Pittsburgh Pirates: Why Joel Hanrahan and Not Evan Meek Was Chosen as Closer

After the Pirates traded their (signed) closer Octavio Dotel for starter James McDonald and outfield prospect Andrew Lambo, they ran a race for closer between Evan Meek and Joel Hanrahan, which Hanrahan won.

We now know that this was the right choice, and that Meek doesn't have the stamina to pitch even the EIGHTH inning.  But the preseason move caught some people, including yours truly, by surprise, because Meek had the better ERA last year.

But other factors come into play. Sometimes a reliever with a higher (worse) ERA will be given the closer role because he is more reliable. For instance, while Hanrahan gave up one run in more innings than Meek late last year, he would give up two runs fewer times. It's easier for a team to compensate for a one- than two-run loss in a "save" situation.

In fact, ERA did not fully measure the abilities of the two men because incorporates a number of short-term "luck" factors, and therefore is not the best measure of pitching ability. A better measure is called FIP (field independent pitching) statistics, or what we call "sabermetric ERA." On this metric, which incorporates home runs, base on balls, and strikeouts, Hanrahan was better last year, 2.62 to 3.45 for Meek.

A pitcher's job is to give his team a chance to compete, which means avoiding "free" (home) runs and bases (on balls). Otherwise, a contact of bat to ball results in in the latter's landing somewhere in the park, where it might be successfully "fielded" for an out, or not.

A pitcher's second job is to get outs while avoiding contact if possible, through strikeouts. If a batter completely misses the ball, there is no chance to put it into play and possibly score a hit or run.

Hanrahan handily beat Meek in this department last year, getting nearly 13 strikeouts every nine innings (27 outs) pitched, versus only eight for Meek. On the other two metrics, base on balls and home runs, the two pitchers were similar.

The reason why Meek had the better ERA, last year despite allowing batters more contact with the ball, was because of a better BABIP (batting average of balls on play). But this can be attributed to small sample size and was rightly discounted by Pirates' management.

Over time, there has been an average BABIP of around .300 across a wide spectrum of pitchers and batters. But this is a LONG RUN average; it takes about three full seasons for a starter's numbers to converge to this figure, and two or three times as long for a reliever.

Hanrahan's BABIP was. 329, a bit more than usual, while Meek's was .224, meaning that he had been "lucky" (relative to .300). Over time, one can assume that the two will converge, meaning that Hanrahan was the better closer, after controlling for this factor.

Management has made some inspired choices this season: Hanrahan for closer, Kevin Correia for opening day starter, both against the "gut feelings" of this writer.  "Sabermetrics" (baseball science) is good for supporting these decisions and for keeping decision-makers "honest."