Barry Zito as a Disappearing Exception

Remember when a seven-year, $126 million deal seemed pretty crazy? Sure, Jayson Werth’s similar contract (signed prior to the 2011 season) does not exactly seem great (although he is having a nice year) these days, either, but back during the 2006-2007 off-season, when Barry Zito signed with the Giants, it seemed utterly insane. It was not simply that it was, at the time, the biggest contract ever signed by a pitcher. Prices go up — hot dogs cost more now than ever before, and I do not see any moral panic about that phenomenon. It was that Zito seemed like a poor choice for such a deal. Sure, he was a three-time All-Star and the 2004 American League Cy Young winner. He was durable, as he had pitched more than 210 innings six seasons in a row for the As. However, Zito was going to be 29 in 2007, and his strikeout and walk rates, which had never been all that impressive, seemed to be getting worse. It looked like it was going to be an albatross.

It was. Although he has had his moments over the last seven seasons, as a final cherry on top (insert a joke about the 2014 option here), Zito’s 2013 season has been one more blow to the contract year theory of player performance. In the last season of his notorious contract, Zito has had the worst year of his career. He lost his rotation spot in August (although he received another shot in the rotation for a few starts later). Zito’s time with the Giants is drawing to a close, and he is not going to get a ceremonial final start at home, if that was something someone wanted. This is not meant as a career retrospective. Instead, I want to look at something that Zito seemed to have in Oakland, something that probably played a big part in the Giants’ willingness to give him the contract, something which seemed to depart after he made the move: his ability to outperform DIPS metrics.

Zito was no uniformlly bad as a Giant. He was roughly average for maybe two or three seasons, depending on how one looks at it, but he was not being paid to merely average. Moreover, he was mostly below average, and that might be generous. Whatever one makes of the Giants’ decision to sign Zito, they definitely deserve credit for working around a decision that at the time of the Vernon Wells trade, seemed to be even worse (although the Wells contract has turned out far worse than it seemed at the time). They won the World Series in 2010 (although Zito did not even make the postseason roster) and then again in 2012. Zito was even a big contributor during the 2012 playoffs. His start in Game Five of the National League Championship Series versus St. Louis was particularly impressive, as he went seven and two-thirds, strikeout out six and giving up no runs against a Cardinals lineups full of good right-handed hitters.

Overall, though, Zito has performed miserably as a Giant. Since 2007, he has totaled 6.2 WAR as a Giant. Even if one does not like FIP-based WAR, he was even worse according to RA9-WAR, with a total of 5.4 since 2007. This is what particularly interests me in Zito’s case, as prior to signing with the Giants, he looked like the sort of pitcher that could outperform his DIPS metrics. In his Oakland years, Zito’s FIP-based WAR was 24.9, but his total RA9-WAR was 33.1, mostly due to his low BABIP (a stunning .262), which enabled him to accumlute almost 12 BIP-Wins with the As.

This does not require dismissing DIPS. As is widely acknowledged, there are some pitchers who seem to be particularly good or bad when it comes to controlling their balls in play. The problem is identifying them over small (even one full season of starting) samples. In certain cases, though, in some cases with a large sample, there is reason to think that RA9 is a better measure of a pitcher’s contribution than FIP or another DIPS-type metric.

Barry Zito certainly seemed like one of those pitchers prior to 2007, a period in which he pitched more than 1400 innings for Oakland. While his FIP was a pedestrian 4.17, his ERA during that time was 3.55. Maybe back in his 2002 Cy Young-winning season one could have seen Zito’s 2.75 ERA versus his 3.87 FIP as a fluke, as he had only pitched a partial season in 2000 and in 2001 his ERA (3.49) and FIP (3.53) were practically the same. As time went on, Zito just kept “outpitching his FIP,” other than a hiccup in 2004.

Zito’s non-DIPS success during those years was not a total mystery. While his Oakland teammate Tim Hudson also mostly outperformed his DIPS metrics, Hudson did so by keeping the ball on the ground. Zito, in contrast, not only did not have Hudson’s control, but was a fly ball pitcher. He was probably helped in part by the As’ park suppressing his home run rate. Mostly, though, simply inducing a high proportion of fly balls — and a high number of those were pop ups — kept Zito’s BABIP down. Zito was not even that great with runners on, as he was about three runs below average in left on base wins during his Oakland years. It was all about balls in play.

What happened once Zito moved to San Francisco? There was talk about his velocity, but his velocity was never good, even during his best years. There was discussion of his delivery. Here, I simply want to focus on his results. While batted ball data is imperfect, it is reasonable to note that Zito did seemed to remain a fly ball pitcher. His batted ball rates were pretty much the same as during his Oakland years. San Francisco pitchers developed something of a reputation for outperforming DIPS metrics during this period (Matt Cain being an example, though he was not nearly as extreme in this respect as the young Zito), even to the point of maintaining good home run per fly ball rates, and Zito kept up with that, too. Yhough Zito’s BABIP was still low in San Francisco at .287, that was not enough to keep him afloat. He had a .262 in Oakland. Some of that was quickly likely simply age-based decline. And it is not as if DIPS metrics are totally irrelevant to guys like Zito. His peripherals still mattered, and both his walk and strikeout rates became worse, as one would expect for a pitcher in his thirties.

Nevertheless, it is still remarkable that Zito’s apparent ability to pitch better than his FIP seemed to disappear after the move. He still did it in 2007 for the Giants, but a 4.53 ERA versus a 4.82 FIP was not great, even for him — his FIP was even 4.89 in 2006 and he still managed a 3.83 ERA. Parks might have been a factor, but though the As’ home park is tough on hitters, the Giants are far from playing in a lesser version of Coors Field.

Even using ERA instead of FIP, Zito’s contract was a bad idea for the Giants at the time it was signed. In that respect, too, things turned out worse than one might have expected. By 2010, it began to seem reasonable to start using DIPS metrics more heavily for Zito, a trend which continued. Barry Zito seemed to be one of the exceptions back in 2007. As of the moment, his ERA for his Giants career has been 4.59, and his FIP has been… 4.59. Whatever he did with the As, with the Giants he has not been exceptional, at least not in the way they probably expected him to continue to be.