October 14, 2010
The Glavine Line
I’m excited to join Baseball Prospectus. If you’ve read any of my previous work, you may know me as something of a PITCHf/x guy. I’ve been learning about and writing about PITCHf/x since the pitch-tracking system was installed in major-league ballparks in 2007, so that description is apt. My interests extend beyond PITCHf/x to the physics of baseball and the details of the pitcher-hitter confrontation.
I wrote at the Statistically Speaking blog back in 2007 and 2008 alongside Eric Seidman and former Baseball Prospectus author Russell Carleton (a.k.a. Pizza Cutter in his Stat Speak days). I’m happy to add to the growing count of Stat Speak alums populating the Baseball Prospectus authorship ranks. From 2008-10, I wrote for the Hardball Times, and I’d like to thank Dave Studeman and the rest of the team there for their contribution to me personally and to the study of baseball.
Can anyone win 300 games again?
Today, I want to investigate something that Seidman mentioned in his column a few weeks ago about baseball records in the context of era. Eric made an offhand statement that he “would bet money that nobody wins 300 games again, and maybe only a couple of pitchers reach 250.” His claim generated quite a bit of discussion in the comments and prompted me to look at who might be considered the active pitchers most likely to reach 300 wins.
It turns out that this is a subject oft discussed. That’s not surprising, I suppose, but I didn’t realize until I did a bit of research for this article how often it has been written about in the past few years, not least by Jay Jaffe and Joe Sheehan on these very pages on the occasion of Randy Johnson’s 300th victory. Jay applied the Bill James Favorite Toy method to identify the most likely candidates for 300 wins, and I’ve done something similar.
It’s interesting to note that only a season and a half later, though the top two candidates are probably the same (CC Sabathia and Roy Halladay), the chances of several of the pitchers right behind them on the list have taken significant hits. Carlos Zambrano was in and out of the rotation this year and fell further behind a 300-win pace as a result. Johan Santana, Jake Peavy, and Josh Beckett all missed time due to injury and had their prospects dim.
Historical 300-win hits and misses
Given that this is a subject about which no one knows the right answers, I’m going to forsake the rigor of the Favorite Toy method, such as it is, and look at this graphically. First, let’s look at how the recent batch of 300-win pitchers made it to that mark, as well as at the fortunes of several of their contemporaries who displayed some promise of reaching the goal at some point.
Greg Maddux leads the pack. He started strong in his early 20s and never let up, winning 15-20 games every year from age 22 until he crossed the 300-win plateau at age 38. He added another 50 wins in his last four seasons to finish with 355 at age 42. Roger Clemens followed close behind, not quite as consistent as Maddux but crossing the 300-win mark at age 40 and finishing with 354 at age 44.
Tom Glavine nosed over 300 wins at age 41 and finished with 305. He didn’t get off to quite as strong of a start in his early 20s as Maddux and Clemens, but he made up ground with three 20-win seasons from 1991-93 and cruised through age 36 winning from 13-21 games every year for Braves teams that went to the playoffs for eight straight season.
Johnson got a late start. He recorded his first major-league win at age 24 and earned his 100th win at age 32. He was nearly 100 wins behind Maddux’s pace for most of his 30s. However, he continued to pitch effectively in his early 40s, winning 16, 17, and 17 games at age 40, 41, and 42, putting him within hailing distance of 300 wins. Johnson battled injuries but pitched effectively enough for another three years to end his career with 303 wins. As Sheehan wrote, “Johnson’s career shows that it’s not that we’ll never see another 300-game winner. It’s just that we’ll never see him coming.”
Dwight Gooden stands out. He had 100 victories by age 24 and 154 wins by age 28. Unfortunately, injuries and hard living took their toll, and Dr. K was unable to reach even 200 career wins.
Pedro Martinez was another pitcher who showed great promise to achieve 300 wins. After his age-28 season, his career record stood at 125-56, and he had just compiled one of the greatest four-year stretches by any pitcher, with an outstanding 2.16 ERA and an average of 288 strikeouts per season. He was only a handful of wins behind the Maddux/Clemens pace, and he appeared to be every bit the pitcher they were, or even more. Though Martinez would go on to post a few more very good seasons, injuries derailed his career, and he finished with a 219-100 record at age 37.
It’s possible that Mike Mussina could have made the 300-win plateau if he had wanted to keep playing and had been able to avoid injuries and retain enough effectiveness to pitch another three or four seasons. Of course, the same could be said of a number of pitchers. Avoiding injuries and retaining effectiveness are challenging at age 40. Pettitte probably faces a similar fate, unless he kicks into gear an early-40s determination to make the Hall of Fame.
Kevin Appier and John Smoltz are on the graph as examples of the head of a large class of pitchers: guys who at one time were near a 300-win pace and were pitching at the high level of quality necessary to sustain that success. Then injuries or role changes came along, and neither ended up sniffing 300 wins, though Smoltz compiled some nice numbers in three-and-a-half seasons as a closer.
Let me show you the rest of the 300-game winners since World War II.
I’m not going to spend much time discussing these pitchers, though it becomes clear that while there are several paths to 300 wins, all of them obviously involve winning a lot of games over some large part of a career.
