Discovery

It’s the worst of times to be in the Red Sox Nation and among the best of times for those who relish baseball’s marriage of probability, improbability and dazzling statistical detail.

One week after the premiere of Moneyball, a movie in which baseball stats play an improbable starring role, the Boston Red Sox concluded a late-season swan dive with a similarly improbable last-minute loss to the Baltimore Orioles.

It has Craig Parks’ gears spinning. Of course you’re thinking, hmm, as a WSU psychologist, he must be wondering what went through the Sox players’ minds as they choked so spectacularly. But no, as a fan of both statistical analysis and baseball, he sees several numbers wonders, including those in the race for the National League batting title and another batting title of a century ago.

His thoughts:

The Red Sox are the first team ever to blow a nine-game lead in September but the Braves came close—they were up 8.5 games on the Cardinals when September started.

An even cooler stat thing that almost happened: With two games left to play, the difference between the #1 and #2 batting averages in the NL was .00006. New York Met Jose Reyes broke away from Milwaukee’s Ryan Braun over the last two games and won the batting title outright, but there was serious discussion about what to do if they had maintained that degree of closeness.

Technically one guy would have had a higher average, but it was unclear whether Major League Baseball wanted to go to that degree of fineness to determine the batting champ.  Or whether it was even legitimate to do so, as that difference is pretty much uninterpretable.  Maybe they would have declared a tie, but the situation has never come up in baseball history, so there’s no precedent to appeal to.

Which calls to mind the 1910 batting title.

Statistics do not treat the 1910 batting title of Ty Cobb kindly.

With two games left in the season, Ty Cobb is hitting .385, Nap Lajoie is hitting .377.  League-wide, everyone hates Cobb, including many of his Tigers teammates, and loves Lajoie.  Cobb sits out the last two games, but Lajoie plays his final two, a doubleheader against the Browns (now Orioles).  Browns catcher-manager Jack O’Connor positions his third baseman back on the outfield grass every time Lajoie comes to bat, which means if Lajoie lays down a bunt, there’s no way the third baseman can get to it in time.  Lajoie reaches base eight times across the two games—five times on bunts—though the last bunt is scored as an error, so he officially goes seven for eight.  This raises his average to .384.  So he misses the batting title, but just barely.  Cobb wins it.  (By the way, the O’Connor manager was fired afterwards.)

Cut to 1981.  A baseball researcher discovers that, in 1910, a box score for the Tigers was counted twice in the official calculation of Cobb’s average.  Cobb had gone two for three in that game.  Subtract out that phantom game, and his average falls to .383.  Now Lajoie is the batting champ.

But, if we’re going to start adjusting hit totals, you can argue that Lajoie’s last two games should be thrown out, since it was clear the Browns were giving hits to Lajoie.  In fact, there is some evidence the Browns tried to bribe the official scorer to change that last at-bat from an error to a hit.  Do that, and Lajoie’s average resets to .377, and Cobb is clearly the batting champ.  Also, the 1910 American League president reviewed and certified Cobb’s average, so some people question whether it’s appropriate to retroactively change it, despite the apparent error.  Others wonder whether there would be this much scrutiny if the player in question was someone other than Cobb, who was indeed a nasty, vicious man.

Lloyd Smith spends a lot of time pondering the performance of bats and balls—aluminum, wood, baseballs, softballs. It’s his job, and he does it well enough that his Sports Science Laboratory is the official bat testing facility for the NCAA.

But while the WSU associate professor of engineering might use his ball cannons and high speed cameras to facilitate an arms race of ever bouncier balls and more powerful bats, the lab focuses more on uniformity, or to mix a metaphor, a level playing field among the tools of the trade.

Flickr photo courtesty of MelvinSchlubman, http://www.flickr.com/photos/pauldineen/

In fact, if you ask him what bat is best, he can’t tell you. That would be a conflict of interest, an implicit endorsement of the people he is supposed to help regulate. Moreover, he says, it just doesn’t matter that much.

It’s a common misconception that there is an enormous difference between bats, he says. By design, the highest-performing softball bat is 10 percent more powerful than a wood bat. The best college bat is 5 percent better.

“The big difference is in player ability,” he says, referring to an on-the-field study showing as much as a 20 percent variation between players.

“When parents come to me and say, ‘Hey, which bat should I buy for my kid,’ I tell them, ‘Go to the weight room and work out. Go play the game. Go work on your skills.’ That’s going to make a lot more difference than spending $300 on the latest and greatest bat.”

Then there are the intangibles that lie outside the realm of measurable physics, like bat comfort. Smith can measure 100 bats and determine the best performer, “but if a player is convinced that this other bat is better, what does that psychology do? What factor does that have?”

That may even have been a factor in the use of illegally corked bats. A study co-authored by Smith in the recent American Journal of Physics found a ball bounced better off a solid wood bat than one hollowed out and filled with a material like cork.

It could be that the lighter corked bat improves a player’s ability to turn on the ball, and a player like Sammy Sosa—caught with a corked bat in 2003—was aiming to improve his batting average, not power. Or it could go back to that intangible psychological factor: He thought the bat worked better, and thinking made it so.

“Suddenly superstition does have a reality,” says Smith, “but we can’t really measure that here, so we stick with the science part.”

To learn more, see “The Physics of Cheating in Baseball” at Smithsonianmag.com.