Learning from history: Revisiting the mirage of the "proven closer"

I'm tempted just to insert deadhorse.png and leave it at that. Because there are again ongoing rumblings that the Arizona Diamondbacks are looking to trade for a closer - perhaps Andrew Miller of the Yankees, perhaps Aroldis Chapman from the Reds, whom Peter Gammons is hearing will be dealt somewhere by Sunday night. Let's hope neither of these deals come to fruition for us, because dealing for a "proven closer" is, more often than not, something that fails to work out.

We looked at this back in early 2014, when the bloom of Addison Reed was relatively fresh. We took the previous three seasons, 2011-13, and looked at the top twenty closers (more or less those who posted thirty-plus saves) in 2011. Just one year later, barely one-third were still in the top twenty for that season. Only 10% - Jonathan Papelbon and Craig Kimbrel - were in the top twenty for all three years. I'm pretty sure a similar study starting in 2013 would show similar results, e.g Jim Johnson who had 50 that season, and two the next. Others in the top ten who were barely visible this year: Grant Balfour, Rafael Soriano and Joe Nathan, and of course, Mr. Reed.

We don't even need to look outside the team's own history to see the folly of trading for a proven closer. Here are the top ten closers for the Diamondbacks by number of saves, sorted by their ERA. I'll get to what the final column indicates shortly.

Any guesses what the last figure on each line is? It's the number of major-league saves the player had, the last full year before joining Arizona [a couple, such as Mantei, were dealt mid-season]. Notice anything? The worst pitcher on the list is the sole example who could be called a "proven closer" at the time he was acquired. with twice as many saves as the other nine combined. Free agents Putz and Olson are the only others with even 20 career saves at the time of their arrival. Putz hadn't closed for over two years when he signed with us, and Olson was more than four years removed from the role when he came to Arizona and became our first closer.

The fact is, the team has done just fine, operating without a "proven closer" doing the job for 90% of its existence - including all the years when we won the division. The shiny save statistic inevitably leads to an inflated value, especially considering the evidence suggests "saves" is, in most cases, not a repeatable skill. Everyone dreams of getting the next Mariano Rivera or Craig Kimbrel, but for every one of those, there are ten Jim Johnsons. Chapman has lasted longer than most, posting 30+ saves four years in a row, so you can reasonably argue he is probably more reliable than many.

But his overall Save% since 2012 is 90.6%; during 2013 + 2015, when he actually closed for us, Brad Ziegler's was 91.5%. Despite this, next season, Chapman will probably cost twice as much as Ziegler, not counting the value of any prospects we have to give up to get the Cuban Missile. Now, there is certainly a case to be made for improving the bullpen. But I'd rather we did it with moves like the Mariners just made, dealing a couple of low-grade prospects to the Padres for Joaquin Benoit, who last year had an ERA+ of 156... and two saves. Getting good pitchers, not adapting The Hunting of the Snark to a baseball theme, seems to me to make eminently more sense.