2010 Pythagorean Wins with BaseRuns
April 29, 2010 20 Comments
Matthew Carruth made some excellent points in today’s Fangraphs article about using BaseRuns as a “sanity check” for a team’s performance at this point in the season.
The theory is that you can break runs scored and allowed into their components and remove some of the “noise” in projecting wins based on run scoring. BaseRuns is probably my favorite run estimator because it ties closely to actual run scoring even in very high and very low run environments (i.e. college baseball or the MLB dead-ball era).
I wasn’t able to duplicate Matthew’s numbers exactly, so I’ll be up front about the formulas I used in case anyone wants to correct my work:
Download the regular Excel version here. The stats will update daily via Fangraphs, and the standings will do the same via Baseball Reference. I didn’t make a regular Google Docs version because the web queries wouldn’t work right.
What you’ll notice on the “summary” tab is that the teams are sorted from “luckiest” to “unluckiest.” Let me be quick to say that this is a rough estimate, so take any notions of “luck” with a grain of salt.
Based on their hitting and pitching components, though, we would say that the Rays have been the “luckiest” MLB team, and that they would finish with a record more like 100-62 (including the games they’ve already played) if they play at their current level the rest of the year. The Orioles, perhaps unsurprisingly, are playing four games below their current BaseRuns expectation.
Braves fans may notice that the pitching staff has looked a little better from a BaseRuns perspective (83 BsR allowed, 91 actual), but good pitching can only do so much, as has been evident over the last eight days.
If you have any questions about what I did or a suggestion for how to improve it, drop me a line in the comments.