The Scott Brosius weirdness continues...

Last year, I invented a formula for predicting the playoffs. The formula correctly picked the White Sox to win it all. Of course, since it worked so well, I decided to use the formula again in 2006.

My formula is based on one little-known fact: the team that committed fewer errors in the regular season wins the division series 2/3 of the time. I've looked at a whole bunch of stats, but I haven't found another stat with a better success rates (fielding percentage comes closest).

This correlation only holds for the division series, not for the whole playoffs. So I set up my formula to work like this: in the Division and World Series, pick the team with the fewest errors. In the LCS, reverse the trend, and pick the team with the most errors.

Quite simple, eh? You'd think, but this year, the formula completely falls apart. Watch:

Division Series (fewest errors wins)
Mets (104 errors) over Dodgers (115).
Padres (92) over Cardinals (98).
Yankees (104) over Tigers (106).
Twins (84) and Athletics (84): it's a tie!!!

Oh, no! We can't predict who is going to win the A's-Twins series!

No matter, we can continue anyway:

League Championship Series (most errors wins)
Mets (104) over Padres (92).
Yankees (104) over A's/Twins (84).

Which leads us to this:

World Series (fewest errors wins)
Mets (104) and Yankees (104): it's a tie again!!!

Yikes! We have no World Series winner, either! The formula fails!

Scott Brosius, what hast thou wrought?