Talking Sabermetrics: Looking at underperforming AA and AAA pitchers

Last week, I discussed three Astros pitchers who are performing better than their ERA, according to SIERA. The pitchers were J.A. Happ, Rhiner Cruz, and Fernando Rodriguez. This week I will spotlight several of the organization's AA and AAA pitchers who are performing better than their ERA, according to SIERA. The SIERA calculations for minor leaguers are available at minor league central.

SIERA , "skill interactive ERA" , is an advanced pitching metric, scaled to ERA and similar to FIP, which has been shown to be a reasonable ERA estimator, particularly for those interested in projecting future pitching performance. SIERA is based on the pitcher's skill exhibited during a performance period, reflecting the interaction of strike outs, walks, and balls in play. SIERA is better than ERA and FIP at estimating the succeeding year's ERA. Of course, we have less than half a season of data for SIERA in 2012, which may limit the accuracy of SIERA in projecting future changes in the players' ERA. However, the SIERA sample may provide us a divergentt perception of these minor leaguers' performance in AA and AAA this year.

The earned run average for each of the pitchers, below, appears disappointing on the surface. However, SIERA tells a different story---namely, that the pitcher's skills have been much better than their ERA.

Some observations:

• Abreu and Zeid are relievers who appeared to have pitched poorly, based on their ERA in repeating the same level this year. However, SIERA indicates that they both have pitched fairly well. Abreu and Zeid came over in the Bourn and Pence trade, with the idea of providing young bullpen reinforcements. Maybe we should expect some improvement which eventually translates to relief pitching at the major league level. Like Abreu and Zeid, Valdez throws hard out of the bullpen. Given his SIERA, he could provide immediate help in the major leagues if a reliever has to be replaced.
• Oberholtzer started the season with some poor results, but his SIERA is quite respectable and seems to confirm conroestros' conclusion that Obie has turned the corner, in terms of improvement. Conversely, Clemens started the season with good results but has had some tough games in recent months. Many of us were hoping that Clemens would "ace" AAA and possibly make an appearance in the Astros' rotation this year. That may not happen, but it is reassuring to see that Clemens' SIERA indicates that he has pitched better than the ERA results.
• Aneury Rodriguez is 24 and Jake Buchanan is 22, and both pitchers provide young pitching depth. Again, SIERA says that both were better than their ERA would suggest.

DIPS theory is an important sabermetric concept which tells us that pitches have relatively limited control over the results for batted balls in play. A corrollary of the DIPS concept is that pitchers will tend to regress toward the major league average BABIP (batting average on balls in play). Using BABIP as a predictive tool at the minor league level is trickier, since the range of skill levels in each league may vary more than in the majors. But a higher than average BABIP could reinforce the SIERA results.

The PCL (AAA) average BABIP is .329. The Texas League (AA) average BABIP is .310. Paul Clemens' and Aneury Rodriguez's BABIP at AAA: .361 and .388, respectively. Oberholzer's BABIP in AAA is .361 and his AA BABIP is .311. Jake Buchanan's BABIP in AA is .346. Josh Zeid's BABIP is .317 in AA. For Clemens, An-Rod, and Buchanan, the BABIP provides support for the possibility that bad luck and/or defensive weakness contributed to the bad results.

Perhaps this sabermetric exercise will make you re-think your perception of these seven minor league pitchers' performances this season. At the least, we have some room for optimism about improvement in their ERA.