from Amazon
Thursday, March 08, 2007
The Four Horsemen
Studes follows the Voros approach in describing some players. It is in fact Voros’ approach that allowed me to create aging charts. (See Legend at the bottom)
As you can see, each rate describes something specific.
Now, there’s no reason that you must look at things this way. It assumes a certain independence that perhaps is not warranted. You could for example, look at things in other ways. Rather than removing HBP from the denominator first, then the BB, then the K, then the HR, you can remove all four right away.
So…
You have your PA split by HBP, BB, K, HR in one group (the unfielded balls), then fielded balls in another.
From the unfielded balls, you can create a rate of HR per unfielded ball. Then, remove the HR, and you can get a rate of K per uncontaced ball. Then walks per unswung ball.
For the fielded balls, you would do H+RBOE per fielded balls. Then 2B+3B per safe-fielded balls, and 3B per extrabase-safe-fielded balls.
Or, you can first break up the fielded balls into GB or air ball. In the air ball, break them up into pop, liner, fly. For each of those, follow the pattern in the previous paragraphs for hits.
Or, you can remove K, BB, HBP first, and then remove HBP from that group, as HBP per non-contacted ball. Then do BB per K + BB. Perhaps this approach makes more sense for pitchers, and the other approach makes more sense for hitters.
The key, as Voros showed us when he introduced DIPS, is to keep things as independent as possible. But, just because we do this, doesn’t mean it’s right. But, it sure seems right. And, at the very least, it create a profile of a player, one that can be used for comparison purposes.
And it was this kind of approach that allowed me to identify comparable players in this old article.
Hey Tango, I like the idea of similarity “scores” based on the “four horsemen.” Mind if I use that for an article someday?