I introduce a new statistic: Adjusted *OPS+. This statistic is a slight variation of Baseball-Reference's *OPS+. *OPS+ works out such that if a player has the league average *OPS, then his *OPS+ will equal 100. Anything higher than 100 represents above average performance compared to the league. However, *OPS makes the mistake of weighing Slugging Percentage and On-Base Percentage equally. According to Moneyball, a single point of OBP is approximately three times more important than a single point of SLG. Thus, my new Adjusted *OPS+ weights OBP three times as much as SLG. The asterisk in all of these stats represents the fact that they are adjusted for the park a player plays in and the lineup that surrounds him.
Below is a list of Sox players and potential off-season targets and their 2008 Adj. *OPS+ (again, above 100 means above the league average).
| Position | Player | Adj. *OPS+ |
| c | Saltalamacchia | 100 |
| c | Varitek | 77 |
| 1b | Teixeira | 146 |
| 2b | Pedroia | 120 |
| 3b | Youkilis | 134 |
| 3b | | 99 |
| ss | Cora | 101 |
| ss | Lowrie | 93 |
| ss | | 92 |
| lf | Ramirez | 157 |
| lf | Holliday | 138 |
| lf | Bay | 128 |
| cf | Crisp | 96 |
| cf | Ellsbury | 91 |
| rf | Drew | 138 |
| dh | Ortiz | 118 |
The biggest surprise here, albeit subject to small sample size, is that Alex Cora outperforms both Jed Lowrie and Julio Lugo. Also interesting is that Coco Crisp over the course of the entire season outperforms Ellsbury. Manny Ramirez is simply on another planet with his Adj. *OPS+, but Teixeira and J.D. Drew are both fairly impressive themselves.
From these numbers, it appears that the Red Sox were correct in not trading Youkilis for Billy Beane and that they were also right to sign J.D. Drew. Further, while Crisp's trade value is as high as ever, it would be wrong to blindly assume that Ellsbury is the better hitter. If Saltalamacchia performed at the league average as a 23-year-old in an off-year, I can't wait to see what good years in his prime will look like.
Finally, it should come as no surprise that Varitek was by far the worst regular player on the team in terms of hitting.
The primary shortcomings of this statistic is that it does not measure fielding performance or baserunning (other than the ability to stretch a single into a double, etc.).
The formula, for those interested, is
Adj. *OPS+ = 100 * (1.5 * OBP/*lgOBP + 0.5 * SLG/*lgSLG - 1)
where *lgOBP is the park- and lineup-adjusted league average OBP, and *lgSLG is the same for SLG.
At some point in the future, I would like to compile a similar stat that compares to the league average for a particular position instead of for all hitters.
2 comments:
Well done, sir.
As far as the exclusion of baserunning and fielding is concerned... perhaps you could take *OPS+ and convert it into runs created over league average. I don't really have any idea as to how one might accomplish this, however. But if you were able to do it, you might then be able to add the batting runs above average to existing metrics in fielding runs above average (pick your favorite) and baserunning runs above average.
... I'm actually kind of surprised that no one has (that I know of) done this yet.
Post a Comment