On the Rumor Forum this morning, someone asked: “[John Smoltz] currently has 167 career wins and 154 saves. Does the split roles affect him one way or the other? Does he need a magical number like 200 wins?”

I replied that I hated team-dependent counting stats; the discussion went forth from there, so I broke out some statistics:

Sorry, I meant by position, and honestly, I consider him a starter, his dalliance as a reliever notwithstanding. I’ll give him bonus points in consideration of his work there, but again, I have a strong distaste for stats like wins and saves because they’re oh-so-very team-dependent. I’d personally seek to remove those effects from an analysis of greatness wherever possible.

I’d look at things like BB/9, K/9, BB/K, HR/9, and *ERA+ [1]; the pitcher controls the first three in large part, and *ERA+ is a pretty decent statistic even if there’s some team dependence in it. Used in combination with more independent statistics, it’d be a good tiebreaker.

When I think of Smoltz’s era, I think of his contemporaries, who I’ll name off the top of my head. If I’m missing someone, throw a name out there—I’ll take my lumps and then run the same kind of analysis on them.

John Smoltz’s Contemporaries:
Roger Clemens
Randy Johnson
Greg Maddux
Tom Glavine
Pedro Martinez
Curt Schilling
Mike Mussina
Kevin Brown

All stats pulled from Baseball-Reference [links given above].

Player: BB/9, K/9, BB/K, HR/9, *ERA+
Smoltz: 2.76, 8.00, 2.90, 0.72, 125
Clemens: 2.92, 8.65, 2.96, 0.67, 141
Johnson: 3.48, 11.12, 3.20, 0.80, 144 [If you'd told me before I started that Johnson had a better career K/BB and *ERA+ than Rajah, I would've laughed.]
Maddux: 1.87, 6.28, 3.36, 0.58, 141
Glavine: 3.07, 5.40, 1.76, 0.69, 121
Martinez: 2.41, 10.40, 4.32, 0.69, 167
Schilling: 2.04, 8.79, 4.31, 0.92, 131
Mussina: 2.02, 7.17, 3.54, 0.95, 127
Brown: 2.49, 6.36, 2.55, 0.57, 130

I’d rank them in decreasing order of brilliance: Clemens, Maddux, Johnson, Martinez, Schilling, Smoltz, Brown, Mussina, Glavine. I could be urged to re-order the Smoltz-Brown-Mussina cohort almost at will, but I’d probably consistently put Moose at the tail end of that group. However, I have a sizable gap from Schilling to Smoltz, one that his fine work as a reliever doesn’t cover. I just don’t value relievers all that highly because they just don’t throw that many innings.

Maddux beats out Johnson because of superior walk and homer rates; you can argue that Johnson’s better and I won’t get too bent out of shape, but I think Maddux has also been better over the course of his career. No doubt that Johnson had a higher peak in overall dominance, even as good as Maddux was at his. Greg’s peak is probably only third-best in this group, as you have to look at Pedro’s peak as well.

I threw Tommy Glavine into this mix for two reasons: his association with Maddux and Smoltz, and the fact that he’s nearing 300 wins. I kinda wanted to hammer home my point on counting stats—remove the wins from the discussion table, and the argument for Glavine over his peers really wilts.

I’m re-posting this here for discussion fodder and because I spent enough time running numbers that I need to re-purpose it to feel justified in the time spent.