Jones for Closing, II

Well, now I've read it, and I kind of wish I hadn't. It's basically the partyline on relieving. It's the standard stuff you hear whenever anyone tries to change how relievers are used:

Boston needs to either name one guy the closer and live and die with him, or go get somebody to be the guy. Bullpen by committee might work in Montreal or Tampa, but it in ain't gonna fly in Beantown.

There are so many errors in Jones' article that it's a bit overwhelming to document them.

First, what is meant by a "closer" has been in flux on , if not a yearly, a decadal basis since free player replacement was allowed in the 1890s. The title "closer" since at least the mid-Seventies when I was a kid. Back then it meant the pitcher than closed out maybe 80-90 games pitched 130 innings and save 20-25 games. He would pitch anywhere from one to three innings at a time. Now it refers to pitcher who comes in in the ninth inning almost only in save situations. He pitches 50-60 games a year, throws about the same number of innings, and saves 30-40 games.

Sometimes closers continued to be used in save opportunities even when their statistics clearly display they are not a good choice for the closer's role. I documented many such examples in my relief pitching series. However, Jones' 1998 season is a prime example: he saved 28 games with a 4.97 ERA (5 % worse than the park-adjusted league average), a 1.48 WHIP (Walks plus Hits per Innings Pitched), and only strikeout 1.5 men per each walk he allowed. Jones was a subpar pitcher that year, but his team, the Tigers, decided in their wisdom that he should be entrusted in holding a lead late in somewhat close games. Jones should praise the current role for the closer; he has been one of the pitchers who have benefited from its use: Jones has 184 saves to go with his 3.75 career ERA and 1.42 career WHIP.

Anyway, I don't know Theo Epstein's or Grady Little's take on the bullpen system employed by the Red Sox this year. But I know Bill James'. First, he does not refer to it as a bullpen- or closer-by-committee, meaning that on any night anyone in the pen could be used in any reliever role. Indeed that does seem doomed for failure as pitcher use would be rather haphazard, and the best pitcher would not be used when most important. Not to mention the unease that it would breed in the bullpen, to quote Jones:

The bullpen works better if guys have certain roles. As the game goes on, if you have set roles and your scenario comes up, you can prepare for it much better. It's awful not knowing when you're going to pitch from night to night.

But this is not what James advocates-at least not what he advocates in his articles. James' study in the New Historical Baseball Abstract calls for the best pitcher, i.e., the closer, to be used when the game is on the line. He found that the most appropriate time for this is from the seventh inning on with the game either tied or the closer's team up by one run (also, when they are down by one run and the pitcher is well rested). He also found that using the closer for more than one inning if needs be is most effective.

For the rest of the bullpen, although James does not really go into this, a hierarchy would be used to optimize the best pitcher for the most important situation (depending on the pitcher's availability). This is not an arbitrary bullpen-by-committee, just a redefinition of the closer's role. The pitcher's all have roles, just not the "traditional" ones. The closer would be used whenever the game is on the line in close games, not just in save situations.

What this could do is redistribute some of the saves to secondary pitchers who happen to finish up a fairly close game after the "closer" pitched the inning(s) that allowed the team to win. Jones goes on to say, "The difference between the eighth and the ninth is mental. As a closer, you are where the rubber hits the road." Well, why is where the "runner hits the road" (and we all know how painful that can be) the ninth inning? The game could be on the line in the eighth and if you don't use your best pitcher, you not lose the save opportunity, you lose the game.

I'm no psychologist, but I understand the comfort that the finality of pitching in the ninth must have for a closer. That's nice for him. As to whether it requires more "mental" toughness, I can't say. All I know is that James' theories make a whole lot more sense than losing a game in the 7th or 8th inning while the best available pitcher languishes unused in he pen. It seems like a bunch of lollygaggers to me.

Now, to evaluate James' theories based on two weeks with the Red Sox personnel seems unfair to me. Epstein has said that he did not have the funds to re-sign Ugueth Urbina and that was one of the reasons that they went with this plan. Perhaps they do not have the right personnel. They are all veteran relievers, but almost to a man they are performing poorly. They have the highest reliever ERA in baseball and Baseball Prospectus ranks them as the worst pen in the game (just behind the Braves). Bobby Howry has already been sent down to the minors. Allan Embree (who's hurt), Ramiro Medoza, and Chad Fox have been awful. Then again if they all stink, does it matter who the closer is?

Maybe evaluating short relievers based on a small sample of innings per season is not the most effective means of predicting future success, or maybe they have all had a slump at the same time, or maybe they are just adjusting to a new system, or maybe Grady Little doesn't understand the system himself, or maybe James' theories don't work in practice. Time will tell. But to say a new system won't work because it has not been tried is erroneous. First, because history tells us that the use of relievers is constantly in flux. And second, because a version of James' system on steroids was in use in the Seventies and the Eighties.

The closers who first put their stamp on the role were used in the way James' prescribes just for an extra 40 or so innings per year. There were also smaller pens and fewer pitchers used per game. You never heard Bruce Sutter, Goose Gossage, Kent Tekulve, and Rollie Fingers complain about their roles back then.

Managers saw that the number of innings caused stress and cut back on the innings that the top closers pitched. James argues that the just cut the wrong innings. Managers followed Tony LaRussa's use of Dennis Eckersley in the late Eighties and tailored the closer role to the definition of a save.

According to James' definition Chad Fox was in essence the Red Sox closer to start the season. Of all the instances in which the Sox either led by 1 run or were tied in the last three innings, Fox pitched the bulk of these. One was handled and blown by Bobby Howry and one was a 2.2 inning win by Mike Timlin, who entered with the game tied. Maybe the choice of Fox was a poor one, what with him returning after a year lost to injury.

Besides this is not exactly new for this team. The Red Sox actually did implement this sort of system with Derek Lowe in his three years as a closer (kudos to Chris DeRosa for pointing this out). Lowe pitched 91.1 to 109.1 innings in 67 to 74 games each year and saved 15, 24, and 42 games. That's more in the Sutter mold than the Eckersley one.

Lastly, I would be remiss if I didn't react to this statement from Jones:

Our skipper plays games against teams with dominant closers as if they are eight innings long. If we're losing, we know who is coming in in the ninth.

Any Rockies team that approaches a game as if it were eight innings long is doomed for failure. I understand Jones' point, but given the fragility of leads in Coors, this statement is ludicrous.