clock menu more-arrow no yes

Filed under:

Revised PECOTA Standings Up

New, 185 comments

I was going to do a post about the issues I felt there are in regards to the BP PECOTA standings that went up earlier this week, but that appears to be moot, as new projected standings are now up.

This version has Oakland at 86 wins, Texas at 85 wins, Seattle at 81-81, and Anaheim winning 77 games.

The "runs allowed" for the Rangers dropped from 842 to 795...at some point, I'll see if Google Cache or someplace has the old page with the individual Ranger projections, to compare and contrast.

But my initial reaction at the time was that the 842 runs allowed probably was a result of the algorithm not properly recognizing the Rangers' strong defense...as I've mentioned before, BP's biggest statistical weakness right now is probably their flawed fielding stats.

Eric Seidman has a post up mentioning a couple of the problems that were involved. 

I'll also take this moment to mention something...there were a few commenters in the original PECOTA projections post that talked about how there's a "bias" against the Rangers, or some projections were just done for shock value, or whatever.

This doesn't really make sense because this is all algorithm-driven.  There may be "biases" built into the algorithm, in the sense that the algorithm uses flawed assumptions on playing time or the projections are not what you or I would project, but the idea that the system spits out three 90-win teams in the A.L. East because there's an East Coast bias, for example, is just wrong.