clock menu more-arrow no yes mobile

Filed under:

Behind the W/L Percentage for 08 Astros

While you are working on an appropriate sabermetric challenge for Dying Quail (see story below), I will write something closer to a statistical rambling regarding the 2008 Astros, based on perusing the Hardball Times' team stats page and's latest installment on PMR results.

We all know the question mark raised by the Astros pythagorean record in 2008, i.e., the pythag stat for the 08 Astros indicates that the team should have been a 77 win team instead of the 86 wins that the team actually achieved.  The significance and uncertainties of such results has been argued at Crawfish Boxes previously, and I won't open that can of worms again.

But a few other stats on the HT team stats page may tell us how the Astros achieved their W / L record.  In "close games" the Astros were 41-31, giving the Astros a 57% winning percentage in close games, which is even higher than the team's overall win percentage. Presumably, the Win Probability Added (WPA) stat may help explain that result. The Astros' win probability added score, .+5.50, was achieved in large degree due to clutch bullpen work.  The WPA for the offense (+2.27) and the starting pitching (-2.33) virtually offset each other. The WPA for the bullpen, though, was a strong +5.56.  The only NL bullpen with a higher WPA was the world champion Phillies (who happened to have a reliever named Lidge).  The next closest bullpen was the Dodgers (WPA 4.68).

Within the bullpen, most of the positive WPA was created by Sampson, Hawkins, Valverde, and Geary.  Byrdak and Brocail also were substantial positive contributors.  Now where will the Astros get similar clutch pitching in 2009?  Sampson underwent arm surgery and is a question mark for 2009.  Hawkins pitched out of his mind, and most likely can't keep up the pace from his 08 Astros tenure.  Brocail may or may not return, given his free agent status.  I don't know if we can depend on Byrdak for a repeat performance.

And that brings up the issue of Valverde.  The rumor is that the Astros are shopping Valverde.  However, it's clear that Valverde is a critical link to maintaining bullpen strength, particularly given some of the question marks.  Given the role of the bullpen in achieving the Astros' winning record in 2008, it seems to me that the team's ability to contend may diminish significantly without Valverde (unless perhaps the trade becomes a vehicle for acquiring Peavy or clearing the deck for Sheets...something which isn't apparent at all).  Put another way, if Valverde is traded, the Astros would need a different formula for winning than the team used in 2008.

And, after Justice's column about trading Wiggy and Valverde, a Valverde to the Indians trade rumor has made the rounds, with some fans suggesting a Valverde for Shoppach trade.  Before you jump on that bandwagon, I suggest you read this article about Shoppach...particularly the part about adjusting his 2008 offensive stats for luck.

Now, turning to the defense topic....

Previously, I have mentioned the defensive metric, Probablistic Measure of Range (PMR), which is rolled out in installments at  PMR develops deficiency efficiency measures which reflect how difficult the batted balls are to field. This installment shows the quality of the defense behind various pitchers.

As shown there, Roy Oswalt was 9th best in terms of favorable defense backing him up. Moehler also had significantly above average defense behind him.  Most of the remaining listed Astros' starting pitchers were closer to average or below average in terms of supporting defense.  Oswalt joins many elite pitchers who seem to enjoy superior defense.  This raises a question (no answers, though) in my mind about the assumptions behind DIPS (pitchers have little control over batted balls).  Notice that some starting pitchers (Oswalt, D-Mat, Wang, Halliday) who have spurred debate in the past over their seeming ability to defy DIPS end up in the upper level of supporting defense. Are they inducing less well hit contact? The predicted DER might be a way of investigating this issue further; however, we would have to separate groundball and flyball pitchers, since the difficulty of fielding flyballs is generally less. (And  no, I haven't done that.)

Defense behind the Astros' starters (100=average, or predicted equals actual).

Oswalt   104.62

Moehler  103.69

W. Rodriguez  99.74

Sampson  99.69

Wolf  97.91

Backe  97.87

Wolf's PMR backing is based on both San Diego and Houston, which makes it harder to judge.  Wolf's predicted DER suggests that he may be inducing batted balls which are easier to field, even though that is also hard to judge since he is a flyball pitcher.  Backe had the worst defense behind him, according to PMR.  Backe seems to slow down the pace and throw a lot of called balls when he is in trouble, and perhaps this diminishes the alertness of his defenders.  Oswalt is definitely a fast worker who puts the ball in the zone; my recollection is that Moehler is similar in that regard.  Maybe this substantiates the well worn saying from broadcasters that fast working pitchers keep the fielders on their toes.

So much for ramblings.