Why Seamheads Keep Under-Rating the O's

In the last three seasons, the Baltimore Orioles have punished their AL East brethren, tallying 93, 85 and 96 wins, and snagging two division titles.

And so this year, once again, the stat-based projection systems are showing the franchise less respect than Aretha Franklin. PECOTA pegs the Birds at 78 wins and a basement finish for the fourth straight season. Not a finished basement, mind you.

The Birds have unintentionally uncovered one way in which projections systems and the seamheads who worship them are for the birds themselves. (But not for the Birds.) Because of pretzel logic baked into the projection system by stat guys who call everything they can't quantify "luck," PECOTA holds a deep and abiding hatred for everything Baltimore, including crabs, Terrapins, Edgar Alan Poe and being called "hun."

In this article for Baseball Prospectus, Zachary Levine explains what the Orioles did to PECOTA (pulled its hair and made it cry in fifth grade) to consistently earn a subterranean ranking. How does PECOTA suggest the Orioles will shed 18 wins from last season? Here's the critical segment of the explanation:

The first five wins are easy. That’s the luck that the Orioles had last year that you would expect to be close to independent year-to-year. A 96-66 team, they had a Pythagorean record based on runs scored and allowed of 94-68 and a third-order record based on the underlying stats that go into run scoring and prevention of 91-71. 

Translation: the O's presented a unique profile in 2014, in which the close matches tilted their way and the blowouts blew in the other direction. Generally, that's more likely luck than skill. Bill James discovered in a previous century that teams are much more like their run differential than they are like their won-loss records; the run differential is a better predictor of future record than past record.

The discrepancy between the won-loss record suggested by run differential (technically, the difference between the square of runs scored and the square of runs allowed, hence the notion of a Pythagorian record) and the actual won-loss record is commonly referred to by sabergeeks as "luck."

But sometimes what appears to be luck is skill lurking in the corner of the dugout. The Os under Showalter have vastly out-performed this measure, which suggests that they know something Bill James doesn't . (Is that possible?) In 2012, Baltimore posted an astonishing 29-9 record in one-run games and 16-2 record in extra-inning affairs, and they repeated the feat, albeit less dramatically, in 2014.

This could be a run of luck akin to flipping heads 17 of 19 times. Or it could be that the team knows something that gives them an edge in close contests. A sharp manager, a good bench and a deep but unheralded pitching staff, all of which Baltimore possesses, could convey that needed margin in close games without tilting the balance in less evenly contested skirmishes.

This is a bit like the "pitching to the score" argument people make for Jack Morris and his pedestrian 3.94 lifetime ERA, so we have to be careful. But PECOTA and its brethren have been wrong for three straight years. Maybe it's time to acknowledge that just because we don't know what it is doesn't necessarily make it luck.

Bet the over.