The Astros' offense has been receiving most of the attention lately.  But the ability of the Astros to stay in contention probably will come down to the starting pitchers.  The Astros' starters as a group have not performed badly, as least compared to expectations.  But  every starter has some question marks, and this leads many to wonder if the rotation will hold its own as the season progresses.  My view is that Roy Oswalt (tonight's pitcher) holds the key to answering that question.  If he can return to form as the dominating ace, his improvement will offset possible declines elsewhere in the rotation. 

As I ponder the starting pitching, I perused some of Bill James' pitching stats.  His on-line web site has many unusual or unique stats, and I have only begun to make my way through them.  So I will make some comparisons of Oswalt, Sampson, Chacon, and Backe using Bill James' subscription data (billjamesonline.net), and see if it is helpful to understanding how the starters are doing.  I left out Wandy and Moehler/Cassel because they have pitched so little that the numbers likely aren't useful.

Percentage of Base Runners Scoring

• Oswalt 40%
• Chacon 35%
• Backe 30%
• Sampson 43%                                                                                              
  

Oswalt and Sampson seem to have some problems preventing runners from scoring.  I wondered if this reflects that they are more likely to have  baserunners with no outs, but that doesn't appear to be the case.  When you break down the scoring by how the runners got on base, the numbers are even odder for Oswalt and Sampson.  Both pitchers have a high percentage of singles turn into runs (36% for Oswalt, and 50% for Sampson, compared to 22% and 29% for Chacon and Backe, respectively).  When Oswalt allows a single, they are scoring at nearly the same clip as doubles (36% vs. 38%) and Sampson's singles score at a higher rate than doubles (50% vs. 32%).  Backe and Chacon show a more normal difference between the probability of singles and doubles scoring. Oddly, for Oswalt, a single is three times more likely to result in a score than a walk.

Perhaps this just reflects bad luck for Oswalt and Sampson.  I'm not sure what conclusion to draw.  I have a suspicion that Oswalt's and Sampson's above average groundball rates may have something to do with the high percentage of singles scoring.  Could they be facing bad luck in terms of groundballs finding holes? Sampson's BABIP is very high (.366) which might support that view. However, in watching Sampson, many of the groundball hits are hit hard.  I suspect that Oswalt may be allowing too many HRs after singles.  Of the 13 HRs allowed, 7 were solo jobs and 5 followed a single. Only 1 HR occurred with a runner in scoring position.

Shutout and High Scoring Innings

Percentage of Innings: Shut Out and 3 or More Runs

Shutout Percent first, with High Scoring Inning Percent in Parentheses

• Oswalt 66% (7%)
• Chacon 68% (3%)
• Backe 73% (8%)
• Sampson 64% (10%)   
  

This data tells us something. Backe  stays alive by cruising along with shutout innings.  One of Chacon's keys to success is avoiding the big inning.  A major problem for Sampson is allowing a high scoring inning; the fact that Sampson is the only starter to allow a 6 run  inning is indicative of his problem.  Sampson and Chacon have the highest percentage of 1-2-3 innings (38% and 37%, respectively), and since they are the best and worst performing starters, that stat doesn't seem to mean much.    

First Batter of the Inning                   

Sampson, Backe, and Chacon have similar stats for retiring the first batter of the inning (63% - 66%).  Oswalt is more proficient at retiring the first batter (75%).  And Oswalt's and Sampson's results are more strongly affected by whether they allow the first batter to reach base in an inning.  If Oswalt retires the first batter, his RA/9 for that inning is 2.51.  If he allows the first batter on base, his RA/9 shoots up to 14.40.  The comparable numbers for Sampson are 3.24 (first batter retired) and 12.38 (first batter on base).  By contrast, Chacon's RA/9 in an inning with the first batter retired is 2.50 and 9.00 if the first batter gets on base.  As Oswalt pitches, watch for the outcome on the first batter in an inning.

If you have any thoughts on these stats (unless you find them boring), make a comment.