Monday, June 12, 2006

ERV Renovations -- Update

As you know, last month I shut down the ERV Boxscores to redesign the process that creates them. The work has been going pretty well, and this weekend I finally constructed the guts of the new database that will hold the information. It can now calcuate the basic RV and WV for each event. Right now it ascribes all of the value to the batter and pitcher, even if the play has errors or other aspects that should be given to other players involved. I have to code the parts that account for errors, fielding and baserunning.

But the real advantage of this new system is that I have data for every team, not just the Nats. So we'll be able to compare RV and WV with other players in the league. For example, here are the 2005 NL Batting WV Leaders:


NameTotal RVTotal WV
Chipper Jones38.31454.142
Jason Bay38.36846.264
Derek Lee50.64544.553
Bobby Abreu54.40342.941
Carlos Delgado40.52741.785
Adam Dunn37.80537.268
Lyle Overbay27.88637.029
Chase Utley38.65936.416
Ken Griffey33.14834.526
Lance Berkman29.00033.585


Where's Albert? Pujols was 14th, with 44.76 RV and 30.42 WV. Nick Johnson was 11th, with 24.89 RV and 33.00 WV.

Here are the best NL pitchers:

NameTotal RVTotal WV
Roger Clemens-50.246-61.176
Andy Pettitte-53.525-56.694
Dontrelle Willis-44.242-56.627
Carlos Zambrano-31.614-50.201
Chris Caprenter-39.947-46.454
Tim Hudson-32.968-45.341
Todd Jones-21.705-42.636
John Patterson-30.653-39.580
Derrick Turnbow-20.319-39.184
Jake Peavy-24.321-38.769

What's interesting to me is that most of these guys are starters. I had thought that because closers pitch in high leverage (i.e. high Win Value) situations, the system would be biased in their favor, and they'd cluster near the top. But apparently that's not necesarily the case.

Keep in mind that these numbers are rough and I haven't checked the data thoroughly. But I do think it tells me that when I'm finished that the data will present some interesting things to think about.

2 Comments:

At 12:47 PM, Anonymous Simon Oliver Lockwood said...

I'm not surprised that there're so few closers on your list. They pitch 1/3 the number of innings as rotation starters. There aren't that many situations where the Win Value is 3 times the average. The firemen of the 60s and 70s might get up there because they pitched a) more innings (up to twice as many as today's closers), and b) in more critical situations (coming in more often in tie games and with runners on).

 
At 2:33 PM, Blogger DM said...

You're probably right. I was biased by last year, when for most of the year Chad Cordero was leading the Nats in WV. I had wondered if the system was biased in favor of closers. But, for the reasons you state, it isn't.

 

Post a Comment

<< Home