Denard Span and Cherry Picking
Let's start off with a simple comparison between two unknown hitters, Player A and Player B. Below are a few stats for each player and their respective rankings in Major League Baseball among qualified batters. Player A
9.8% K% (8th lowest)
28 doubles (4th)
52 runs (t-21st)
6.7% BB% (106th)
.313 wOBA (111th)
19 RBIs (t-156th)
It should come as no surprise that Player A is Denard Span and considering the title of this article it also probably shouldn't be a surprise that Player B is also Denard Span. This is what cherry picking looks like. By one set of stats Denard Span looks like the perfect leadoff hitter, by another he looks like one of the worst hitters in baseball. The truth, as always, lies somewhere in the middle.
The first thing that should stand out to you when look at those numbers again is how low both Span's strikeout rate and walk rate are. It should therefore come as no surprise that Span has a contact rate of 92.6% this season according to PITCHf/x, the highest contact rate among qualified hitters in Major League Baseball along with Victor Martinez. If Span is swinging, then he's likely making contact.
At it's face that isn't a bad thing and Span's contact rate is right in line with the rates he put up with the Minnesota Twins. However, when Span left Minnesota for Washington he had a career batting line of .284/.357/.389, a 9.5% walk rate and a 12% strikeout rate. In his two years with the Nationals, Span is batting .275/.323/.385, a 6.5% walk rate and a 11% strikeout rate. For a leadoff hitter the drop from a well above-average .357 on-base percentage to a near league average .323 OBP is substantial. So what's behind this insidious drop?
Mostly it's Span's walk rate and if you read my post earlier this year on plate discipline and strikeout and walk rates then you probably have a good idea where the problem is. While Span is making the same contact he always has, his swing rate at pitches outside the strike zone has jumped from an average of 20.5% with the Twins to 27.5% in 2013 and 24.2% in 2014. The immediate issue, as detailed in the aforementioned post is the drop in walk rate. The larger effect though isn't quite as obvious.
Since Span is swinging more but is making contact at the same rate as before that means he's putting the ball in play more. This might sound like a good thing, but the problem is that it isn't a one-to-one transaction. Denard Span's BABIP stays relatively the same regardless of how many balls he puts into play. So while he saw his walk rate drop from 8.3% in 2012 to 6.3% in 2013, his BABIP (.315 to .313) and hit rate (25.7% to 25.7%) stayed about the same. And voila, a once potent on-base threat is reduced to merely average. And unlike some hitters, like Martinez, who can make up the difference by hitting for more power, Span doesn't have that luxury. Even with the 28 doubles Span's isolated slugging (.125) and slugging percentage (.393) are well below and at league average respectively.
So now we have a hitter who is entirely dependent on getting on-base who isn't getting on-base, let's ditch him right? Well, not so fast. As I said before, despite Span's problems at the plate he's still just about league average with a 98 wRC+. And hitting isn't the only part of offense, there's also baserunning where Span excels. His 3.8 baserunning runs above average are tied for 12th in the Majors among qualified hitters allowing Span to contribute 3.0 offensive runs above average. Now that isn't a huge number, but it certainly isn't killer for a guy who has been worth over 9.0 defensive runs above average in each of the past four seasons at a premium defensive position. So Span certainly isn't hurting the Nats, and his 1.4 fWAR puts him squarely on pace to be a solid starter this season. He probably just shouldn't be batting leadoff, but that's a discussion for another day.