[Editor's Note: I've invited Kyle Boddy of Driveline Mechanics to analyze an Indians pitcher. Here's his statistical and mechanical analysis of Cliff Lee's amazing start to the 2008 season - Ryan]
Phrases that I heard about Cliff Lee over the course of the 2007 season:
- He’s a bum!
- Good riddance - enjoy Double-A, you idiot.
- We should have traded you for Matt Murton while we had the chance!
Phrases that I have heard about Cliff Lee over the course of the 2008 season:
- We love you!
- Never leave!
- Thank God we didn’t trade you for Matt Murton!
Funny how things change, isn’t it?
Of course, when you’re posting numbers that would make former Indians’ hurler Cy Young blush, people tend to notice. Cliff Lee’s stats currently look something like this: 53.2 IP, 0.67 ERA, 4 BB, 44 K, 1 HR. Yes, you read that right - a 11:1 k/bb ratio and allowing only one home run (which, need I remind you, barely got out of Progressive Field courtesy of Wladamir Balentin). Lee has thrown a complete game shutout this year and would have had another if the Indians could have scored a single run in the first nine innings of the nightcap against the Blue Jays. (In fact, the Indians starters almost combined for back to back complete game shutouts that day by Fausto Carmona and Cliff Lee.)
So how’s he doing it? Well, besides striking out tons of batters, not walking anyone, and not giving up extra-base hits, he is sporting a quite-low BABIP of .224. Though the Indians’ rate 8th overall in Defensive Efficiency (DefEff) at 0.713, that still means the approximate expected BABIP of the pitchers on the Tribe should be around .287. To further calculate Lee’s expected BABIP (eBABIP), we can take his GB/FB/LD rates and plug it into this formula:
.763LD% + .265GB% + .131FB%.
That formula represents the best fit line for all batters with 300+ PA in 2006, and while it has probably changed since then, it’s still a good and quick tool to estimate BABIP.
For those who would like to know more on BABIP and what it means, you can read this excellent primer on it. In short, pitchers show very little ability to control what happens after the ball is put into play, and therefore BABIP remains stable for all pitchers - including Cy Young winners like Brandon Webb as well as bottom of the barrel guys like Matt Morris. A few pitchers have demonstrated the ability to depress BABIP rates to a small extent, such as Barry Zito and Jamie Moyer. Stadiums with above-average foul ground territory aid in this process, as does the ability to generate large amounts of Infield Pop-Ups.
Lee has allowed 17.3% line drives, 47.5% ground balls, and 35.3% fly balls (source: Fangraphs). Using the above formula, Cliff Lee’s eBABIP is 0.304 - quite a bit higher than his current .224. (The current major league average BABIP is about .299.) Through that, we can expect Lee to regress back towards the mean, which will also correspond with an increase in his ERA/WHIP. Lee’s Fielding-Independent Pitching stat still rates at 1.83, which is basically his expected ERA calculated from BABIP regression.
Here’s what Cliff Lee’s GB/FB/LD rates have looked like over his career (source: Fangraphs):
Notice anything? In 2008, he has swapped his GB/FB rates! Lee has always been an extreme flyball pitcher, leading to a career-high 30 HR allowed in 2004, but this year he is generating tons of ground balls. I wish I had Pitch f/x data, but I have had no luck trying to download it and parse it despite following various walkthroughs available on the Internet. Hopefully Josh Kalk updates his Pitch f/x tool for lazy people like myself.
Another important thing to note is that Cliff Lee typically allows 10% of his flyballs to go over the fence. This year, he’s at just 2%. That is also not a sustainable number, as HR/FB is usually very stable across wide ranges of pitchers. Lee is also stranding 90% of runners who reach base, which is simply ridiculous - the MLB average is just over 72%, and again, this is a stable number from year to year as well.
By now I hope you have realized that while Cliff Lee is very lucky this year, he is also doing the three things that sabermetricians look for in a front-line pitcher:
- Strike a lot of batters out.
- Don’t walk a lot of batters.
- Don’t give up extra-base hits.
Pitchers who do these things are aces year in and year out. Some pitchers do two things exceptionally well, which is good enough to be a top of the line starter. Pitchers who do all of the above are your Johan Santanas and Jake Peavys.
As an Indians fan, I sure hope he keeps up the torrid pace!
Now that we’ve dissected Cliff Lee from a statistical point of view, let’s check out his pitching mechanics:
Since I have nothing to compare him to, I can’t tell if he has made mechanical changes in 2008. If anyone has video from previous years (particularly 2007), please forward it to me.
Tempo: Lee is pretty slow from maximal leg lift into footplant, clocking in at 20-21 frames. However, he’s not a power pitcher, and he’s a lefty, so this isn’t too major of a concern. I’ll grade it as Good.
Arm Action: Lee reverse-rotates his shoulders as he starts to bring the ball out through his hand break. He does not use a classic pendulum swing, but neither does he break his hands with his elbows, so there’s no need for concern there. Lee has an aggressive lower body motion, “sitting” down as he drives to the plate. As readers should know by now, I like aggressiveness into footplant through tempo as well as this motion (you can see perfect examples of both in Roy Oswalt’s delivery). At footstrike, Lee’s arm has passed through the horizontal but it is not into the high-cocked position. He will experience more stress on his UCL and anterior structures of the shoulder than a pitcher like Greg Maddux, but his arm is an acceptable enough position for me. I’ll grade it as Good.
Ball Release: Lee aggressively pulls down with his head and points the PAS shoulder at the target. Like I said about Max Scherzer, I think this is a very good thing. The glove is up and around his GAS shoulder. Lee gives himself every opportunity to have a great followthrough phase. Grade: Excellent.
Followthrough: Like Max Scherzer, Lee has an aggressive shoulder rotation. Unlike Max Scherzer, he actually follows through with his arm and does not actively brake it. However, Lee pulls his elbow behind his body rather than keeping his front side firm. Pitchers can firm up the front side by focusing on pulling their hand back, not their elbow. Regardless, it’s not that big of a deal since I like his followthrough and recovery of the pitching arm. Grade: Very Good.
All in all, I like what I see, and it matches up with Lee’s clean bill of health (his DL stint was due to an abdominal strain, nothing arm-related). This Indians’ fan hopes that he keeps his mechanics clean and continues to put up plenty of zeros on the scoreboard!
For more statistical and mechanical analysis of pitchers and hitters, check out my website: Driveline Mechanics .