Projecting Felix In 2006
I’ve done a significant number of posts breaking down Felix Hernandez’s statistical performance through his first four starts. What I haven’t done, however, is project what I expect from King Felix going forward. Until now. I’ve been working on a pitcher-projection-system behind the scenes, and I’m finally to the point where I trust the data enough to publish it.
Now, keep in mind, pitchers are weird beasts. No projection system in the world would have forecasted Jamie Moyer’s late career development. Statistical projections are a tool that can be useful, but when combined with pitchers, we need to know that they are still a blunt tool, and especially with a kid who won’t turn 20 until next April, absolutely anything is within the realm of possibility. However, we can use past performance to give us a reasonable window of what to expect. That’s what I believe this projection for Felix provides; an array of reasonable performances for the King in 2006, any of which could occur.
As for the nuts and bolts of the system, I’m going to lean towards brevity. It’s nothing radically different than what others have done before me, I’m sure. Ron Shandler’s projections are based on most of the same assumptions, and I’m actually expecting his postseason projection for Felix to be similar to this one. So why create my own? Well, two basic reasons: I’m impatient and don’t like waiting for someone else to tell me what I want to know, and I don’t have to pay myself for this information. I’m not claiming this system is better than what Shandler or Prospectus or whoever else puts out there. I’m simply claiming as far as accuracy, it’s pretty darn close, and the fact that its available now for no cost makes it a winner in my book.
So, anyways, here’s the basic gist of the system. In any given plate appearance, there are five potential scenarios, one of which will occur each time: a walk/hit by pitch, a strikeout, a ground ball, a fly ball, and a line drive. Of these five outcomes, we know that a pitcher has significant control over the first four, and though the research is still new, it appears that pitchers can exert some control over the amount of line drives they allow as well, though it is less than their control over the other four scenarios. As most people who have read the blog for quite a while know, we believe the current evidence shows that, with some exceptions, most pitchers do not have a repeatable skill that allows them to control the outcome once a player makes contact with the ball. In other words, a pitcher can induce the opponent into hitting a groundball, but he can’t consistently compel the batter to hit a groundout.
So, what I decided to do was create a system that converts each batter faced into a percentage of likely outcomes, much like the last chart I used to break down Felix. We know Felix has significant control over his walk, strikeout, and groundball/flyball rates, so those are what we’re projecting, and the rest of the data populates itself when we put in the league average results based on hit types. Keep in mind, however, these numbers are context neutral, not dependant on the park he plays in or the teammates around him. Pitching in front of the M’s defense and in Safeco should only serve to make his raw numbers even more impressive.
Anyways, that’s a basic overview. For most people, what you really care about are the results, right? Well then, here you go.
Median Projection: The Middle Of the Road Safe Pick
213 innings pitched
128 singles allowed
38 doubles allowed
2 triples allowed
16 home runs allowed
GB/FB rate: 2.20
Opponents BA: .244
Opponents OBP: .290
Opponents SLG: .324
Component ERA: 2.86
Fielding Independent ERA: 2.99
2005 Performance Comparison: A.J. Burnett
Optimists Projection: The Best We Should Plan For
244 innings pitched (and yes, I realize the team will never let him near that number)
143 singles allowed
42 doubles allowed
2 triples allowed
14 home runs allowed
GB/FB rate: 3.00
Opponents BA: .237
Opponents OBP: .268
Opponents SLG: .303
Component ERA: 2.38
Fielding Independent ERA: 2.21 (!)
2005 Performance Comparison: No one was this good in 2005. Not even Clemens.
Pessimists Projection: What To Expect If He Struggles
178 innings pitched
116 singles allowed
35 doubles allowed
2 triples allowed
11 home runs allowed
GB/FB Rate: 1.67
Opponents BA: .257
Opponents OBP: .331
Opponents SLG: .328
Component ERA: 3.39
Fielding Independent ERA: 3.56
2005 Performance Comparison: Carlos Zambrano
Based on Felix’s professional career to date, the pessimistic projection for him next year would make him something like the 5th or 6th best starting pitcher in the American League. The optimistic projection grades him out as better than what Clemens is doing in Houston this year (which, to me, is hard to believe), and would be one of the most remarkable pitching seasons we’ve seen in the history of the game.
I feel the best about the median projection, and that’s along the lines of what I’m expecting next year. And, realistically, if he puts up his median projection next year, he’s going to finish in the top 3 in Cy Young voting. People talk about the need to acquire a #1 starter in the offseason. I disagree. We have one.