<
>

Clemens vs. Maddux

7/27/2001

He's one of the greatest pitchers of all time, a first-ballot Hall of Famer with an excellent chance to win 300 games.

He's one of the greatest pitchers of all time, a first-ballot Hall of Famer with an excellent chance to win 300 games.

But which one is better?

I'm writing, of course, about the Twin Towers of modern pitching, Roger Clemens and Greg Maddux. The two have combined for nine Cy Young Awards, and at 38 and 35 years old respectively, are among the front-runners to pick up more hardware this year:
W-L IP ERA H BB SO SNVA
Maddux 13-5 152.0 2.43 138 18 127 3.3
Clemens 14-1 142.2 3.72 138 43 139 2.3

(SNVA: Support-Neutral Value Added, a measure that isolates a pitcher's performance from league and park factors, as well as bullpen and run support, expressed in wins over an average pitcher.)

At ages when many of their peers are coaching in Double-A or calling games for local radio, Maddux and Clemens are still holding their own with the best in baseball. That alone is a sign of greatness: inner-circle Hall of Famers are mostly those who have long careers with extended declines from their peak, something both players have done. Let's take a look at their career numbers:
W-L Pct. IP CG ERA H BB SO TPI ERA+
Maddux 253-140 .644 3470.0 102 2.81 3124 751 2477 60.0 145
Clemens 274-143 .657 3809.1 116 3.10 3239 1229 3643 57.5 146

(TPI: Total Baseball's Total Pitcher Index, TB's tool for rating pitcher performance. ERA+ is ERA, normalized for park and league, also from TB. It shows Maddux as 45 percent above the league and Clemens at 46 percent. Both these measures are through 2000.)

In terms of raw numbers, Clemens holds a small edge, primarily due to his three-year age advantage over Maddux. That's the basis for his lead in wins and innings, and adds to his massive edge in strikeouts. Maddux leads in ERA, but the difference of less than three-tenths of a run is almost entirely swallowed up by Clemens pitching his entire career in the DH league, and Maddux in the non-DH league. The Total Baseball tools call the players close to even, rating Maddux's very best seasons as a bit better than Clemens' very best.

What jumps out at me in Maddux's line is the strikeouts. While Maddux's shot at 300 wins is well known, how many people realize he has a good chance to reach 3,000 strikeouts, and should get there late in 2005? For a pitcher who was never regarded as a strikeout guy -- despite good rates throughout his career -- that's a significant achievement.

We've seen that Clemens has an edge in the traditional stats. How do the pitchers look using some of the other ways of comparing great pitchers?

Clemens Maddux
Black Ink 87 69
Gray Ink 252 244
HOF Monitor 225 196.5
HOF Standards 59 55
Fibonacci
Win Points 311 276
Cy Young Awards 5 4
MVPs 1 0

(Check out Baseball-Reference.com for data and explanations.)

Roger Clemens has edges, albeit small ones, across the board. This is in part a function of his extra seasons and advanced age. Still, those extra seasons have value, and his advantages in both raw stats and career-evaluation measures are a strong indication he has been the better pitcher.

I admit that I had a bias coming into this article. While I appreciate Clemens's talent and his body of work, I don't hold him in the same regard as Maddux. That's not a knock on him; I have just enjoyed watching Maddux pitch in a way that I haven't enjoyed Clemens. And based on his insane peak in 1994-95, when he had ERAs of 1.56 and 1.63, I'd felt that Maddux was the better pitcher of the two by a small margin.

That's wrong. Clemens has been the greater pitcher to date, although again, the difference between the two is small and not insurmountable.

Part of the reason for my misperception is that Maddux concentrated his best seasons into one extended run, while Clemens's best years have been distributed throughout his 17-year career. From 1992 through 1995, Maddux had a four-year peak that matches up with that of any pitcher in baseball history:

W-L Pct. IP ERA CG ShO H BB SO TPI
'92-95 75-29 .721 946.2 1.98 37 11 726 176 733 26.2

Four of Maddux's best five seasons are included in this run, with only his 1998 (187 ERA+, 7.1 TPI) being left aside. He probably should have been the NL MVP in 1995, finishing third behind Barry Larkin and Dante Bichette in a very splintered vote.

The raw numbers don't tell the entire story of this peak. Maddux lost about 15 starts in 1994 and 1995 to the players' strike. Had he had those starts, he likely would have won 20 games both seasons; given the offensive context, his 1994 season is one of the greatest of all time, but suffers for the shortened season in comparison to other great performances. Clemens, on the other hand, had his best years in non-shortened seasons, and while 1994 was a good year, losing a third of the season doesn't have any impact on how we would view him.

Clemens' peak is harder to define. His Cy Young Awards have come in 1986, 1987, 1991, 1997 and 1998 -- which in itself is a sign of greatness. It's not like Clemens was ineffective in those other seasons, either: he led in AL in ERA in 1990 (when he deserved the Cy Young Award over Bob Welch, anyway) and 1992, in strikeouts in 1988 and 1998, and spent most of the period among the league leaders in innings pitched.

Because Clemens' peak is harder to define, we can compare the two pitchers by lining up their careers side by side, ranking their seasons from best to worst. The following table lists each pitcher's adjusted ERAs (ERA+) for their careers:

Clemens Maddux
225 272
212 262
177 191
176 187
175 170
168 165
164 162
153 151
141 128
140 126
131 118
130 116
130 114
117 76
104 73
103
97

As this table shows, Maddux's peak seasons have been better than those of Clemens. But Clemens has more seasons of non-peak, effective pitching than Maddux, and has been better, season-for-season, than Maddux in those years.

Both pitchers are great, among the six or seven greatest ever, but at this point, Clemens retains a small edge in career value.

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. Joe Sheehan can be reached at jsheehan@baseballprospectus.com