Watching the Game: Cliff Lee
With a loss yesterday afternoon at the hands of the struggling Miami Marlins, the Phillies dropped to 36-45, 10.5 games out in last place in the NL East. With a 9-19 June and a loss to start July, the season is all but over for the Phillies. It would require divine intervention to make the playoffs at this point: to reach an assumed 88-win threshold for the new second Wild Card, the Phillies would have to win 64% of their remaining 81 games — the pace of a 104-win team over a 162-game season. It’s been this kind of season:
The frustration of this season overall, combined with the unexpected under-performance of several key players, has resulted in some unfair lashing-out by Phillies fans. Cliff Lee has been one such target, but more so than anyone else, it seems that the criticism directed at him is most unwarranted. Looking at the factors most in his control — strikeouts and walks — Lee has been doing his job nearly as well as he did last year. When you look at things he has little or no control over — RISP, W-L, ERA — he doesn’t look so hot.
Philly fans have a tendency to blame the better players for a disappointing team performance, as Donovan McNabb can verify. Coupled with the proneness to error simply by being human, we have an ever-growing myth on our hands. With the perception that Lee is irreparable, fans start magnifying every little mistake and assigning more blame to Lee than is warranted. I saw no acknowledgement of any outside factors on Lee’s results on Sunday, so I want to illustrate that by going through Lee’s last start against the Marlins in which he lasted four and two-thirds innings, on the hook for six runs on ten hits.
Lee allowed one run in the first, two in the third, and three in the fifth, so I will focus on those innings. Lots of .gifs after the jump.
- Jose Reyes double down the left field line
Lee missed his spot here by about four inches. Even so, this is an easy 5-3 ground out if Polanco is playing in a normal defensive alignment (5-10 feet behind the third base bag), instead of preparing for a potential bunt from a known-bunter and very fast runner in Reyes. This is simply bad luck. With the bases empty and one out, 0.26 runs are expected; with a runner on second and no outs, 1.08 runs are expected, per Baseball Prospectus.
- Hanley Ramirez fly out to center field
This just shows how bad luck can compound itself. If Reyes had been retired on his grounder, as most other right-handed batters would have, then this is an innocuous fly out to center field by Ramirez. Or if a slower player had doubled, the runner doesn’t tag up and go to third base on this play. Both of which Lee has no control over, as validated by many years of independent research on balls in play. Expected runs with no runners on and two outs: 0.10; with a runner on second base and one out: 0.66; with a runner on third base and one out: 0.90.
- Logan Morrison 4-3 ground out
Here is where it’s very easy to muddy the waters. There is a fundamental difference between “pitching badly” and “fielding badly”. Lee made a great pitch to Morrison, but because of the outcome of this play, we simply assign general blame to Lee. When compounded with later results, this play is simply thrown in the mix. Moreover, Lee was not assigned an error on this play because he technically didn’t touch the ball, so it counts against his ERA. If Lee fields the ball properly, it’s an easy inning-ending 1-6-3 double play.
- Hanley Ramirez line drive double to left field
This was not a very good pitch by Lee. Still, it’s just a single with most batters and with most left fielders. However, Juan Pierre has one of the worst outfield arms in baseball (if not the worst), and opposing teams know this. The Marlins frequently challenged Pierre’s arm throughout this series, and Ramirez did so here. If Giancarlo Stanton was instead the batter, it’s just a single. Likewise, if it’s Hunter Pence in left field, Ramirez stays at first base. Lee has zero control over this, even though he made a bad pitch. Expected runs with a runner on first base and one out: 0.51; with a runner on second base and one out: 0.66.
- Giancarlo Stanton single to left field
Lee was just completely unlucky here. If Stanton’s grounder is hit five feet to the left or five feet to the right, the Phillies have an easy play at first base, or perhaps can catch Ramirez between second and third base. Additionally, Pierre boots the ball, so there isn’t even a play at home. Knowing Pierre’s weak arm, the Marlins were sending Ramirez home all the way. Compounded with the unluckiness of the hit, there was zero chance that the run is even challenged as opposed to a partial chance if Pierre doesn’t boot the ball, and a likely chance if the left fielder had at least an average arm. Again, Lee has no control over any of this.
- Logan Morrison single to left field
Again, just simple chance at fault here. Lee makes a decent pitch and induces a terrible swing from Morrison. Still, it finds a hole in the vast outfield at Marlins Park. From an amateur perspective, it also seems like Pierre was playing too deep for a left-handed pull-hitter facing a left-handed pitcher. Here is Morrison’s hit location chart against left-handed pitchers, going back to late July 2010:
There must have been a reason Pierre was playing where he was, but it seems like it was incorrect.
