The New York Times > Sports > Baseball > Keeping Score: Baseball’s Leading Man of Math Has Some Second Thoughts About the Numbers
In February SABR’s Baseball Research Journal published a Bill James story called “Underestimating the Fog” in which he racants his previous assertions that there are no such things as clutch hitting and hot and cold streaks. His reasons are mature, humble, and well worth noting, especially because they illustrate all sorts of problems SABRmetrics have encounted because of small sample sizes and the “fogginess” of the baseball data.
James wrote: “We ran astray because we have been assuming that random data is proof of nothingness, when in reality random data proves nothing.”
If the intensely complicated calculations required to translate player performance in each stint in each ballpark is used to rationalize entire careers, but if much of that data (or the resulting translations) is actually noise, how conclusive can any conclusions be?
The NY Times link above will turn into an advertisement for you to pay to read the story in their archives in a couple of days. University of Nebraska publishes BRJ, but their website doesn’t seem to have a link to this edition. One reason to get quickly to the Times’ story is because David Leonhardt’s conclusion strikes me as right, in re clutch performance.
I think so because John McEnroe once said to me in an interview about rising to win the big points, “The best players win the big points, because they’re the best players.”
James is probably right about the fog, but oddly that really illustrates more of the problems with Win Shares than it does the existence of clutch hitting.