Today’s leading candidates
Instead, let’s turn to some of today’s promising pitchers. Here I’ve shown their career win paths as compared to the paths of the last batch of four 300-game winners.
Halladay, with the highest win total among active pitchers younger than Pettitte and the probable 2010 National Cy Young Award winner, gets a lot of attention as the next likely 300-game winner. However, we can see here that’s he behind the pace of Maddux, Clemens, and Glavine. Perhaps Halladay will follow the Gaylord Perry late-bloomer path to 300 wins, but right now he appears to be fighting an uphill battle to get there.
Sabathia is another favorite. He has nosed ahead of the Maddux and Clemens pace after posting 21 wins in his age-29 season. Sabathia is only 12 wins behind Halladay and is a little over two years younger. Of course, as we saw with Gooden, a strong early pace is no guarantee. We’ll look at this in a bit more depth in a moment.
There are several other contenders on a similar pace to Halladay. The pack is led by Roy Oswalt, Mark Buehrle, Barry Zito, and Jon Garland. Though individually none of them seem to have a good shot at 300 wins, collectively perhaps one of them might break through with a power run through his 30s. Buehrle is actually a little ahead of Halladay’s pace.
I was surprised to see Garland’s name there. He’s been roundly criticized in sabermetric circles as a pitcher with inflated win totals masking mediocre run prevention skills predicated on poor strikeout rates. He’s a very unlikely candidate to make it another 10 or 15 years in the major leagues because of that, but an early start and consistent innings have him in the discussion at this point.
Are pitchers today further from 300 wins?
So what’s a good way to winnow candidates for 300 wins from pitchers who are a decade or more away from accomplishing the feat? Jay had his Jaffe Blind Optimism method. I’ve concocted something similar: the Glavine Line.
It occurred to me that if you throw out the late bloomers like Phil Niekro and Johnson, whom we’re going to have a tough time assessing accurately at age 30, Glavine’s path represented the minimum standard for a pitcher to make 300 wins. He started with two wins at age 21 and reached 303 wins at age 41. I approximated his career with a straight-line 15.5 wins per year from age 22 to age 41, and it turns out that Glavine was mostly within 5-10 wins of that line during his career. It’s far from a perfect assessment of a pitcher’s chances, but it gives us an idea of who is in the ballpark and how far ahead or behind a 300-win pace they might be.
Here are the active pitchers with at least 50 wins and their standing relative to the Glavine Line. I used integer season age as of July 1; you could fine-tune this method by using ages with fractional years.
It looks from this view like Sabathia and Hernandez are the only ones with a good shot at 300 wins. There may be some truth to that, but it’s not the whole truth. It’s instructive to look a similar table from 1996, when the latest crop of 300-game winners was establishing its bona fides.
This list is interesting for something besides the fact that it contains three of my favorite Royals pitchers. Three of the four eventual 300-game winners are in the top six spots. Johnson is missing from the top 20, all the way down at No. 52, 67 wins below the Glavine line. To put that in perspective, with 104 wins at age 32, Vicente Padilla today is tied with Johnson’s win total at the same age in 1996.
We’ve already discussed Gooden and Pedro, but Alex Fernandez and Ramon Martinez are another couple pitchers whose stars faded quickly.
One thing that sticks out to me, though, is that pitchers today don’t seem to be winning that many fewer games than pitchers 14 years ago. I don’t see a striking change in the career win totals of pitchers in their late 20s and early 30s. The top 20 pitchers in 1996 were, on average, 10 games behind the Glavine line. The top 20 pitchers in 2010 are, on average, 11 games behind the Glavine line. The odds may be marginally steeper today than they were for Maddux, Clemens, Glavine, and Johnson, but I don’t see signs of a seismic shift.
Post-war Glavine Line leaders
However, if you go back to the 1970s, there were quite a few pitchers who looked like good candidates for 300 wins, among them Catfish Hunter, Jim Palmer, and Bert Blyleven. Though very few of them panned out in the end, the post-war Glavine Line leader board is well stocked with pitchers from that decade.
For the preceding table, I included every pitcher since 1945 between the ages of 24 and 33.
We learn at least two things. First, the Glavine Line is not a very useful tool for predicting future 300-game winners. Only two of the top 20 went on to win 300 games—Maddux and Seaver. One, Sabathia, is still active. Five of them failed to reach even 200 wins. That’s not to say that other eventual 300-game winners were not ahead of the Glavine Line pace. Clemens was at +28 in 1992, Steve Carlton was at +21 in 1979, Don Sutton was at +21 in 1976, and so on. But how far ahead of the pace a pitcher was in and of itself was not a good predictor of eventual success.
I doubt, though, that even more sophisticated tools like the Favorite Toy or full-blown projection systems like PECOTA would do much better. I say this because of the second point, that most pitchers who look like promising candidates, even the cream of the crop, will never make it.
I came today not with a crystal ball to identify the next 300-game winner but to investigate the paths that past pitching titans and flameouts have taken in their quest for that mark. As far as I can tell, there’s no reason to think we’ve seen the last one yet. If you want to bet on Sabathia, Halladay, King Felix, or even Moyer to make it, be my guest. As long as there are pitchers like Niekro and Johnson to defy the odds, I’ll be happy to bet on the field and waiting to cheer our next unlikely champion.