Is Mariano Rivera a Hall of Famer?

If you’ve been following THT Live this week, you know that I’ve been waffling back and forth with regard to Mariano Rivera‘s Hall of Fame-worthiness. On one hand, he’s only got 128 career Win Shares, and probably won’t crack the 200-mark. Only five pitchers have made it into the Hall with fewer than 200 Win Shares – Sandy Koufax, Dizzy Dean, Rollie Fingers, Lefty Gomez, and Addie Joss – and some of those guys are among the more questionable Hall of Fame selections.

On the other hand, Rivera has been the best closer of his generation, and he’s been particularly lights-out in the postseason. On Tuesday, David Pinto of Baseball Musings said, in discussing the Hall of Fame candidacies of Goose Gossage and Rivera, “Yes, Gossage belongs in the Hall. And if you define a Hall candidate as greatest at a position over a sustained period of time, Rivera will belong also.”

Wait a second. Before we define a Hall candidate, let’s first define “position.” There are nine of them, and Mariano Rivera’s is “pitcher.” He’s not “closer”; that’s not a position, it’s a role. His position is pitcher, the same as Pedro Martinez and Greg Maddux. This significantly widens the field of competition. I’m pretty confident that Rivera is the best closer of his generation, but I’m just as confident that he’s not the best pitcher.

The classic argument, of course, goes like this:

Pro: “Rivera’s been lights-out dominant.”
Con: “But he only pitches 70 innings a year.”
Pro: “But those 70 innings have been high-leverage.”

Thank goodness, analyst-extraordinaire Tangotiger from Baseball Primer has come up with a brilliant tool called Leverage Index, which adjusts Rivera’s innings based on the fact that they come in high-leverage situations. An average inning has a Leverage Index of 1.00; a Leverage Index of 2.00 means that the inning’s “importance” is double that of an average inning.

Mariano Rivera’s Leverage Index from 1999-2002 (the only years available right now) is 1.96 – the fourth-highest figure in baseball during that time. What do we do with this? Well, simply multiply Rivera’s career stats by 1.96, and you get his “leveraged stats.” In other words, stuff we can compare to starting pitchers. Here are the leveraged stats for Mariano Rivera:

    IP        H      BB       SO      ERA
1273.1     1009     347     1141     2.49

Since his Leverage Index is 1.96, Rivera’s counting stats all increase by 96%. Still, that gives him less than 1,300 innings pitched, which is too low for a Hall of Famer.

Ah, but that’s before taking into account Rivera’s postseason dominance. “Sandman” has pitched 96 postseason innings, with a remarkable 0.75 ERA.

Considering that a) manager Joe Torre tends to use Rivera in even higher-leverage situations in the postseason, and b) every postseason inning is high-leverage, let’s assume that Rivera’s Leverage Index for the postseason is 50% higher than that of the regular season. That is, since his regular season innings are about 96% higher-leverage than average, his postseason innings are 144% higher, for a postseason Leverage Index of 2.44.

Yes, that’s an arbitrary figure and could be higher, but it’ll do for our purposes. Remember, while postseason innings might be much higher-leverage than their regular season counterparts, Rivera has benefited from opportunity – it’s not really Trevor Hoffman‘s fault the Padres haven’t made the playoffs nearly as much as the Yankees. This adjustment turns Rivera’s 96 postseason innings into 234, with the same 0.75 ERA. Add the postseason numbers to his leveraged career stats, and we get this:

    IP        H      BB       SO      ERA
1507.2     1156     376     1328     2.22

Well, that’s pretty good — about as good as Pedro Martinez has been since 1997 (1,408 IP, 2.20 ERA). Thing is, Pedro pitched before 1997 and has almost 2,100 IP in his career. One way to bridge the innings gap is to find the “residue ERA” – the ERA that Adjusted Mariano would have to put up to equal Pedro’s innings and ERA. The difference between Pedro and Adjusted Mariano is 571 innings and 225 earned runs, for an ERA of 3.54. Take a look at the residue ERA Rivera would need to match some of his contemporaries:

G. Maddux         3.30
P. Martinez       3.54
R. Clemens        3.72
R. Johnson        3.93
K. Brown          4.08
T. Glavine        4.33
J. Smoltz         4.74
D. Cone           4.80
C. Schilling      4.88
M. Mussina        5.23
K. Appier         5.81
D. Wells          6.15
J. Moyer          6.33
A. Pettitte      13.03

What all that means is it would take Adjusted Mariano, plus a 2,461-inning, 3.30 ERA pitcher (say, Allie Reynolds) to equal Greg Maddux. On the other hand, to equal Jamie Moyer‘s career stats, only a replacement-level pitcher would be required. What this suggests is that Mariano Rivera is roughly equal to Mike Mussina; that is, he’s good, but he needs a few more good years before he’s truly Hall-worthy.

While we’re on the subject, let’s look at the other great modern closer, Trevor Hoffman. His Leverage Index from 1999-2002 is 2.08, second only to Troy Percival. Adjusting his stats, we get the following career line for Hoffman:

  IP        H      BB       SO      ERA
1477     1108     451     1681     2.78

That’s not quite enough for the Hall, if you ask me. His postseason stats don’t really help (4.09 ERA in 11 unadjusted innings). Of course, Hoffman is still pitching and will add to his totals, but he’s also coming off two shoulder surgeries.

In my opinion, Mariano Rivera will be inducted into the Hall of Fame. Even if he retired today and went in, he wouldn’t be a disgrace to the Hall by any means. But, while he might be the best closer since Eck, he’s only been about the tenth-best pitcher of this era. He’s still playing and adding to his numbers, but right now, Mariano Rivera is just a borderline Hall of Famer.

Using Recurrent Neural Networks to Predict Player Performance
Technology is rapidly advancing possibilities in decision-making.

Print This Post

Comments are closed.