SNY.tv: Online Home of All Things Sports

Goose Gossage is a Hall of Famer now and not shy about saying that today's closers have it easy. So what better time to stack up great Yankee closers of the past and present? Who's better: Gossage or Mariano Rivera?

Most sabermetricians think it's a slam dunk for Rivera on the basis of Rivera's ERA relative to the league average and Gossage's performance in that same stat.

Rivera consistently has an ERA that's less than half the league ERA and sometimes, adjusted for his ballpark, nearly a third. Gossage did have one great strike-shortened year when his ERA was 0.77 compared to the league average of 3.55. But his best Yankees season other than that was in 1978, when his ERA wasn't even half the league average -- a feat Rivera has met or exceeded many times.

Before we get much further, I want to make clear that I saw Gossage pitch at his peak and my recollection is that Gossage was Rivera's equal. But I realize that memory is far from a perfect representation of the past. Numbers don't lie when used properly. Let's open the toolbox and see if we can fairly assess players from vastly different eras who were used very differently.

Let's start with ERA+, the ratio of a league's ERA to that of a pitcher. I believe that it's much easier to post a higher (better) ERA+ number today when runs are more prevalent than it was when Gossage dominated in the 1970s. As proof, I'll note that only one qualifying starting pitcher in the 1970s sported an ERA (adjusted for home ballpark) at least half of the league ERA -- Ron Guidry in 1978. Conversely, since the offensive explosion in 1994, qualifying starters have accomplished this 10 times.

I can't argue whether the top pitchers since 1994 were just better than the top pitchers of the 1970s. But, I think it's reasonable to say that they shouldn't be that much better than their contemporaries. So, ERA+ doesn't seem like a fair barometer to use in measuring Gossage and Rivera.

It is fair to look at peak seasons and say simply that Rivera had a lower ERA than Gossage despite pitching in an environment where more runs were scored. For example, Gossage in 1978 had a 2.01 ERA. Rivera has beaten that seven times. And in his best year, 2005, he had a 1.38 ERA.

What's most different about these two eras is homers and strikeouts. In the AL those years (same number of teams), homers are 50 percent higher in 2005 than in 1978 and strikeouts are 36 percent higher. Walks actually were six percent higher in 1978, which surprised me.

So what can we do with that. I'm ballparking here, not trying to write a SABR paper -- so cut me some slack in your e-mails. The 1978 Gossage gets 44 more Ks, giving him 166 in 134.3 innings. He walks five less guys and you can reasonably take off all of his eight intentional walks (Rivera had none in 2005). Given that the league hit about .271 off him on balls in play (excluding homers), those extra Ks mean 12 less hits. Gossage's ration of hits to homers means we knock one or two homers off his total of nine allowed, but then we have to bump that up 50 percent for 2005 and give him about 12 homers allowed. When we crank all this together using Bill James's component ERA formula, Gossage gets and ERA of about 1.91 if we were able to magically transplant him in the 2005 environment (again, Rivera's Sgt. Pepper's season).

Now here is where things get really interesting. Gossage pitched those 134.3 innings in 1978. Rivera, in 2005, pitched 78.3. I'm quite confident that Gossage would have been at least as valuable as Rivera pitching slightly worse but for almost 60 more innings.

Aside from the obvious fact that it's harder to pitch more innings (which is why the best relievers typically have lower ERAs each year than the best starters), Gossage had I think a tougher job.

For example, in 1978, Gossage faced the same hitter more than once in about 25 percent of plate appearances against him. Rivera in 2005 faced the same hitter twice two percent of the time. (All stats cited now are 1978 for Gossage, 2005 for Rivera.)

Gossage averaged over two innings per appearance compared to 1.1 innings per appearance for Rivera.

Gossage faced 139 batters in the seventh inning or earlier compared to 182 batters in the ninth inning. Rivera faced no batters in the seventh inning or earlier and just 17 in the eighth compared to 257 in the ninth.

I believe it's more stressful to pitch in a tie game than with the lead. Gossage faced 176 batters in tie games compared to 53 for Rivera.

By looking at these two players in these two years, you get a strong sense of how bullpen usage has evolved over 30 years. I believe Gossage was leveraged better by Billy Martin then than Rivera (and all other closers) are used today.

But that comes at a price, of course. Gossage had five years as a reliever where he pitched 99 or more innings, going over 130 innings twice (back-to-back in '77 and '78). Rivera has one year over 100 innings, his first year as a reliever in 1996. Since then, he's never logged more than 80.7 innings.

I also believe that Rivera was never more dominant than in 1996, when he was used in a manner more similar to how Gossage was used in the 1970s. Note that Rivera's ERA that year was 2.09, just the eighth-best of his career. I'd bet that's strictly a function of pitching more innings.

Gossage is a Hall of Famer now. Rivera will follow him one day. Both, at their peak, were dominant. Contextualizing their stats not just for their own era but for the other leads me to conclude that their peak levels of greatness were essentially equal.