September 29, 2010

Saves Are Dumb

Peter Abraham, Globe:
Closer Jonathan Papelbon blew eight saves, a career worst. Outside of Bard, the bullpen was a liability for much of the season.
Papelbon was a huge disappointment this year. I don't think any Red Sox fan would disagree with that.

And Daniel Bard has certainly been fantastic -- he leaves Bot in the dust in both WAR and WPA -- but let's not forget that he has blown seven saves -- in only 10 chances*. At that rate, Papelbon would have 32 blown saves this year. (He doesn't even have that many in his six-season career.)

* Bard seems to be tiring. He has pitched 12 times in September and has allowed 12 hits and 8 walks in 12 innings; he has allowed at least one hit in 11 of those 12 games. However, only two of his blown saves have come in the last six weeks.

In Papelbon's amazing 2006 season, when he had an 0.92 ERA and a 0.776 WHIP, he blew six of 41 save chances for a 86% conversion rate. His rate in 2010 is not much worse: eight out of 45 (82% success).

The point is not that Papelbon has actually been as good as usual or that his blown saves have not cost the team a handful of sorely-needed victories. It's this: saves are dumb, a useless counting statistic that tells us nothing about what happened on the field.

You can get a save for (i) retiring the side in order in a 1-0 win or (ii) if you are up by three runs, allowing two runs and leaving the bases loaded. And there are blown saves of all types, from flushing a generous three-run lead (coughjos1cough) to recording an out that scores an inherited runner from third to tie the game (as Rivera did in 2004 ALCS 5)*.

* That game was something else. Fuckin' A! My favourite game of the 2004 post-season, I think.

I don't give a shit whether Papelbon finishes the season with 10 saves or 50 saves. As long as he keeps guys off base and doesn't allow inherited runners to score (and dials down his ridiculous post-game fist pumping, which I think he has done, toned it down, but maybe there simply have not been a whole lot of those types of opportunities lately), I will be happy. There are statistics that keep track of those things, allowing us to more accurately assess his performance.

There is no reason to ever consider saves or blown saves. If we did, Bard's 2010 season would rank as one of the worst of all-time. There are 45 relief pitchers in MLB with at least 10 save opportunities this year. In save percentage, Bard's 30% puts him #44 out of those 45. Only Washington's Tyler Clippard is worse (having converted only 9%; he's blown 10 of 11). Most everyone on the list is over 75%, with 20 of the 45 over 90%.

Abraham can point to Papelbon's eight blown saves as evidence of a poorly-performing pen, but to then say that, in contrast, Bard has been the only (?) non-liability in the pen, while ignoring his seven blown saves is disingenuous, at best. If you are going to concentrate on blown saves, Bard has been just about the worst pitcher in baseball.


FenFan said...

Totally agree with your assessment. I do believe that there is value in having a closer but here are three completely different situations that earn a relief pitcher a save:

1. You come into a game with two outs in the ninth, the bases loaded, and your team holding a six-run lead. You give up five runs before the final out is made and your team wins the game.

2. You come in at the start of the seventh with your team holding a one-run lead. You pitch the final three innings to preserve that lead and give your team the win.

3. You come in with two outs in the eight and a one-run lead, make the final out, then watch your team score ten runs in the next half-inning. Then you come back out in the ninth and get the final three outs; in the process, you allow ten runs to score but your team wins by one run.

I could go on, obviously... but those all seem like VERY different situations. Is one more or less deserving than the others? Doesn't matter because, in each situation, the closer earns the save.

Amy said...

Great post. Of course, your last post showed that he DID allow a lot of inherited runners to score and put more runners on base than in past years. So he did suck by that standard as well.

allan said...

did i look at inherited runners?

Amy said...

Here's the stat you posted in inherited runners:

He has come into a game with runners on base 9 times. He has allowed 6 of 15 inherited runners to score.

I don't know how many games this amounted to, but allowing 40% of the inherited runners to score seems high to me.

allan said...

i can't remember what i write!

that does seem higher than it should