Statistical oversight
Relying on the wrong numbers, in baseball and other areas of life, can lead to the wrong conclusions
CALL THIS A CURVE BALL, IF YOU WILL, OR A change-up. It's WORLD's publisher acting as an interloper in territory normally reserved for WORLD's editor-in-chief. It's WORLD's publisher daring to talk about baseball.
But it is, as my colleague Marvin Olasky always reminds me during the late fall and winter months, time for the hot stove league. What else is there to talk about?
I am forced to pull myself up to the warming stove because of a book recommended to me by a pastor friend, a thoughtful theologian who I never dreamed bothered with inconsequential things like baseball box scores. Indeed, because of Moneyball, the book he recommended, neither he, nor I, nor many others will ever again take box scores very seriously.
Moneyball might even persuade you to be more careful with, and more skeptical about, traditional statistics in all walks of life. That is my main point here.
Moneyball, by Michael Lewis, is an artful (and occasionally vulgar) series of vignettes of folks who challenged the conventional wisdom about how excellence in baseball should be measured. Some such challenges come from people who are merely hobbyists of the sport, aficionados who pride themselves on knowing one more detail about baseball than the next expert knows. Other such challenges come from people who have a huge vested interest in baseball-people who stand to make or lose millions of dollars just by getting the numbers right. Oddly, Mr. Lewis seems to be saying in Moneyball, the hobbyists have often worked harder at getting things right.
Actually, this new style of thinking challenged more than box scores. It took on all the conventional wisdom. For instance, baseball insiders absolutely knew that scouts should look to young high-school stars to find the next great players. But anyone interested in reality had to concede what Bill James, the guru of the new school, observed when he wrote that "college players are a better investment than high-school players by a huge, huge, laughably huge margin."
Moneyball is full of such comparisons, and it's fascinating stuff. Two issues, though, stand out for most people after reading Moneyball. First, the comparison of traditional batting averages (which the new school disdains) and of "on-base-and-slugging" averages (which the new school applauds even though such numbers never appear in newspaper box scores). And second, the special focus of the new schoolers on the negativity of outs in baseball. "You're out!" should be a call any thoughtful baseball player should avoid at all costs. The new schoolers see a strategic sacrifice bunt, for example, as a silly waste.
Mr. Lewis tells how the Oakland Athletics, largely following the new-school philosophy, have in the last few years put together highly competitive baseball teams at a fraction of the cost of those teams that have followed traditional statistics. All this has the potential for taking the discussion into some fairly complex levels of higher math. But it isn't necessary to go there to appreciate the main point: Don't get hung up on the wrong numbers.
So if that's true in baseball, how much truer might it be in the rest of life? The November issue of The Atlantic Monthly, for example, devotes a whole section to helping set straight similar misconceptions in the realm of higher education. National correspondent James Fallows discusses one of the basics: "Changes in demographics, technology, and society have saddled the most selective colleges with more applications than they know how to handle. More applications means an ever higher rate of rejections, which makes a college statistically more 'selective.' Perversely, this makes it all the more attractive to the next crop of applicants, and the cycle goes on."
In the same issue of The Atlantic, Nicholas Confessore reminds college-bound students, and their parents, to be careful with statistics from sources like the popular rankings of "America's Best Colleges" in U.S. News & World Report-even though he defends the statistics as accurate. He summarizes at one point: "A university may well be rich in faculty resources [ranked in the popular surveys] and poor in actual teaching [not ranked]."
"I wonder," says Bill James, one of Moneyball's heroes, "if we haven't become so numbed by all these numbers that we are no longer capable of truly assimilating any knowledge which might result from them?"
In what other critical areas do we make this mistake? At what important junctures do we follow traditional statistics and conventional wisdom rather than more significant numbers and real wisdom? Over the next few weeks, I'll be assembling a list of such wrong turns along the statistical road. If you have a favorite example, I'd love to hear from you.
Please wait while we load the latest comments...
Comments
Please register, subscribe, or log in to comment on this article.