I thought it would be an interesting exercise to calculate WAR for each Marlins player on my own and see how the results turned out. It was mostly a practice for me for future attempts at determining stuff like MVP awards for the BBA. On a whim, I decided to start with pitcher WAR, since I already had some results and was intrigued to find out how the Marlins stacked up.
Now, pitcher WAR is interesting because there is a lot of variation in the way different sources calculate it. There are two primary sources for WAR, FanGraphs and Rally’s WAR database. The two sites use very similar methods of calculating position player WAR, essentially differing slightly only in inputs. The method to determine pitcher WAR is similar as well, but the inputs are significantly different. As you may already know, FanGraphs uses Fielding Independent Pitching (FIP), a statistic developed by Tom Tango that considers the runs pitchers allow from purely fielding-independent components such as strikeouts, walks, and home runs, then puts it on the scale of ERA. Rally’s WAR statistics, on the other hand, more directly consider runs on a team context. Rally takes actual runs allowed by a pitcher and uses his TotalZone defensive metric, prorated for the pitcher’s balls in play, to account for defensive contributions, removing the prorated defensive runs from the equation, then converting this into wins.
Both versions definitely have their merits, and in most cases they probably would not diverge much from each other. However, in the particular case of Ricky Nolasco and his 2009 season, both versions would likely yield large discrepancies. As we all know, Ricky hit a supposed “rough patch” early in the season, culminating in a 9+ ERA and a demotion to Triple-A. When Ricky returned to the big leagues, he once again looked like the 2008 Ricky Nolasco, except that FIP and its defense-independent brethren never dropped Ricky from “darn good pitcher,” even as he struggled early. By the end of the season, Nolasco recoded 4.2 WAR for the year, with a FIP of 3.35. FG’s tRA measurement also seems to be OK with him, posting a tRA of 4.11, some 0.4 runs above average per nine innings.
I did both calculations using park factors provided by Patriot. For defense-independent statistics WAR, I averaged tRA from StatCorner and FanGraphs, then averaged that value with FIP/0.92 and stuck into Pythagenpat. If you checked out my MVP article, you saw a list of WAR for pitchers calculated using that; that list contains all pitchers with more than 4.0 WAR. Using that method of evaluation, I had Ricky at 3.8 WAR for the season, a very good total. I then calculated WAR using Rally’s method, using team bUZR from FanGraphs as my defensive metric. Using that total, I got Ricky totaling 0.8 WAR on the season.
That’s a difference of 3 WAR between the two methods. This is not surprising given what we know of Ricky’s problems; early in the year he couldn’t strand a runner to save his life. His BABIP against was astronomically high as well, and as we all know pitcher BABIP tends to regress fairly closely to .300 over time, since pitchers don’t have a whole lot of control over it. Still, those runs did still score, and removing defense is supposed to give you the defense-independent runs allowed by Nolasco on the season.
So I pose this question to you, Maniac readers: Which WAR value or win total is closest to correct for Nolasco’s production on the year? We know that FIP/tRA/tRA*/xFIP are going to be closer to his “true-talent” level than his runs allowed this season, but when measuring Ricky’s pure production on the year, which method pegged it better? Is timing important when we evaluate a pitcher’s production, since he has more control over his environment, or should we remain context-neutral and still use components and either linear weights or, to be perhaps more accurate, BaseRuns?
So once again, I want to hear the Maniac readers’ opinions here: Was Ricky Nolasco a four-win pitcher or a one-win pitcher this season? Have at it folks.