see current methods and models and break them down and differentiate them into tractable and fantasy. Credit structures who've relied on standard cash-flow and default probability metrics would have done well to start with Thorp to see how what they construct can be de-constructed by clever boots who see both the strengths of the original construct, and the copula methods and correlation assumptions in the structure (and its decay) to make arbitrage opportunities. In other words; if they read Thorp and "get it" they have a lower likelihood of being hoodwinked going forward.