Printer-friendly

Contact Author

January 18, 2013

Overthinking It

The Scurrilous Lie About the WBC

by Ben Lindbergh

Baseball players are often described—or describe themselves—as creatures of habit. And at no time is their adherence to routine more evident than during their methodical preparation for the season, when they shake off a winter’s worth of rust and ramp up for the coming campaign. Pitchers, especially, are dependent on spring training to build up arm strength, incorporate new offerings into their arsenals, and learn to work with their batterymates. But every three (or, starting in 2009, four) springs, including this coming one, an event takes place that threatens to disrupt that routine: the World Baseball Classic. The timing of the tournament has caused concerns that pitchers who choose to participate in it could be adversely affected, which likely explains why the United States squad that was announced on Thursday, while stocked with star position players, is relatively short on impact pitching talent.

There’s some basis for this fear. In May of 2006, not long after the first World Baseball Classic concluded, Nate Silver observed that the pitchers who’d taken part in the tournament had performed far worse to that point in the regular season than PECOTA had predicted, prior to Opening Day, that they would. The pitchers had posted a collective 5.08 ERA in over 1,000 innings compared to a projected mark of 4.10. Some of the difference was attributable to a higher-octane league-wide offensive environment than PECOTA had anticipated, but even after accounting for that discrepancy, the WBC pitchers’ performance still stood out as particularly poor. Starting pitchers with WBC experience began the regular season with especially disappointing results.

No pitcher threw more than 17 innings in the 2006 tournament—in 2009, the high was 20—and pitch counts were strictly regulated. (In the 2013 tournament, pitch counts will reportedly return to their 2006 maximums of 65 pitches per game in the first round, 80 in the second round, and 95 in the semifinals and finals, after rising five pitches per round in the 2009 event.) Given those light workloads, it seemed like a stretch to say that the WBC pitchers’ lousy start to the season as a group could be blamed entirely on overuse. Instead, Silver proposed an alternative explanation:

I think the problem is that the pitchers had been taken out of their routines. They didn't get to see their trainer, pitching coach, or manager for a couple of weeks, they weren't able to relax in the low-pressure environment of the Grapefruit or Cactus Leagues, and so forth. The pitcher may have been throwing in a relief role for his national club, when he hadn't done that in the major leagues in years. If a certain pitching coach doesn't like his pitcher to throw breaking stuff early in camp, or his manager wanted to pull him out after a high-effort inning, there wasn't the opportunity to play it safe. This is why I think that the starting pitchers have been disproportionately affected as compared to the relievers—starting pitchers are creatures of habit and routine.

As worrisome as that article was, its findings were based on a fairly limited sample of performance—roughly one sixth of a season for 58 pitchers. Now we have much more data to go on: not just the rest of the 2006 season, but the entire 2009 season, too. So has the pattern of WBC pitchers struggling in the regular season persisted over time, or were the initial concerns overblown?

To find out, we can generate “retro” projections for each pitcher who appeared in one of the World Baseball Classics and went on to pitch in the majors in the same season. Such a projection for, say, Felix Hernandez, who pitched in two games for Venezuela in the 2009 tournament, would tell us what his 2009 regular-season stats would have been projected by PECOTA to be, based on his pre-2009 performance and scaled to the 2009 major-league run environment. By comparing each pitcher’s pre-season projection with his actual performance, we can assess whether the group as a whole underperformed expectations.