[DesertCat]

Having some minor relation to the topic, and containing poker strategy!, today's Wall Street Journal book review.

[ QUOTE ]
The Right Way to Choose
By DAN SELIGMAN
November 30, 2005; Page D12

We all make bad decisions at times, even those among us who are Nobel laureates. Evidence for that statement appears on page 33 of "Making Great Decisions in Business and Life" by David R. Henderson and Charles L. Hooper. They recount the tale, notorious in certain circles, of the laureate in economics (unnamed) who was to be guest of honor at a dinner party sponsored by a university. A professor was sent over to the honoree's hotel to escort him to the party. As they were leaving, the eminent economist announced that he had a headache and wished to stop at the hotel's small store to buy some aspirin. But the store's price was $4.95 a bottle, which he refused to pay, demanding instead a visit across town to a "regular drugstore" whose price was said to be half that figure. Envisioning disastrous delays, and presumably marveling at the obliviousness of the laureate to the cost of the search time, the professor instantly put up the $4.95 himself.

Ignoring search costs is one of several obstacles to wisdom encountered in this lively work by an academic economist (Mr. Henderson) and a marketing consultant (Mr. Hooper). To get to "great decisions," they draw on economic concepts, on the discipline called "decision analysis" -- which is mainly about making choices in the absence of good information and in the presence of risk -- and on common sense.

Lessons even Nobel laureate economists sometimes forget.

The book comprises 15 chapters, each centered on some thinking pattern that leads to poor outcomes. It rapidly becomes clear that, like the laureate with the headache, most Americans tend to have problems with costs that are hidden or at least not intuitively obvious. A prime example is "opportunity costs," which should be part of every family's calculation about the cost of college: It's not only tuition, room and board -- it's also the income that the student forgoes by not exercising the opportunity to work full-time during the academic year. Another "hidden" cost arises with insurance. Most people think of premiums as the sole cost to be considered, but Messrs. Henderson and Hooper require you to think also about the cost of not insuring valued objects, which is the value multiplied by the percentage probability of their being destroyed.

A common source of cloudy thinking is "sunk costs" -- i.e., the investment you have already made in some venture. The book cites a coin shop in Kulpsville, Pa., whose owner kept losing money -- the city simply wasn't large enough to support his business -- but refused to move because he had already sunk so much into it. But sunk costs are always irrelevant to current decisions, and you can cite this principle as your reason for walking out on boring Broadway shows whose tickets cost a bundle.

"Making Great Decisions" is especially interesting on choices made with incomplete information. Do you buy a $350 nonrefundable airline ticket now for a future event that may or may not be canceled or do you wait until the last minute, when the price will be $1,200? Since nobody wants his ticket cost to be a complete waste of money, there is a bias in favor of wait-and-see. In this example, however, it pays to buy the ticket early unless you can plausibly tell yourself that the likelihood of cancellation is as high as 71%. But even if you have absolutely no clue about the likelihood of cancellation, you should still buy the early ticket. In all such situations -- when you must make a choice but have no information about a probability -- you should estimate it at 50%. You are at least limiting the extent to which you can be wrong.

A recurrent theme in the book is the importance of focusing on the right objective, which is not necessarily the objective that's on your mind when you're making a decision. The authors come down hard, and properly, on people who confuse their economic interests with their moral and ethical positions. Writers and intellectuals are especially prone to mess up this way. A dozen years ago, my wife was seriously injured by a company van running a red light. Given that our medical costs were covered by insurance, and given also my own repeated writings deriding "only in America" litigiousness, I made the dumbest decision of my adult life: I decided not to sue the company that owned the van. Luckily I was brought to my senses by a friend who heard the relevant details and asked whether we really wanted to throw away $60,000.

I had just one quibble with the authors of "Making Great Decisions." In a discussion of gambling, they argue that probabilistic decisions cannot be judged solely by their outcomes. You may, after all, make the right decision but lose anyway. OK so far. But then they exemplify the point by describing a poker game in which a player draws a queen-high straight flush, bets everything he has got and is beaten by a royal flush (which is ace-high). Their conclusion: The player's loss was a fluke, and if he ever again gets that hand he should again bet everything he has got, because only six hands out of 2,598,960 can beat him, so the probability of his winning is 99.9977%.

Realistically, the odds are not anywhere near that figure. The calculation assumes that the player's straight flush is competing with a random selection of poker hands. But that's crazy: The weak hands will have long since dropped. Based on our player's previous betting and the size of his final bet, there is virtually no possibility of his getting a call from players who do not also have strong hands. A larger problem is that the analysis focuses on the player's probability of winning the hand. It should focus instead on maximizing his income. With this insight, we instantly see that the real problem of players with tremendous hands is getting the other players to bet with them. Betting everything you've got is not ordinarily the best way to accomplish this. There is a lesson here for life beyond poker, if there is such a thing.

[/ QUOTE ]