- Justin Ruggiano single to left field
This is eerily similar to the Stanton single, finding the same hole between third base and shortstop. Likewise, if it had been hit five feet to the right or to the left, it’s an easy 5-4-3 or 6-4-3 inning-ending double play. Instead, the bases are loaded as the Marlins’ third base coach inexplicably held Stanton at third base. (The Marlins’ broadcasters, as the play happened, shouted, “Send him! Oh, you gotta send him!”) Did Lee make a great pitch? No, but every pitcher makes mistakes — even Roy Halladay. They just don’t pay for them at the rate Lee has paid for his this year. Pitchers still get an out more often than not on mistake pitches. For instance, between 2009-11, opposing hitters had a .302 average on pitches Lee left in the middle of the plate, compared to .247 overall.
- Omar Infante sacrifice fly to center field
Lee made another mistake and Infante barely missed hitting a grand slam. It was one of the rare instances where Lee had a fortunate outcome. However, it was only compounded by the bad luck he’d experienced prior to this at-bat. It could have been a harmless fly out, but instead, a run was able to score on the play and both of the other runners advanced.
- Giancarlo Stanton line drive single to left field
All of the credit goes to Stanton here. He put a great swing on a decent pitch by Lee and served it into left field. Not much else you can say about this.
- Logan Morrison ground-rule double to right-center field
(The .gif is slower here because the MLB.tv condensed game didn’t show a full-speed version of this play for some reason.)
Morrison puts a great swing on a decent pitch by Lee, sending it into deep right-center. Hunter Pence took an absolutely awful route to the fly ball, which absolutely should have been caught and would have been caught 99% of the time. Instead of having a runner on first base and two outs (expected runs: 0.32), the Marlins had runners on second and third with one out (1.30). Additionally, Pence doesn’t get an error for this play because he never touched the ball. As a result, the runs Lee subsequently allows are earned and count against his ERA.
- Justin Ruggiano doubles to center field
This is a flat-out awful pitch by Lee and Ruggiano made him pay. Still, because of Pence’s bad play in right field, two runs score on this play instead of one, and because Pence wasn’t credited with an error, both runs and all subsequent runs are earned and count against Lee’s ERA.
- Justin Ruggiano steals third base
Presumably, the Phillies weren’t holding Ruggiano on and were playing deep on the left side because there were two outs and behind only four runs in the fifth inning. Still, this play ends up having an impact and Lee couldn’t have done anything to prevent Ruggiano from stealing third base here. Gaby Sanchez, who is batting while Ruggiano steals third, ends up drawing a walk to extend the inning.
- John Buck single to left field
This was the final straw for Charlie Manuel as he pulled Lee from the game after this hit. It was Lee’s 96th pitch and he was presumably gassed after so many high-stress pitches, thanks to shoddy defense and bad luck. It wasn’t exactly an awful pitch, but a very hittable one and Buck put a good swing on it. However, if Ruggiano hadn’t been allowed to steal third base so easily, Pierre might have had a play at home, or the Marlins’ third base coach might even have decided to hold Ruggiano. After all, the ball did have velocity upon entering left field. If Pierre had charged the ball in this event, it might have changed things.
Did Lee pitch well in this game? Not really. He only struck out three of the 26 batters he faced (11.5%) and walked two (7.7%). His strikeout and walk rates on the season are 20% and 6%, and were 26% and 5% last year, respectively. Did he pitch like a 11.56 ERA pitcher (6 ER in 4.2 IP) would perform? Not at all.
What this exercise should show you is how faulty the popular methods of evaluating pitching really are. Earned runs are subject to the scorer’s decisions and in the case of this game, Lee allowed three runs that should not have counted against his ERA. 3 ER in 4.2 innings (5.78 ERA) matches up slightly better with his performance. Similarly, Lee’s defense did him no favors even discounting the should-be errors. The Marlins took advantage of Pierre’s weak arm, the Phillies’ seemingly-illogical outfield alignment, and their lack of concern over base-advancement late in the fifth inning.
Outings like this reinforce why stats like BABIP and ERA retrodictors like xFIP and SIERA are so important — they remove the impact of a pitcher’s defense and luck from the equation. Outings like this also reinforce why older stats like ERA and W-L are faulty — they are impacted by many factors out of the pitcher’s control, including luck, defense, and in the case of the latter stat, run support. Lee may be 0-5 with a 4.13 ERA, but he has not been pitching bad this year, even despite his last three starts. Anyone who says otherwise simply isn’t watching the games closely enough.