Crooked Numbers: No Team without M-E

Let’s just get this out of the way now: Derrek Lee is not going to win the MVP award. Andruw Jones set a well-publicized franchise record by hitting his 50th home run, his sterling defensive reputation is still intact, and he’s perceived as leading a dynasty reborn back to its rightful place atop the division. If that’s not good enough for the writers, Albert Pujols is once again the best player on the best team in the league and he’ll likely garner the sympathy vote from those writers who know that in a Barry Bonds-free world, Pujols would already be a three-time MVP.

Lee’s candidacy has several problems, but the three main obstacles are 1) his failure to win the triple crown–he’s leading AVG by a handful of points over Pujols, is second in HR to Jones, and is seventh in RBI; 2) his truly outrageous performance came in the season’s early months unlike Jones’ late-season surge; and 3) his team is not in contention.

It’s this third point that I want to focus on today. The nadir of the Cubs’ season came on August 10 when they lost their eighth game in a row and dropped six games under .500 at 54-60. Since then, they’ve actually been playing well, going 18-13 over the last month, sitting at 72-74 yesterday morning. Despite the better play of late, the last time the Cubs’ odds of making the playoffs were higher than 10% was July 31. The Cubs are a distinctly mediocre team, but just imagine how bad they would be without Lee.

Answering that question is a tricky one. It would be convenient to simply see what percentage of the Cubs’ WARP1 Lee accounts for but, according to WARP mastermind Clay Davenport, simply adding player WARP1 scores to create team WARP1 scores is similar to adding batting averages to generate a team batting average. Instead, team WARP1 scores are computed from the component runs above replacement, be they batting (BRAR), pitching (PRAR), or fielding (FRAR).

Thus, we could drop back to those individual runs above replacement and see how much Lee contributes compared to the rest of his team. This methodology brings to light another issue: you cannot take a percentage of a negative number. Of course, the last team to post a negative WARP3 was the famous 1899 Cleveland Spiders. (It should be noted that WARP1 is the applicable metric when comparing players playing in the same season while WARP2 and WARP3 are adjusted to compare players across multiple seasons, the difference between the two being the “length of season” adjustment in WARP3. Thus, comparing the 1899 Spiders to the 2005 Cubs, WARP3 is the appropriate metric, but to compare Lee to his teammates this year, WARP1 is best.) Regardless, we have the issue of players with negative WARP scores–what percentage of the team’s value did they contribute? And if the team without the player in question was below replacement level, the player would contribute more than 100% of the team’s value over replacement level.

Instead, we can try to view Lee’s performance relative to his teammates in a slightly different way. To determine the Cubs’ expected WARP1 without Lee, we instead need to recalculate their WARP1 score minus Lee’s BRAR and FRAR. We’ll skip the math for now, but removing Lee from the Cubs and replacing him with a replacement level player, the Cubs drop from 54.5 to 43.6 WARP1, a loss of 20.0% of their value. By contrast, Pujols nets the Cardinals 11.8% of their total value, and Jones 13.6% of the Braves’. Not only is Lee having the best year of the three according to WARP1, but he’s far and away the best player on his team.

When looking over recent performances that have garnered hardware, Lee is far exceeding historical precedent. In 2003, Alex Rodriguez garnered the MVP for last-place Texas while accumulating 16.6% of the team’s total WARP; Andre Dawson‘s MVP campaign for the basement dwelling Cubs in 1987 totaled a mere 10.2% of his team’s WARP.

However, two hitters just last season totaled a higher percentage than Lee and they were both in the NL West: Bonds and Todd Helton. In fact, quite a few players have turned the trick since 1990:

NAME            YEAR   TEAM    WARP1  Team WARP1    W/O_Player    PERC
-------------   ----   -----   -----  ----------    ----------    -----
Todd Helton     2004   COL-N   12.3      48.1          36.9       23.2%
Barry Bonds     2004   SF_-N   14.9      66.5          52.1       21.6%
Todd Helton     2003   COL-N   12.3      50.6          39.3       22.4%
Dmitri Young    2003   DET-A    6.5      25.9          20.1       22.2%
Brian Giles     2002   PIT-N   11.2      47.4          36.0       24.0%
Jim Thome       2002   CLE-A    9.5      42.8          33.4       22.0%
Brian Giles     2001   PIT-N    8.5      33.8          26.1       22.6%
Barry Bonds     1996   SF_-N   11.7      46.2          35.2       23.8%
Barry Bonds     1995   SF_-N    9.8      39.3          30.8       21.6%
Jeff Bagwell    1994   HOU-N   11.5      49.1          39.2       20.2%
Cal Ripken      1991   BAL-A   12.6      49.1          39.2       20.2%

Take Bonds off that list and we’re left with the list we’d expect: rare performances on very bad teams. This isn’t to say that Lee deserves the MVP simply because he’s contributing more to his team than any player this year. Whether or not a percentage of total team wins is your definition of “value” is up to you. But in relation to his teammates, Lee is the most valuable player this season. But as we all know, to win the MVP, a player’s performance must differentiate himself from the league, not his team, for if the latter fails to keep pace, his season is deemed to lack the prerequisite meaning of an MVP season. Lee’s teammates have failed to keep pace with his impressive performance and as a result, the voters will send the trophy somewhere else this winter.