Over the past several weeks, the discussion concerning the AL Cy Young race has turned into a discussion concerning Felix Hernandez and CC Sabathia, and the discussion concerning Felix Hernandez and CC Sabathia has turned into a discussion concerning the value and importance of pitcher wins. It seems like there's a new article or three every single day, and each of them end up reaching one of two conclusions.
I've tried to stay out of it, both because there's still time left in the season, and because I'm personally unsure of what the Cy Young represents. However, as long as people are still talking about wins, I thought I'd throw in a little nugget of my own.
As the season winds down, much has been made, and much will continue to be made, of the difference in wins between Felix and Sabathia. Right now, that difference stands at eight, as Sabathia reached 20 and Felix has 12. Now, I don't need to go to great detail to tell any of you about the problems with wins. They depend on luck, they depend on offensive performance, and they depend on things that happen after you're already out of the game.
So as far as I'm concerned, wins should pretty much be ignored entirely. But as long as some people are unwilling to do that, I offer something similar to wins that adjusts for that last point - the one about stuff that happens after the starting pitcher is out. I think everyone can agree that no pitcher should benefit or be penalized if his offense later stages a rally, or if his bullpen blows a lead. Those are completely and utterly independent events.
To adjust for that, we go to Win Expectancy. For example: on May 13th, in a game against the Orioles, Felix allowed one run over seven innings, and departed with a 5-1 lead. Brandon League then blew it in spectacular fashion, and the M's fell 6-5. Felix didn't get a win. However, the ' win expectancy at the time of his departure was 95.8%, and that's way more important.
So that's what I'm going to focus on. I went through all 32 of both Felix and Sabathia's starts and noted the win expectancy after they faced their final batter of the game. Following are the averages:
We see, on average, a 16.2 percentage point advantage for Sabathia. Over the span of 32 starts, which each of them have made to date, that comes out to a difference of 5.04 'deserved' wins. Which is a good deal lower than the difference we currently see of eight 'actual' wins.
Which is interesting. And then you can start talking about the breathtaking difference in run support.
This isn't, of course, how I'd go about thinking through my hypothetical Cy Young vote. And I know win expectancy is based on the average team, and doesn't adjust to take into account that you're dealing with the Mariners or the. But this is one of those things that anyone who takes wins seriously should consider. The eight-win gap we see right now isn't really reflective of anything.