Is Cliff Lee For Real?
Published by Nick April 30th, 2008 in Pitch F/X, Sabermetrics
One of the most surprising things about the start to the 2008 season has been Indians’ LHP Cliff Lee’s utter dominance of opposing hitters in his first four starts. Coming off a 2007 in which he posted an ERA of 6.38 in 97.1 innings, Lee has a 0.28 ERA in 31.2 innings thus far this season. Naturally, the question on everyone’s mind is a more or less profane version of “what the hell is going on?” Has Cliff Lee turned a corner at age 29 and returned to his 2005 form (if not better) or is this just a run of good luck? In order to answer that question, let’s break down Lee’s numbers in 2007 compared to 2008:
2007: 97.1 IP, 66 K, 36 BB, 16 HR, 38% GB, .309 BABIP
2008: 31.2 IP, 29 K, 2 BB, 0 HR, 46% GB, .153 BABIP
To me, there are two obvious things that stand out. Most obviously, Lee’s BABIP is absurdly low so far this year, at .153. What this means is that Lee has been getting pretty damn lucky on balls in play, and that a regression to the mean is imminent. The other obvious thing is that Lee’s control has improved dramatically from last season. Lee walked 8.1% of hitters last season, and has walked only 1.9% of hitters so far this year. His strikeout rate is much better too, at 27.9% compared to 14.9%. His HR rate this season (zero) is obviously way too low, and will undoubtedly rise; especially considering that his FB rate is not much different from last season (46.6% as opposed to 49.7%). What has changed is that many balls that were line drives last year are now GB, and those rates will probably regress toward the mean as well.
What all this implies is that yes, Cliff Lee has certainly been the beneficiary of good luck (and honestly, who hasn’t that has a 0.28 ERA?). At the same time though, he has made some significant improvements in his game that bode well for his performance the rest of the way. His FIP ERA, which normalizes BABIP, still gives him an absurd 1.36 for this season due to his ridiculous K:BB and low HR rate. Assuming that he starts to allow HR at a somewhat normal rate, his ERA would still likely be around 3.
So now that we’ve established that Cliff Lee is indeed a much improved pitcher from years past, a secondary issue is why exactly he has been so much better. Thanks to Pitch FX data at Fangraphs, we can see what Lee is doing differently in terms of pitch selection and velocity. While his velocity is only slightly higher, he has made significant changes in the rates at which he throws his pitches. Last season, Lee threw his fastball 68.4% of the time, his changeup 15.8% of the time, and his curve/cutter/slider all between 4 and 7% of the time. This year, Lee has thrown his fastball at a much higher rate (78.4%) and his changeup at a much lower rate, now the same as his curveball (7.0%).
In other words, Lee has become much more reliant on his fastball, and apparently has been locating it very well, as evidenced by his ridiculous BB rate. According to a friend who is a rabid Indians fan, the anecdotal evidence is that Lee has been effectively painting the corners with his fastball and daring hitters to swing (and miss).
A final criticism that has been leveled against Lee is that he has been successful against below average MLB lineups. While this is certainly true given that he has faced the Royals, the Twins, and the Athletics twice, it also true that this is a run of dominance almost unprecedented in MLB history. As BP’s Rany Jazayerli correctly points out in that post, “half the pitchers in the majors today would struggle to put up those numbers if they got four starts in the Midwest League.”
Taking all this into account, it appears to me that Cliff Lee is indeed a decent MLB pitchers once again. He will not put up these Bob Gibson-esque numbers the rest of the way, and I would guarantee that every single one of his peripherals will regress to the mean, but I would not be at all surprised if he finished the season with an ERA between 3 and 3.75, with 3.75 being more likely. He almost certainly will hit a few roadblocks soon, but it is clear that he has regained some of his magic. With Carmona and a Sabathia that is returning to form, Lee gives the Indians a scary front of the rotation— one that could make some serious noise in the post-season.
Stay tuned for Lee’s fifth start tonight (against the Mariners). My prediction: 7 IP, 3 ER, 5K, 2BB.
2 Responses to “Is Cliff Lee For Real?”
- 1 Pingback on May 2nd, 2008 at 11:01 am
Leave a Reply
You must login to post a comment.
Nice work.
Who is regression to the mean going to hit harder? Cliff Lee or Fausto Carmona.
I really wouldn’t be surprised if Carmona ended up getting TJ or something at the end of the year, because his control sucks right now and has since the end of last year. He’s got 26 walks and 13 strikeouts in 34 2/3 innings.