This was a pretty nifty win for the Bucs. Zach Duke is reaching the point where he deserves a little slack--I'm skeptical of his ability to continue to post good starts with his ugly K/BB numbers, but so far he's doing his job. Tonight he posted 7.2 unspectacular but effective innings; he's now gone at least six innings in seven of his last eight starts (as Thunder pointed out in the gamethread). That has value. Basically, he's like Kirk Rueter plus some groundballs and minus some homers right now. If Rueter hadn't worked so fast, he would've been about the most boring pitcher ever, but he got the job done for a long time. 

When the theory of DIPS--the idea that pitchers had little control over what happened once balls were put in play--was introduced several years ago, people really marveled about it, and one of the arguments against it was Kirk Rueter, who (it was argued) couldn't possibly put up such low ERAs if all that mattered were strikeouts, walks and homers allowed. I can't find the article now, but someone did a study that suggested (if I remember correctly) that Rueter was able to get away with obscenely low strikeout rates because he was a very intelligent pitcher who attacked batters with the bases empty, resulting in bunches of solo homers and few bases-empty walks, then tried to paint the corners with runners on, resulting in some walks with runners on but few homers. He was thus able to avoid big innings and keep runs off the board. I checked his splits and this appears to be basically true throughout the portions of his career when he was good (from 1996 through 2003). In 2004, this stopped working and he found himself out of the league pretty quickly. 

Duke seems like a pretty sharp guy too, so I checked his splits. He doesn't have the problems Rueter had with homers, but I wondered if he might be just throwing the ball into the zone kind of indiscriminately with the bags empty in order to prevent walks, then trying to be more fine once there were men aboard.

Overall, major league pitchers are allowing a .254 batting average against this year with none on. That number rises to .267 with runners on. In 2007, it was .262 with the bases empty and .276 with runners on. In 2006, it's .264 and .276. So there appears to be some consistency--pitchers allow slightly more hits, on average, with runners on.

Duke's average against with none on this year is .326, thanks to the Pirates' awful defense and the fact that he never strikes anyone out. But his average with runners on is only .293 (and with less power, too). His walk rates are about the same in both situations.

This is the only year of Duke's career where the trend holds, however. In fact, in 2007 he allowed a .337 batting average with no one on and a .385 average with runners. Of course, when the numbers are that bad, you may have to just say he had no control over anything and leave it at that. (Those remarkable averages show what happens when you mix a horribly confused pitcher and a terrible defense. Water into acid. Kablooey!)

So it looks like Duke's success so far this year has been the result of a pretty Duke-specific application of the Rueter method. Whether this is a skill Duke has acquired or whether it's just the result of small sample size remains to be seen. I'm pretty sure it's the latter, although I'm mildly encouraged by the fact that Duke has basically been even better this year when pitching with two men on instead of one, or with two outs. My guess, though, is that Duke's ERA will inch up as the year goes on. His groundball tendencies will keep him from being too terrible, but ultimately it's going to be very hard to maintain a 4.19 ERA while striking out only three batters per nine innings.