The new FDP metric at FanGraphs provides useful information about extra wins earned by pitchers due to non-FIP events. How do Warren Spahn, Tim Hudson and Phil Niekro stack up?
Hey there, and welcome back to the third piece in my series on FDP and great Atlanta Braves starting pitchers. Part 1 of this series was posted on Monday, and Part 2 went up yesterday, in case you'd like to circle back and get some more context. And please, don't forget to drop by FanGraphs and read Dave Cameron's two introductory pieces on FDP: Introducing Fielding Dependent Pitching and FDP and Pitcher WAR. Are we all up to speed now? Awesome. Let's talk a little about two historical Braves starters, as well as one current Atlanta starter.
Ok, so Warren Spahn had every opportunity during his career to rack up copious amounts of WAR and FDP-Wins. The guy is eighth all-time in innings pitched with 5243.2 over his distinguished 21-season career. But can you believe he was worth a whopping 35 FDP-Wins over that career? To put things in perspective, Brandon Webb, Greg Swindell and Mike Hampton all put up less than 35 RA9-Wins over their career -- and remember that RA9-Wins include both FDP-Wins AND fWAR!
I brought up Tom Glavine yesterday as an example of a pitcher who benefits greatly from value added through FDP-Wins. Well, compared to Spahn, Glavine's total doesn't even seem all that impressive. Spahn has nearly 1000 innings on Glavine, but also has about 10 FDP-Wins on him.
Due to this massive load of FDP-Wins over his career, Spahn sits at 12th all-time in RA9-Wins (114.4), despite only landing at 27th all-time when it comes to fWAR (79.3). That's quite a jump, especially considering that many players atop him on the fWAR leaderboard get substantive bumps from FDP-Wins as well.
According to the component parts of FDP-Wins, Spahn was able to get the lion's share of those FDP-Wins by limiting hits on balls in play; his BIP-Wins score is 23.7. He was also consistent in logging these BIP-Wins, only four of his 21 major-league seasons saw him log negative BIP-Wins. There's a definite pattern here of Spahn adding value due to the placement of balls in play ... however much of that value you wish to assign to Spahn, in comparison to his defensive teammates or park, is up to you.
Braves (and especially Athletics) fans probably don't need me to tell them this, but Tim Hudson kind of rules. Hudson has, very quietly, amassed 51.2 fWAR over his career. That's no joke -- Tim Hudson has a world-class, historically-relevant career. That's more fWAR than Dave Stieb, Roy Oswalt, Ron Guidry, and Jamie Moyer. But Hudson has also racked up 14.6 FDP-Wins over his career, most of those coming as BIP-Wins (10.9), and many of those coming in the last five seasons. That amps up his RA9-Wins to 65.8 over his career -- which puts him in a tier of starting pitchers that sits right below the Hall of Fame. It's just one metric, but it's one that has Hudson providing more career value than all but 79 other pitchers in MLB history. Whoa.
Think about that for a second. According to RA9-Wins, Tim Hudson is one of the 80 most valuable starting pitchers in major league history. Being able to pitch 200 innings a season over a long stretch of time, at an above-average level, is worth a heck of a lot. And, for what it's worth, his career value numbers like fWAR and IP compare favorably to the very-similar Mark Buehrle. As of this writing, the two pitchers have almost identical IP numbers, but Hudson has been worth about 3.5 more fWAR over his time than the Marlins hurler. Hudson gets the benefit of a lot more BIP-Wins over his career than Buehrle, and their LOB-Wins are pretty close, so Hudson also has an edge in RA9-Wins. Hudson, though, is almost four years older than Buehrle, so the lefty has plenty of time to pad his career stats.
It's a relatively well-known axiom that pitchers who throw the knuckleball are supposed to induce weak, weird contact on balls in play. Well, the great knuckleball artist Phil Niekro's BIP-Wins score may help support this argument -- or perhaps not. Niekro, in his 24 seasons and over 5.4k innings pitched, logged 11.9 BIP-Wins, which is a fairly high number. Like Tom Glavine, one might expect it to be higher given his reputation and his number of innings thrown, but it is still nothing to sneeze at.
Tangent alert! I took a little jaunt through the BIP-Wins of several other prolific knuckleballers (since 1947), in order to see if BIP-Wins for knuckleball pitchers was consistent, or if other knuckleballers had an easier / harder time in regards to BIP-Wins.
Well, that settles it. Most of these pitchers had pretty substantive bumps in BIP-Wins, including Niekro, though it was over a pretty sizable number of innings. Hough really jumps out at you, doesn't he? Same with Wakefield. And boy, did both of those guys have trouble stranding runners. I wonder how much of it has to do with walks? Or if knuckleballs are tougher to field? Or if it's because runners on base have more freedom to run when a slow-acting knuckler is in the air? So many questions.
Keep in mind that, Niekro also racked up more fWAR (84.6) than Glavine, Smoltz, Spahn, or any other pitcher in this article save Greg Maddux. So he was in no way a "fielding-dependent" pitcher, despite relying heavily on his knuckler. Like Smoltz, Maddux and Hudson, Niekro also didn't benefit greatly from a high LOB-Wins score, only racking up 0.9 wins through that metric over his career.
The Big Finish
In closing, most MLB teams should be so lucky to have such a tremendous history of pitching talent like the Atlanta Braves do. And FDP can help tell us just a little more, statistically, about these historical aces.
Perhaps Tom Glavine and Warren Spahn really were greater than the FIP-based metrics make them out to be. Perhaps Tom Glavine wasn't quite as good at inducing weak contact as many thought. Perhaps John Smoltz really was one of the top 25 or 30 pitchers in baseball history. Perhaps Tim Hudson deserves some down-ballot Hall of Fame consideration. Perhaps Tom Glavine really was one of, if not the best, pitcher in modern history at stranding baserunners.
There's a lot of thinking and examining left to do before coming to any definite conclusions regarding FDP, but for the time being, it's another useful way to look at pitching performance from a different angle. And it's also a reminder that, while we haven't figured everything out yet in terms of separating a pitcher's performance from that of his teammates', the sabermetric community is still moving forward in a couple of new and interesting ways. I'd love to hear more voices exploring this FDP data.
Update: After I finished this article, and thanks to a kind re-posting on Tom Tango's The Book Blog, Michael A. Humphries, author of Wizardry, presented me with one of his online appendices, which can be found here. This appendix speaks to the fielding capabilities of four of the six pitchers I've discussed in my articles: Glavine, Maddux, Spahn, and Niekro, using DRA. Check it out, and help me parse through some of this info. If Niekro's associated teams had defense that was really as bad as Michael implies, what does that tell us about his BIP-Wins number? Maybe it should be even better than it really was? If so, well, cool.