How many over the typical replacement player does Brett Gardner provide the Yankees? That’s a tough question to answer. One of the best easily accessible methods to determine this is fWar, over at Fangraphs. Fangraphs will tell you that Gardner was worth 5.1 wins over replacement in 2011, and 6.1 in 2010. A lot of people think that this doesn’t pass the smell test, since it would make Brett Gardner right about as valuable as Robinson Cano. Intuitively, that doesn’t make a lot of sense to us.
The problem lies in measuring defense. That’s been the hardest thing for people to measure for a long time. Scoring runs is easy to measure, because the batters are solely responsible over the long run. Preventing runs, on the other hand, comes from a combination of the pitcher and 8 defenders in the field. How much credit do the defenders get, and how much credit do the pitchers get?
Fangraphs uses Ultimate Zone Rating (UZR), which has its problems. UZR, in the basic explanation, takes the number of plays that could possibly be made by a defender and divides it by the number of plays that they successfully make, then converts that into runs and compares it to other players at the position. Brett Gardner has consistently posted a very high UZR score of 31 UZR / 150 games, or between 3 and 3 and a half wins per season.
That’s probably not news to many people reading this site. UZR has its problems, but I don’t think that Brett Gardner suffers from those problems. UZR is prone to sample size issues. Because there are 9 fielders (including the pitcher) in the field, even a 162 game season isn’t a sufficient enough sample size to overcome random statistical noise. That’s why we shouldn’t trust Alex Rodriguez’s 4.2 fWar this season, despite playing so few games and not hitting as well as he usually does. His defense improved, but there’s not enough of a sample size in his 2011 UZR to make the judgement that he would be worth +20 runs on defense over 150 games. That’s too big of a jump to believe.
However, I don’t think that UZR’s normal problems apply to Brett Gardner. Gardner has been a consistently excellent defender for three years now, and our eyes confirm this. There’s no reason to doubt that he’s one of the best, if not the best, fielder in left field.
What about that left field thing? I’ve heard a lot of people say that Gardner’s UZR is an illusion, because he plays left field, and is thus compared to other left fielders. He makes a ton of plays, and that raises his average far above mediocre competition. My problem with this critique is that it doesn’t make any sense. Sure, Brett Gardner should probably be a center fielder. His skills would still be strong (possibly Gold Glove caliber) in that position, but he’d likely be in the +10 to +15 range, not +30. While that may make some sense, its also worth remembering that he’d be compared to different replacement players. Simply manning left field gives Gardner a -5.8 run positional adjustment. Curtis Granderson’s time in center field earned him a 2.2 run adjustment. The LF-CF thing evens out for the most part. If anything, his value would actually increase in center, given the marginal increase in opportunities.
Here’s the real leg to stand on if you think Brett Gardner’s value is overstated: left fielders haven’t hit much lately. Brett Gardner, of course, is one of those light-hitting left fielders, but he’s was actually better than the average LF in 2011, who hit .251/.311/.393. Left field was the second-lightest hitting position after catcher. That’s not going to last for long. Gardner’s value will take a hit over time as the average left fielder stops being a hole in the lineup and starts providing offense again. But that doesn’t mean that Gardner’s real value is low to a team. We can only really compare these things against the league’s competition. And even given that kind of regression to the mean, Gardner would still be a very valuable player.
If you enjoyed this article, please consider sharing it!
One Response to On Brett Gardner and Those Enigmatic fWar numbers.
Leave a Reply
TIQ IQ
Well, no one else is commenting on this so I’ll take a hack at it.
Just to be clear, I like Brett Gardner and think he is a valuable member of the team. I think he should be an every-day player (who is strategically rested against some of the tougher lefties) and batting at or near the top of the order.
That said, I have a lot of difficulty accepting, at face value, the numerical valuations that I’ve seen attached to him, frequently accompanied by comments like “all-star” and “elite player”. I feel particularly that way about the Fangraphs numbers, which, IIRC had pegged Gardner as having >90% of the value of Albert Pujols (yes, I know Pujols had a down year). While I don’t pretend to understand the defensive metrics at the level of you guys at TYA what I have read about them does not give me a lot of confidence in a player valuation which owes the lion share of it’s foundation to them.
I know Brett Gardner is an uncommonly good outfielder just by watching him but even with the positional discount Fangraphs gives to Gardner I have a lot of trouble seeing how he could add that much value over the typical LF in terms of defense. After all, a batter will get his 3-5 chances to hit per game. A fielder can only have value if the ball is hit to him within his range which may or may not happen in a particular game and even a superior fielder can’t add value over a generic replacement unless the plays he’s presented with are in the upper range of difficulty. Given that good fielders are more common than good hitters and that LF defense is not as difficult as say SS or C I find it hard to understand how the defensive contribution of even a LF as good as Gardner can be weighted as high as Fangraphs weights them in its player valuations.