How Much Is Fielding Weighted in WAR?
Occasionally (okay, rather frequently), I’ll see people debate the accuracies between the WAR displayed on FanGraphs and Rally’s WAR on Baseball-Reference.
Joe Posnanski speculated on the differences in a recent article about Josh Hamilton’s MVP chances:
*I could be reading this wrong, but Fangraphs seems to put more emphasis on defense. For instance, Carl Crawford’s WAR at Baseball Reference is 3.7 — his defense is worth eight runs above average. But Fangraphs credits him for 22 runs above average, which thrusts his WAR up to 5.6 and into the No. 4 spot in baseball.
I’ve seen similar sentiments echoed throughout the blogosphere and on Twitter.
In reality, on a per-player basis in 2009, UZR distributed 441 fewer runs than TZ did, excluding pitchers and catchers. And there is not a year that UZR is available where its absolute value has been higher than TZ.
In 2009, the maximum spread of UZR was +31 to -37 and TZ showed a similar spread of +31 to -34. Here’s a graph of the full spread. The blue overlap shows the points at which TZ starts showing a greater spread.
This might not be a perfect comparison in how much defense actually contributes to WAR and a better one might be how much as a whole does fielding contribute to total runs. In 2009, fielding made up about 14.2% of all positive and negative runs according to FanGraphs WAR, while Rally’s WAR made up about 15.5% of all positive and negative runs.
All in all, they are similar in how fielding is weighted as a whole. The biggest difference between the two is how each individual player’s fielding is evaluated.
David Appelman is the creator of FanGraphs.
Just a thought, have you guys ever thought of regressing the fielding numbers, based on their inherent volatility?
I like that idea. I also like the idea of expressing defense in a range (something like -7 to +7). I just think as popular as this site is, there needs to be something that shows that the UZR or TZ is not as reliable in a small sample as offense is. I think much of the disagreements start over people saying so and so was worth 9 runs using UZR and -6 runs using TZ. People look at that and say they can’t be right so all defensive metrics should be ignored. They shouldn’t ignored, but I think the unreliability of the metrics should be acknowledged in their measurements. I also think they have to be regressed. UZR/150 is the same as looking at a guy who had 2 home runs after the first game and saying he’s on pace to hit over 300 home runs. Get rid of UZR/150 and add an metric error column.