Lucky Dogs

Dave Paisley

For the next couple of weeks I?m going to take a look at pitching and luck. We all know that some pitchers are luckier than others in terms of the run support they get from their team. To some degree it?s not just luck at all, as good pitchers tend to end up on good teams with decent offenses. But even on the Yankees or Mariners, the distribution of runs by starting pitcher is very uneven.

But there?s another part that luck plays in pitching success, and that?s how the opposition bunches their offensive production together to get runs. Take ten hits and two walks in a game - fairly typical. Spread those out evenly over nine innings and the pitcher willikely only give up a couple of runs - maybe a shutout if he?s lucky enough. However, concentrate them in a couple of innings and the pitcher will tend to give up five, six or more runs. Same raw material, different result.

We could argue that the former pitcher knows how to scatter hits, while the latter can?t pitch under pressure. There may be a grain of truth in that, but as the myth of the clutch hitter has been exposed, can there be any such thing as ?clutch pitching?? Or to clarify, can a pitcher, by sheer will and concentration, decide when to suddenly pitch better? The answer is likely ?no?.

Defensive plays made behind him , just how the opponents batting order stacks up, who leads off each inning - all of these things have a subtly different effect from game to game and are beyond the control of the pitcher, hence I?m going to attribute them eventually to luck.

In the old days, all we had to evaluate pitchers by was ERA and more recently, WHIP (walks plus hits per inning pitched). But now, through the marvels of the information superhighway (?l Gore) we have access to a more reliable indication of pitching performance - OPS Against. Just as a batter?s OPS is the best simple measure of his efficiency, OPS Against allows us to see what the pitcher is allowing in terms of raw offense.

You would expect OPS Against to correlate very well with ERA, and to prove the point, here are this year?s AL qualifiers:

Note I used a quadratic curve fit because it makes logical sense that the trend would tend to a zero ERA as OPS declines to zero. The quadratic nature of the curve does give a slightly higher correlation, although the area of interest is pretty well approximated by a straight line, too.

From the curve we can see that indeed there is excellent correlation, but there are also a few pitchers whose numbers defy the curve. Take the orange diamond at the top center. That?s Mike Maroth of Detroit, who has managed to give up a decent .712 OPS Against, but features a 5.35 ERA. Contrast that with the red diamond in middle center, and that?s Ryan Franklin of the Mariners, giving up a .711 OPS, but an ERA of only 3.18. I looked at the numbers and could see no good reason why Maroth should be so bad and Franklin so good. Franklin has given up 23 walks, 40 strikeouts and 14 homers (and four more homers on Tuesday night in an 8-4 win - his ERA ballooned up to 3.29) to Maroth?s 17, 37 and 11. If anything you?d think those numbers would favor Maroth.

Other notables way off the line include the green diamond at lower left (Pedro Martinez), the blue diamond mid right (Detroit?s Nate Cornejo, proving it?s not necessarily a Detroit effect), the yellow diamond at top right (Joe Mays of the Twins) and the purple diamond upper right (Brian Anderson of the Indians.)

So I thought, what if I calculated the difference between actual, ERA and curve fit ERA, and combining it with the difference between league average run support and actual run support, I?d have a ?Luck Factor? for each pitcher. Some pitchers have good ERA luck and bad runs support luck and vice versa, but the real interest lies in those who are at the far ends of the spectrum.

You might think that Maroth is the ultimate poster boy for bad luck - what with the undeservedly crappy ERA and what is no doubt brutal run support from his Tiger offense. And you?d be wrong. No, teammate Adam Bernero wins the honors hands down. Like Maroth, he loses a bit in ERA luck, but not as much (.7 to Maroths 1.3). But the real difference is that Bernero is getting just 2.24 runs per nine innings in support, over three runs off league average, for a total Luck Factor of -3.9 runs. Maroth gets almost a run more support, negating his ERA luck advantage for an LF of -3.5.

Believe it or not (and I know you will), the next two are both Detroit starters, Gary Knotts (-2.6) and Jeremy Bonderman (-2.6). In the case of these four guys, the luck is in being in the wrong place at the wrong time. Maybe even the wrong planet. Looks like this may have been a good year for them to spend a year dead for tax purposes instead.

Not all the bad luck pitchers are on bad teams. The next guy in line is Ted Lilly of the A?s, with his +.4 ERA and -1.7 run support for a total LF of -2.1, followed by Bartolo Colon (-1.8), then teammate Mark Buehrle (-1.7) and Jarrod Washburn (-1.5). These guys can pray to the law of averages that things start evening out soon.

The guys who are currently flipping the bird at the law of averages are headed by Roy Halladay, with his .6 ERA bonus and 2.2 run support bonus for a total Luck Factor of +2.8. I?m sure Halladay is whistling quietly hoping the law of averages doesn?t notice him. Next up is Rick Helling with +2.5 runs of luck and David Wells and Ramon Ortiz with +2.4. That torrid Toronto offense is as much a bonus as the Tigers offense is a boat anchor, and so we have Mark Hendrickson and Cory Lidle next with +2.1 units each. The Mariners? Freddy Garcia also checks in with +2.1, but he actually has bad ERA luck, combined with very generous run support. That run support has been patchy, though, as he has had games where his offense got him 15, 13, 12 and 11 runs, but also games where they garnered only 0, 1, and then 2 runs three times.

So, to sum up, if you?re looking for guys to fall off a cliff the rest of the way, these guys areas good candidates as any.

about the author

Spare a passing kind thought for the beleaguered Detroit pitching staff. Send you virtual sympathy cards to Dave Paisley at drdjp@strikethree.com and he'll be sure to forward them on.