Monday, July 2, 2007

New Metric

So i've been tinkering with team quality metrics and wanted to share the results of one attempt with you. This is an old method but it says a lot about relative quality. What i'm doing is making a cumulative score for team rank in the NL of important stats. Dumb i know, and stats heads like me cringe when people try and compare stats like this but amazingly enough this method does give a good and simple metric for team quality.

To set up the data set i've taken all the major team stats from ESPN as of noon today and assigned a point value for rank with 1pt being first in the league and 16pt being last. I've ranked teams based on R, H, HR, TB, BA, and OPS then summed their scores. Here are the results:

1 Philadelphia 17
2 Milwaukee 22
3 Chicago Cubs 25
4 Colorado 30
5 Florida 30
6 Atlanta 32
7 NY Mets 37
8 Cincinnati 41
9 Houston 48
10 LA Dodgers 53
11 Arizona 70
12 St. Louis 72
13 San Francisco 77
14 Pittsburgh 80
15 San Diego 80
16 Washington 94

By this metric the braves have the 6th best offense in the NL. Now for the pitching. Same method of analysis using ERA, R, BB, SO, and BAA. The results:

1 San Diego 15
1 LA Dodgers 15
3 Chicago Cubs 24
3 Milwaukee 24
5 NY Mets 26
6 San Francisco 38
6 Arizona 38
8 Atlanta 40
9 Houston 52
10 Pittsburgh 54
10 St. Louis 54
12 Florida 59
12 Cincinnati 59
14 Colorado 60
14 Philadelphia 60
16 Washington 61

Here the Braves are 8th. Now we combine metrics to find:

1 Milwaukee 46
2 Chicago Cubs 49
3 NY Mets 63
4 LA Dodgers 68
5 Atlanta 72
6 Philadelphia 77
7 Florida 89
8 Colorado 90
9 San Diego 95
10 Cincinnati 100
11 Houston 100
12 Arizona 108
13 San Francisco 115
14 St. Louis 126
15 Pittsburgh 134
16 Washington 155

And amazingly enough the braves are 5th, or just out of the playoff race. Now, personally i feel that the Cubs and Brewers are over performing but this method of analysis shows that the Braves are going to be in a dogfight for the wild card with the Cubs... scary. Fortunately we have 1/2 a season to go. Let's hope those numbers improve.

No comments: