Cliff Lee and BABIP
Cliff Lee left the seventh inning of Game Two down 5-4 with runners on first and second, no outs. It was quite a disappointing start for the Phillies’ second ace as the lefty allowed 12 hits and forked over a four-run lead. It was the third consecutive mediocre post-season start for the lefty, which screamed “Narrative!” for writers across the nation.
For a fun game, I will provide two lines of pitching statistics. Your job is to pick one of them to start Game 7 of the World Series.
- Pitcher A: 20.2 IP, 5.66 ERA, 14 K, 10 BB, .286 BABIP
- Pitcher B: 17.2 IP, 7.13 ERA, 22 K, 3 BB, .463 BABIP
If you picked Pitcher A, you picked Steve Carlton. If you picked Pitcher B, you picked Cliff Lee.
Carlton put up that line in his three starts prior to the 1980 post-season; Lee compiled his in the 2010 World Series and Game One of the 2011 NLDS. Most people will recognize that the gap in ERA is meaningless given the small sample, and they will also recognize that Lee’s strikeout-to-walk ratio is much superior and that he has been BABIP-unlucky.
Carlton had a great post-season in 1980, finishing with a 2.30 ERA in four starts as he led the Phillies to their first championship in franchise history. Those who are giving up on Lee now would have given up on Carlton after the 1978 post-season, wrongfully so.
Arguably the most important thing to come out of the Sabermetric movement has been the realization that pitchers have very little control over the outcomes on batted balls (thanks, Voros!). Sure, there are exceptions like Nolan Ryan and Matt Cain, but the overwhelming majority of pitchers will find themselves close to the league average around .300. For instance, Roy Halladay‘s career BABIP is .295 while Adam Eaton‘s is .301. The difference between one of the best pitchers of this generation and one of the worst is six hits over a sample size of 1,000 batted balls. Halladay, of course, separates himself from the pack with tremendous defense-independent abilities: not issuing walks, missing bats, and inducing weak ground balls.
Throughout the game Sunday night, Saber-minded tweeters kept cursing the luck dragons for Lee’s outing. Meanwhile, traditionally-minded fans cursed Lee for pitching poorly in a crucial game. Naturally, debates rose over the validity of BABIP. Lee’s defense-independent stats (nine strikeouts, two walks, 45 percent ground balls) were good and didn’t merit his fate. BABIP wasn’t taking into account just how hard Lee was getting hit, some suggested.
If you compare the outcome of batted balls last night compared to Lee’s average on the season, you can understand why some felt he was the victim of bad luck. I posted the data in the comments of the game recap:
Batted ball distribution: 9 ground balls, 4 outfield fly balls, 3 infield fly balls, 4 line drives
Expected hits based on Lee’s 2011 BABIP…
Ground balls: .241 * 9 = 2
Actual: 5 (+3)
Outfield flies: .132 * 4 = 1
Actual: 2 (+1)
Infield flies: .010 * 3 = 0
Actual: 0 (even)
Line drives: .661 * 4 = 3
Actual: 4 (+1)
Allowed three more ground ball hits, one more outfield fly ball hit, and one more line drive hit than expected.
So, let’s actually watch the game again and see if Lee was really hit hard. I’ve made .gifs of each of the 12 hits Lee allowed. Afterwards, I’ll provide my own judgment on the hit. You, of course, are allowed to disagree but make a solid case in the comments as to why.
Rather than overload your browser with 12 .gifs, I’ve linked them next to the description in brackets. Click on it to open the .gif in a new window.
- Verdict: Well-hit fly ball, absolutely preventable by Lee pitching better.
- Verdict: Well-hit line drive. However, Hunter Pence could have judged the ball better and held it to a single as opposed to a double. We’ll give half-credit on this one (single).
- Verdict: Not a particularly well-hit ball. Given that Molina is not fleet of foot, it’s very likely Rollins throws him out if he fields this cleanly. Certainly, the Jimmy Rollins of 2009 or prior would have been able to get to the ball without so much as diving, so this isn’t really Lee’s fault here. The outcome of this grounder was very dependent on the quality of the shortstop’s defensive abilities.
- Verdict: This is just bad luck for Lee. Because there were runners on first and second, Ryan Howard was able to play off the first base bag. If, instead, there was just a runner on first base, Howard would be on the bag holding the runner and Theriot’s hit instead turns into a 3-6-3 or 3-6-1 double play (assuming Howard makes a good throw to second base). This hit was completely defense-dependent.
John Jay singles on a ground ball to right field. [Link]
- Verdict: Jay just smothered this ball, not particularly well-hit but just placed well. If it’s five feet to the right or to the left, it’s an out.
Rafael Furcal singles on a line drive to left field. [Link]
- Verdict: This was labeled a line drive but I don’t think it was hit well enough to merit the classification. In reality, it was a weakly-hit fly ball with a low trajectory. This goes back to Colin Wyers’ suggestion of bias in batted ball data. In an article for Baseball Prospectus, Wyers said, “no matter how much we massage the data, there simply is not a way to objectively define the difference between a fly ball and a line drive. It is inherently a subjective and somewhat arbitrary distinction.” In looking at the .gif, I don’t think Lee did anything wrong. He made a good pitch (an inside cut fastball off the plate), did not miss his spot (check out Furcal’s triple in the first inning to see him miss his spot), but Furcal just put it where the fielders were not. I can’t fault Lee on this.
Ryan Theriot doubles on a line drive to left field. [Link]
- Verdict: Lee simply caught too much of the plate here and Theriot put a good swing on it. Lee could have gotten an out on the pitch if he had located it better.
John Jay singles on a ground ball to left field. [Link]
- Verdict: As with his single in the fourth, there was simply nothing Lee could have done about this. He made a good pitch and hit his spot, but Jay simply put it where there were no fielders. Just bad luck for Lee.
- Verdict: Just by the description, you know it’s not Lee’s fault. This is heavily influenced by the positioning of the middle infielders, who were leaving the middle of the field open in order to corral any ground balls to the right of third base and to the left of first base. If Schumaker was a bit slower, or if Utley makes a slightly better throw, or if Schumaker doesn’t dive head-first, the call may have been different. Lots of variables in this one, but none of them can be pinned on Lee. He made another good pitch and hit his spot.
- Verdict: This is a tricky one. Although Lee missed his spot, he didn’t make a bad pitch. Give Craig credit for putting a good swing on the ball. Additionally, if Shane Victorino reads the ball better off the bat, he catches the fly ball easily. Instead, he took a bad first step and it cost him as the ball glanced off of his glove. I think this is an out 100% of the time if Victorino reads the ball better, so I’m not assigning Lee any blame here.
- Verdict: This is a tough one to judge since the TBS broadcast team was busy looking at shiny objects while baseball was being played. It looks like Lee just left a first ball cutter over the middle of the plate and Pujols hit the ball hard. I give Lee the blame on this since he could have had a better outcome with better pitch selection and better location.
- Verdict: Nothing you can do about this. Not a terrible pitch, mediocre location, but it induced a horrible swing out of Lance Berkman, but the weakly-hit fly ball dropped before any fielders could get in range. Not at all Lee’s fault.
- Furcal triple: Lee’s fault
- Freese double: Lee’s fault (but just a single)
- Molina single: nope
- Theriot double: nope
- Jay single: nope
- Furcal single: nope
- Theriot double: Lee’s fault
- Jay single: nope
- Schumaker single: nope
- Craig triple: nope
- Pujols single: Lee’s fault
- Berkman single: nope
In the end, I only have Lee on the hook for four of the 12 hits. Now, to be fair, I’d have to do the same exercise for the eight batted ball outs, but the point of this exercise should be evident. As mentioned, Lee pitched well in the areas that didn’t involve his fielders: he struck out nine in six-plus innings of work, walked two, and induced ground balls at a 45 percent rate. If you roll the dice again with the same defense-independent numbers, Lee will have a much better night a majority of the time. As many will point out, though, baseball games are only played once and Lee’s results are what they are.
However, we can be a bit more fair and rational when assigning blame for the Phillies’ loss in Game Two. The Phillies’ offense, for example, went down in order in four consecutive innings between the third and sixth innings. Even if Lee forked over the lead, the Phillies should have been expected to push across at least one more run after they went up 4-0 after two innings. Unfortunately, they did not and the Phillies lost an otherwise winnable playoff game.