Ask a baseball fan which of today's pitchers will go down as among the very best ever, and you'll generally hear three names in response: Roger Clemens, Greg Maddux, and Pedro Martinez. Those three are usually put on a pedestal above everyone else, with the other great pitchers of the day -- Tom Glavine, Kevin Brown, David Cone, and others -- are relegated to the second tier.
What about Randy Johnson? Certainly everyone knows that Johnson is a great pitcher, given his three Cy Young awards and two ERA crowns. His strikeout prowess and intimidating presence have assured him plenty of media attention over the years. But when it comes to discussions of all-time greatness, my observation is that Johnson is usually lumped in with Glavine, et al, in the second tier rather than with the "Big Three" of Clemens, Maddux, and Martinez.
That was probably the right assessment of Johnson's standing a couple of years ago, but not any more. A close look at the numbers reveals that while the Big Unit probably isn't in Clemens/Maddux territory yet, he's putting more and more distance between himself and the second tier, and has reached the point where he's comparable to some of the elite pitchers in the history of the game.
We'll look at Johnson's standing in a couple of ways. The first is the rate at which he allows runs. ERA is the usual way of measuring this, but we'll make a few adjustments. First, since no one has ever demonstrated that "unearned" runs are any less due to pitching than "earned" runs, we'll use all runs allowed (RA instead of ERA). Second, since some parks are more pitcher-friendly than others, we'll adjust the pitchers' run totals according to the parks in which they pitched. Third, we'll express a pitcher's park-adjusted RA as a percentage of the league average. A pitcher's career percentage is the average of his single-season percentages weighted by the number of innings he pitched in each season.
Here are the top 10 active pitchers in career RA as a percentage of league RA, through Wednesday's games (minimum 1500 IP):
Pitcher %LgRA
Pedro Martinez 60
Roger Clemens 70
Greg Maddux 71
Randy Johnson 75
Mike Mussina 75
Curt Schilling 78
Kevin Appier 80
Kevin Brown 81
David Cone 83
Tom Glavine 83
Martinez is so good here it's scary. The closest all-time career %LgRA to Martinez's 60 is Lefty Grove, with 69.
Aside from Pedro's dominance, the thing to notice here is that Johnson is ahead of everyone on the list except the aforementioned Big Three. The pitchers who are closest behind Johnson on the list -- Mike Mussina, Curt Schilling, and Kevin Appier -- have all pitched significantly fewer innings than Johnson has. Among the pitchers with 2,500 or more innings pitched, Johnson is closer to Maddux and Clemens than he is to Brown, Cone, and Glavine.
Of course, a great rate of run prevention by itself doesn't make a pitcher's career great. If it did, then Tim Jones (career totals: 10 innings, no runs allowed) would blow all these guys away. A better measure of career greatness combines a pitcher's rate of run prevention with how long he maintained that rate. That will let us attach some value to Johnson's IP advantage over Martinez and Mussina (as well as Clemens' and Maddux's IP advantages over Johnson).
We'll combine rate with longevity by calculating how many runs the pitcher prevented over what a replacement-level pitcher would have allowed in the same number of innings. Here a replacement level pitcher will be defined as one who gives up 20 percent more runs than average. We'll then convert the resulting runs total to wins by using Clay Davenport's Pythagenport formula. We'll give the resulting measure the generic name "Wins Above Replacement" (WAR). Here are the active leaders:
Pitcher WAR
Roger Clemens 101
Greg Maddux 87
Randy Johnson 63
Tom Glavine 57
Kevin Brown 57
David Cone 54
Pedro Martinez 54
Chuck Finley 52
Mike Mussina 51
Kevin Appier 47
Johnson is unlikely to ever catch Maddux or Clemens in this measure, especially since Maddux and Clemens are still adding to their totals at a pretty good clip. On the other hand, Johnson has put some distance between himself and the Glavine/Brown/Cone cluster over the past few seasons, and that gap shows every sign of continuing to grow in the short term.
Whenever you do career evaluations of active players, there's always the caveat that the numbers are going to change, and some are more likely to change significantly than others. For example, no one would expect Pedro Martinez to finish his career with a WAR rating anywhere close to 54, given that he's just 29 and currently putting up mind-boggling numbers. However, David Cone, at age 38 and currently struggling, looks like a long shot to see his career WAR grow very much.
At first glance, it might seem that Johnson, age 37, is another pitcher who will have a hard time adding much to his career value. But if there was ever a pitcher for whom you should throw out the rules about aging, it's Johnson. After all, Johnson had his first really good season at age 29, his first great season at age 31, and his best season to date (by WAR) at age 35. He won't be able to throw 95 mph forever, but given his age-defying track record and the general tendency of power pitchers to have longer careers, I wouldn't look for a collapse anytime soon.
Perhaps the best way to make the case that Johnson is nearing all-time elite status is to compare him to one of the widely acknowledged all-time elite pitchers, Sandy Koufax. Here are the career totals of the two pitchers, using the measures we introduced above:
Pitcher IP %LgRA WAR
Koufax 2,324.1 74 55
Johnson 2,590.1 75 63
When you adjust for league scoring levels and park effects, Johnson just about matches Koufax's rate of run prevention. And Johnson has a higher number of Wins Above Replacement, by virtue of his advantage in innings pitched.
But maybe we should compare the two in a different way. After all, Koufax is generally celebrated not for his career in its entirety (which included some so-so seasons in Brooklyn at the start), but for the great string of seasons he put together in Los Angeles at the end of it. So let's look at each player's best five seasons by WAR:
Koufax
Year IP %LgRA WAR
1966 323 55 10.6
1963 311 56 9.8
1965 335.2 65 9.3
1964 223 53 7.5
1961 255.2 83 4.9
Johnson
Year IP %LgRA WAR
1999 271.2 55 9.5
1995 214.1 52 8.0
1997 213 51 7.9
2000 248.2 64 7.7
1993 244.1 76 5.8
Koufax has slightly higher WAR totals for his best three seasons. That's as much an artifact of the four-man rotation and the resulting extra innings pitched as anything else, but those extra innings still have value. On the other hand, if you look at the rates of run prevention in their best seasons (%LgRA), Johnson is equal to or even a little better than Koufax, indicating that he matched Koufax in quality if not quite quantity.
Any way you want to slice it, Johnson's career to date looks remarkably similar to Koufax's. You can interpret that one of two ways. Either Johnson's career has been more valuable than commonly thought, or Koufax's career wasn't quite as valuable as commonly thought. I think there's probably some truth in both statements. But anyone will acknowledge that Koufax was a remarkable pitcher during his short career, and the fact that Johnson's career has essentially paralleled Koufax's gives you an idea of the outstanding company he's keeping.
But back to our original question. Which class of today's pitchers does Johnson belong to, the Big Three or the Second Tier? I think the numbers show that Johnson's career currently hasn't reached Clemens' or Maddux's level, but it probably does him a minor disservice to group him with Glavine, Brown, and Cone (good as those three pitchers are). In other words, the numbers show something you probably already knew before you started to read this column: The Big Unit is in a class by himself.
The team of writers from the Baseball Prospectus (tm) will be writing twice a week for ESPN.com during the baseball season. You can check out more of their work at their web site at baseballprospectus.com. Michael Wolverton can be reached at mwolverton@baseballprospectus.com
| |
|
