Monday, March 28, 2011

Oh snap.  I think Nate Silver's methodology went 0-for-4 in predicting this past weekend's basketball games.  The new projected winner is Kentucky.    

Incidentally, I found his original bracket:

Also incidentally, Nate Silver wrote a fresh New York Times article this morning, but unfortunately I couldn't get past the first sentence without an intense feeling of rage overwhelming my core.  That sentence reads:

Somehow, my N.C.A.A. tournament bracket still ranks in the 76th percentile nationwide, a result which it owes to having performed very strongly in the first couple of rounds.

Oddly enough, Nate didn't link to his original New York Times bracket, which shows he's in the 33rd percentile.  Nor did he link to an ESPN bracket (, which shows, very clearly, any bracket's percentile--and which is also, I think, the largest bracket compiler in the country.  Instead, he linked to his "Yahoo Fantasy Sports" bracket, which conveniently doesn't list a bracket's current percentile, nor does it tell you how many brackets the system is tracking.  (The New York Times contest tracks about 35,000 brackets, while ESPN tracks about 6 million brackets.  No data could be found for Yahoo Fantasy Sports, because their navigation system is fucking terrible, but my guess is somewhere around 100,000 brackets.)

It took me a while to figure out why Nate linked to his "Yahoo Fantasy Sports" bracket, rather than his New York Times bracket.  After all, the Times is his employer, and both his blog and methodology article are located on the Times server.  It turns out his New York Times bracket has predicted 31-of-60 winners correctly, while his Yahoo bracket has predicted 38-of-60 winners correctly.  So clearly Nate Silver filled out multiple brackets and only linked to his most successful one.

Well, no shit.  If I had several different bingo cards, odds are one card would perform better than the others.  When you fill out multiple brackets, odds are one of them will not look pitiful.  Maybe Nate Silver will have some complicated reason for why his personal New York Times bracket isn't official, but his "Yahoo Fantasy Sports" bracket is.  But, truth be told, I don't feel like giving this C-3PO knockoff the benefit of the doubt.

Nate should try to have some integrity, like President Obama, who filled out his one NCAA bracket on national television.  As it happens, Obama, like Nate, didn't get any of his Final Four participants correct.  But Obama's bracket is still in the 94th percentile over at ESPN by dint of his having done well in the earlier rounds.  And that bracket is linked to on the front page for everyone to see.

Tomorrow I'll write more about how Nate Silver pissed me off in the first place.

No comments:

Post a Comment