Baked Zito
Following my post last week about Carl Pavano‘s contract a lively discussion ensued in which it was argued that Barry Zito‘s was much worse. The reasoning for this being that Pavano’s absence didn’t necessarily hurt the Yankees whereas Zito would actually be hurting the Giants with his poor pitching each start. Yes, it bordered on the absurd to offer him a 7-yr/126 million dollar deal following the 2006 season, but, in looking at his numbers I found that his 2006 seasonal line and current numbers in 2008 shared a similarity.
In 2006 he produced a 4.89 FIP; so far this year he is at 4.98. In 2006, however, his ERA ended up 3.83, a number much lower than the current 5.80. Essentially, both of these seasons resulted in very similar metrics of controllable skills yet one involved an ERA a full point lower and the other is featuring an ERA almost an entire point higher. What’s the difference?
For starters, he has allowed a ton of baserunners this year, posting a 1.80 WHIP following his poor performance last night. He always walked plenty of hitters but this year his walk rate is higher and his BABIP, which hasn’t ever been higher than .300 in his career, is currently .332. On top of that, his strand rate isf 65.2%, meaning that he is allowing baserunners at an insanely alarming clip and allowing them to score at a rate well below the average. In 2006, his strand rate was well above average at 78.5%.
In 2006 he walked 4.03 batters per nine innings while fanning 6.15 per nine. This year, the strikeout rate is similar, coming in at 5.88 K/9, but the BB/9 has risen to 5.63. Despite the vast increase in walks, his home run rate is down, which has allowed the FIPs in each of these seasons to be within .09 of one another. Using the in-season Marcel projector, Zito is expected to produce this over the remaining two months:
2008 Remainder Marcel: 12 GS, 60 IP, 64 H, 30 BB, 44 K, 1.57 WHIP, 4.55 FIP
Compare then his overall line with this projection to his 2006 season:
2008: 33 GS, 169.0 IP, 192 H, 98 BB, 115 K, 1.72 WHIP, 4.82 FIP
2006: 34 GS, 221.0 IP, 211 H, 99 BB, 151 K, 1.40 WHIP, 4.89 FIP
Dan Szymborski, creator of the ZiPS projection system, was kind enough to send me his in-season projector this weekend and so what happens when we plug Zito into that?
2008 Remainder ZiPS: 12 GS, 68 IP, 67 H, 34 BB, 49 K, 1.54 WHIP, 4.55 FIP
Substitute ZiPS for Marcel in the 2006 vs. 2008 comparison to get the following:
2008: 33 GS, 176.2 IP, 195 H, 102 BB, 120 K, 1.70 WHIP, 4.82 FIP
2006: 34 GS, 221.0 IP, 211 H, 99 BB, 151 K, 1.40 WHIP, 4.89 FIP
And, in comparing the two projections for Zito over the remainder we see virtually the same results: He is not as bad as his numbers thus far would suggest but he isn’t necessarily good either:
Marcel: 12 GS, 60 IP, 64 H, 30 BB, 44 K, 1.57 WHIP, 4.55 FIP
ZiPS: 12 GS, 68 IP, 67 H, 34 BB, 49 K, 1.52 WHIP, 4.55 FIP
It seems that Zito was not as good as some of his 2006 numbers would suggest nor is he as bad as some of his 2008 numbers suggest. He’s not a very effective pitcher anymore but he is a very expensive and handsomely paid one. If we didn’t know his name and just heard about a pitcher with his numbers coupled with an 84-85 mph fastball and a lack of control, I would tend to think we would all question how he is a major league pitcher. We may not be at that point completely yet with Zito, but that corner may be turned quite soon.
19
I just thought it was funny when Sunday Night Baseball on ESPN screwed up for the Sox / Yanks game. When they brought the diamond up showing the defense for the Yankees, it showed Pavano as the pitcher when Ponson was starting.
I don’t think you meant to say he’s letting baserunners score “more often than not” that would put his strand rate at less than 50%, no? That would put his ERA at somewhere around 9.
As in he’s letting them score at a rate well below the average.
The p-value for Zito’s BABIP for his career until 2007 compared to 2008 by itself being equal is .009. This means (roughly) there is only about a .9% chance that Zito’s “ability” to turn balls into play into outs has not diminished now compared to the rest of his career. Given that there are so many major league pitcher’s we are bound to find cases this extreme, but it does make you wonder if Zito is really pitching so bad that his BABIP is actually inflating due to more than random chance.
Dave, the p-value you’re offering is a tremendous sign that Zito’s ability or true talent with regards to his BABIP hasn’t changed, though it doesn’t disregard or make up for the fact that it currently is where it is, which is contributing to his lack of success.
From here on out, based on what we know about his true talent level it is likely to regress. He has definitely lost something, however. His FIP was never higher than 4.50 and from 2006-now has been in the 4.80s; his 87-88 mph fastball is now down to 84.5 mph; his K/9 has decreased and his BB/9 is at an all-time high for him. So the BABIP may be more chance than “suckiness” but his numbers elsewhere are not exactly showing signs of improvement.
Over the course of his career with the A’s, Zito induced tons and tons of popups. In his two years with the Giants he’s been more pedestrian in that regard. I don’t know if this due to a change in velocity, fielding, park, or luck (or all), but it’s part of the reason for his decline.
Remember that popups are just about as good as strikeouts, so the decline in his K-rate was somewhat mitigated by the fact that he was getting all these infield flies. Now that he’s not striking anybody out and no longer inducing those popups.
More info:
http://www.insidethebook.com/ee/index.php/site/article/solving_barry_zito/
http://www.lonestarball.com/story/2006/12/17/20939/021
Sal, thanks for the links. I hadn’t seen that work on Zito before but it’s fascinating. For anyone coming here who doesn’t have time or hasn’t read the articles linked, essentially, Barry Zito has had a league average BABIP against lefties in his career but is ~40 points lower against righties. A standard deviation would be 8 points, meaning Zito is five standard deviations from the mean which, as Tom Tango said, is as significant as significant gets.
The reason was that he induced so many weak flyballs. It wasn’t luck that the balls were being caught but rather somewhat of a skill that he was inducing popups and weakly hit balls.
Now, without the popups and without the strikeouts it’s no wonder his results are declining.
Are caught fouls considered pop-ups? Because the Oakland stadium has so much more room for those to get caught than SF.