X

Pittsburgh Pirates' Zach Duke: Typifying a Team Enigma

Tom AuSenior Analyst IIMay 19, 2010

PITTSBURGH - APRIL 13:  Zach Duke #57 of the Pittsburgh Pirates pitches during the Opening Day game against the Houston Astros at PNC Park on April 13, 2009 in Pittsburgh, Pennsylvania.  (Photo by: Gregory Shamus/Getty Images)
Gregory Shamus/Getty Images

Besides Paul Maholm, Zach Duke is the longest serving Pirate. And he is still hard to typecast.

On a good night, he is capable of beating a Cy Young award winner like the New York Mets' Johan Santana, or the Phillies Roy Halladay. On a bad day, he can be truly awful.

An example of this was Duke's last outing against Cincinnati. He gave up five runs, two of them homers, in five innings, for a 9.00 ERA.

Last night, Halladay pitched a quality start against the Pirates, two runs in a complete game, a basically "average" performanceβ€”for him. He's done both better (fewer runs in the same nine innings), and worse (the same two runs in fewer innings).

Given that he struck out six and walked one, nine hits and two runs represented an "average expectation."

But on a pro rata basis at least, Duke pitched even better: one run on six hits in six innings, while striking out five. Last night was one of his better nights, meaning that's an atypical performance for him.

And given the shutout relief that has been the staple of the Pirates' bullpen, that was good enough for a win. One might argue that Duke, who didn't pitch as long, wasn't better than Halladay. But Duke and the bullpen clearly were.

The Pirates had just enough offense to win. Back to back doubles by Andy LaRoche and Ryan Doumit accounted for the first run. An RBI single by Garrett Jones, who had three hits, scored the second (Lastings Milledge from second base).

Tonight's matchup, in Pittsburgh against the Milwaukee Brewers, could tell a similar story. The Brewers' Randy Wolf will be facing Brian Burres. They have similar ERAs, between 4.50 and 5.00, but the similarity ends there.

Wolf's ERA has a normal consistency around the trendline. Burres is the exact opposite, with a 0.00 ERA in his best two games totalling 12.1 innings, and a 9.25 ERA in his worst three outings totalling 13 innings. It's either feast or famine against him.

Statistically, such a wide dispersion of outcomes, weighted toward both ends, is a property called skewness. This is a pattern exhibited in an extreme form by Charlie Morton.

But other Pirate pitchers are not far behind in this regard. For now, at least, the team has a whole rotation of "Charlie Mortons."

🚨 SPORTS NEWS ➑️ YOUR INBOX

The latest in the sports world, emailed daily.