Injury Takes Holliday
It looks like it’s not an April Fool’s joke – Matt Holliday is having his appendix out Friday afternoon. A tough spring has just gotten worse for the Cardinals. Approximately how many wins has the team lost in the last month? Let’s count it up.
First they lost as many as five wins when Adam Wainwright went down. Even if Kyle McClellan can replicate his league-averageish FIP work from the rotation (which isn’t likely, considering the penalty usually associated with such a move), the team lost at least about three wins when their young ace grabbed his elbow in pain.
This Holliday situation won’t be as dire. For one, he’s not out all season. The team is saying two weeks, or as few as “six or seven days,” but the typical recovery time can go as many as four-to-six weeks. Let’s be safe and call it three weeks. In those busy three weeks, the Cardinals play 20 games, or about 12% of their season. Pardon the wonky math, but Holliday has averaged 671 plate appearances and 6.33 wins per season. Steal 12% of those plate appearances, and the excellent outfielder might still manage 591 plate appearances. If he had an ‘average’ season despite the injury, he might put up 5.57 wins. That’s not even a full win of a loss.
In his place, Jon Jay and Allen Craig will play a little more over the next three weeks. Jay doesn’t walk much (7.4%), strikes out just a hair below average (17.4%), has below-average pop (.122 ISO), but might be a scratch glove on the corner outfield. His own fans projected him to have about an average-ish wOBA (.328), so if he combined that with neutral defense, he could put up an average-ish WAR. Let’s call him a 2-win player over 600 PAs, or a .27-Win player over 80 PAs. Subtract that from the Holliday loss and you’re looking at losing a half a win.
We’ve shown this sort of thing before. A star player goes down for a while, and we tend to over-react. Zooming out, it may only cost the team a half a win or so. Taken in a vacuum, it wouldn’t be a huge deal, especially if the recovery time trends towards the timetable suggested by the team.
But look at this in tandem with the loss of Wainwright, and it does change things a little. If we were more likely to say the team was only going to lose about three wins (instead of five) with the drop from Wainwright to McClellan, we’re now a little more likely to push that number in the other direction. In a division that is likely to be decided by fewer than four games in the final standings, this April Fool’s day surprise was not a pleasant one for the Cardinals.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
Using WAR in this way does not seem very useful. WAR is an abstract calculation – it doesn’t reflect reality in small sample sizes.
I agree and disagree. Surely, we have no idea ‘which’ three weeks we would have missed out of Holliday’s season – might have been a sweet stretch or a poor stretch. But, also, the sample sizes behind the WARs used are fairly big (four seasons), so using it to paint broad strokes (what might the team miss) seems appropriate to me.
Perhaps he’s also suggesting that you’re taking the “wins” part of WAR too literally, given the uncertainty of the true “replacement level.”