How often closers lose their jobsMay 25th, 2011 by Derek Carty in General Guidance, Theoretical
It would be an understatement to say that I've struggled with drafting competent closers in the CardRunners Experts League (a fact that Eric has little problem reminding me of on a semi-regular basis). Last year, I drafted Frank Francisco, who lost his job two weeks into the season and, despite being one of the top non-closers in the game the rest of the way, never reclaimed the role. This year, I drafted Matt Thornton and Francisco again. Thornton was arguably the best non-closer in baseball before finally getting a crack at closing this year, but he had a rough (arguably unlucky) start to the season and lost the job in April as well. Francisco managed to get injured after our draft and finally took the role back a couple weeks ago, but after a rough few days may be on thin ice again.
Given my storied unsuccess drafting closers, Eric suggested that I run some studies on closers and examine just what kind of return on investment one can expect from a closer. Over the next few weeks, I'll be digging into the data and answering a number of questions about closers that should prove extremely useful for both myself in figuring out where I keep going wrong and for the population at large in their own closer decisions.
Throughout my series on closers, I'll be looking at all pitchers since 2001 who started the year as his team's closer and grouping them into the following categories:
• Sole closer (CL)
• Closer who will begin the year on the DL (INJ)
• Replacement closer while primary closer is DLed (TEMP)
• Closer-by-committee favorite (CCF)
• Closer-by-committee underdog (CC)
I can't tell you how difficult it was to go back ten years and try to figure out who began the year as the closer, but I think I've done a pretty good job of doing so. If you'd like to check my work, you can see the full list of classifications here. If you see anything that's wrong on there, I'd love to be corrected, though I'd guess this list is close to 95% accurate. I put it together using a combination of preseason auction prices (a huge thanks to Peter Kreutzer for these) and team-by-team April saves leaders cross-checked with Google searches for articles from the period in question.
"Sole closer" stats
To start with, we'll look at what happens to pitchers who begin the year healthy and with the closer's role to themselves. The table is broken down by the number of saves accumulated and the percentage of pitchers who do so. So for the first row of the table, we see that 41% of these closers save at least 35 games and these closers average 42 saves.
Overall, we see that if a pitcher begins the year healthy and at closer, he will save an average of 29 games. And if we use 30 saves as a proxy for keeping the job the whole year, we see that 54% of closers manage to go the whole year in the closer's role.
"Closer who will begin the year on the DL"
Our sample size drastically diminishes here as no more than a handful of closers usually begin the year on the DL. Adding to the noise in these numbers is the fact that I've made no attempt to account for injury severity, how long the closer is expected to be out for, or how good the closer is (i.e. an injured Mariano Rivera is more likely to take back his closer's role once healthy than an injured Brandon Lyon would be)
Caveats aside, we see that these closers average just 14 saves per year — half of what healthy closers do.
"Reliever who will close while primary closer begins the year on the DL" stats
Again, our sample size is tiny as we're dealing with the same exact teams as in our previous grouping.
Interestingly, these closers average 13 saves per year — just one less than the closers they're replacing. You may notice that these two groups account for just 27 of a team's total saves (teams usually average around 40 saves per year). This is likely because these "replacement closers" aren't elite relievers to begin with (if they were, they'd be full-fledged closers in the first place) and are more likely to lose the job before the real closer even comes back from the DL, which brings a third closer into the mix (or potentially more than that).
The table to the right gives the stats for the reliever who is the favorite to emerge from a committee situation.
Another tiny sample because most teams don't begin the year with a closer committee, but we see that it doesn't bode well for pitchers who are a part of them, even if they're the favorite. If they're not the favorite, their expectation doesn't change much:
For these pitchers, we see that they average a mere save less than the reliever who is the favorite to emerge from a committee situation. So no matter whether a pitcher is the favorite or not, if a reliever is a part of a committee of closers, this data suggests he'll be extremely lucky to top even five saves and will average just 7 or 8.
Making sense of everything
Based on what we've just seen, it appears that a closer who is healthy and has the role all to himself is — as we'd expect — the best bet. Still, it appears that it's just a 50/50 shot that this closer will remain healthy and effective enough to keep the job for the entire season. With an average expectation of 28 saves, though, they don't appear to be the worst investment in the world.
Injured closers, however, need to be severely discounted, on the whole. Obviously there would be some exceptions, such as if the pitcher is firmly established as the closer and is dealing with a non-serious injury that should allow him to return shortly after the start of the season. If that isn't the case, though, the replacement closer might end up being a good bargain. Closers who are a part of a committee are likely to be poor investments.