Offensive Winning Percentages

This is an email I sent to Rob Neyer on August 15, 2000, reformatted for publication.
ESPN.com calculates “offensive winning percentage,” which I assume is the
same basic formula used in your book. Here are the Mets/Braves starting
lineups through Sunday, with ranks among guys at the position in the NL
qualifying for the batting title:

Pos Mets OW% Rank Braves OW% Rank
C Piazza .835 1 of 4 Lopez .528 4 of 4
1B Zeile .617 9 of 17 Galarraga .640 8 of 17
2B Alfonzo .744 2 of 15 Lockhart .346 DNQ
SS Bordick* .560 3 of 11 Furcal .623 DNQ
3B Ventura .536 6 of 9 C. Jones .723 1 of 9
RF Bell .611 9 of 16 Jordan .521 12 of 16
CF Payton .522 11 of 19 A. Jones .673 7 of 19
LF Agbayani .692 DNQ Surhoff* .565 11 of 18
P Hampton .391 DNQ Maddux .089 DNQ

*-Based on full year stats
I seem to recall from the book that very few of the all-time great teams you listed had a regular at every position with an OW% over .500, let alone .520, but if Veras were healthy, both of these teams would meet that standard at this stage of the season.
I had to put in the pitchers to show that Hampton is having a better year with the bat than the Braves’ starting 2B. The Braves fail to stack up to the Mets primarily due to the absence of Quilvio Veras and the fact that Chipper would be the Mets’ third-best hitter. The Mets have a better ERA in the starting rotation and have scored more runs than Atlanta; the Braves’ only statistical advantage is in the bullpen (!).
Also: Benny Agbayani’s OPS, for 1999-2000, is now .891 (comparing favorably to the .876 figure posted by Bubba Trammell, 1998-2000). Agbayani was not an impressive minor league hitter. Is this a real improvement, or are we still waiting for a larger sample size for Agbayani to return to earth?