##
Non-Stochastic Reading *June 17, 2007*

*Posted by Brian L. Belen in Academically Speaking, Books, Reviews.*

trackback

trackback

Finding a good book on heady topics such as mathematics or science is no mean feat. Often, books on the subjects are either too technical that they themselves require a strong background in the field just to understand them, or sacrifice too much content in an attempt to become accessible to the casual reader. As such, any book that manages to demistify the complex concepts behind these subjects without lacking in depth categorically must be a gem.

In search of light yet substantial reading on probability, I thought — nay, hoped — that Michael and Ellen Kaplan’s *Chances Are…(Adventures in Probability),* would be that gem. Sad to say, it wasn’t; but not for lack of trying on the authors’ part.

The trouble with *Chances Are* is that it is, depending on how one looks at it, only tangentially about probability. Yes, many of the key concepts surrounding statistics and probability are covered and their origins discussed. Yet in the desire to present a manuscript about “adventures in probability”, the authors have put together a narrative that, while interesting, is also obtuse and long-winded. As such, the book comes across more like a reader on the applications of thinking in probabilistic terms in the context of specific issues (see below), many of which, ironically, can be dissected more simply without the very discussion on probability that is the book’s *raison d’etre*.

*Chances Are* is divided into several chapters that give the authors occasion to elucidate upon the important applications of probabilistic reasoning. In terms of what one would expect from a book on the subject, the strongest of these are the insightful introduction that sets the tone for the overall piece (“Einstein famously remarked that he did not believe God would play dice with the Universe,” the Kaplans write, “The probabilistic reply is that perhaps the universe is playing dice with God.”), as well as a thoroughly entertaining chapter on gambling, certainly the most natural topic to cover in any discussion on probability. The rest of the book may be said to go steadily downhill from there.

Downhill, that is, for the reader looking for a solid non-textbook on the nuts and bolts of probability theory. That *Chances Are* isn’t, mostly because from a broader perspective there is no sense that the book makes for an organic whole. One might say this is due to the authors’ writing style: they inject unnecessary erudition into the text at largely inopportune moments to the point that the discussions come across as forced, perhaps even pretentious. For instance, there is a terrible chapter on “Fighting” that attempts to present rudimentary game theory in probabilistic terms only to end up being quite convoluted. Notwithstanding these excesses, there are nuggets of wisdom to be gleaned from the book when each chapter is considered on its own. By way of example, a chapter on “Healing” that discusses health care and medicinal trials is actually quite engaging (particularly if one has an interest in econometrics), and there is a truly thoughtful chapter on “Judging” that goes through how probability should and shouldn’t factor into legal decisions. But to pick up the book and expect more than missives peppered with some thoughts on probability is a formula ripe for disappointment.

In many ways, the book suffers from the inherent problem of its subject: probability tends to be difficul to grasp, often relegated to second-class status even as a tool of analysis. As the authors themselves point out, the human mind seems to be wired to think more in terms of absolutes; uncertainties — and what is probability if not a science of uncertainty — tend to be a tad unnerving. Yet the authors fall into the very trap they seek to overcome: if *Chances Are* shows anything, it is that thinking probabilistically can sometimes be more trouble then its worth, that there are simpler ways to go about solving problems even if these are less accurate (though that is a conclusion that cries out for some probabilistic analysis; but I digress).

The sobering conclusion is thus that the book simply does not quite live up to what the authors evidently have meant for it to be. This is especially true given the obvious effort to explain otherwise complex statistical concepts that somehow falls short of being satisfying (there’s also no list of references or footnotes throughout the text; a shame, that). On its own merits, *Chances Are* will perhaps appeal more to liberal arts majors with rusty math skills and who enjoy a good yarn, but not so much to those seeking a straightforward book on probability albeit couched in layman’s terms.

[…] is a follow-up to a prior entry on “Non-Stochastic Reading“. To read the earlier piece, click […]