Wednesday, April 25, 2007

Time to place your base-ball thinking caps on your heads, gentlemen

The Myers/bullpen situation has the potential to become one of the more fascinating baseball subplots of the season, if only because it's such a rare occurrence that you see a top starter in his prime get moved out of the rotation. When I first heard of the move, I chortled, like a lot of people. But the more I've thought about what the move can accomplish, I actually wondered if it wasn't really the kind of forward-thinking club management that seemed completely anachronistic to the SOP for this team since I can remember.

What's important to consider, in this situation, is the true value of having an "ace" reliever, a role that should not be confused with "closer." Contrary to what a lot of people think, the ninth-inning is not automatically the most vital frame in a close baseball game, though it often is. What determines the most important inning, or cluster of innings, is leverage, a value explained by BP as such:

Leverage is calculated as the ratio of the impact of one additional run at a point in time to the impact of one additional run at the beginning of the game averaged over all plate appearances.

Essentially, measuring leverage is calculating how important scoring (or preventing) a run is in any given situation, weighted against the remaining number of outs for the team in question. Every out becomes more valuable as the game progresses, since there will be fewer subsequent opportunities to score. That's why such a premium is placed on having studs in late-inning roles; there's little question that the Padres' 2006 division title came down to the Linebrink-Meredith-Hoffman trio, as they combined to provide a little more than 13.5 additional wins over replacement level last season (according to BP's WXRL table), an astounding (and way, way, way league-leading) figure. However, most teams make do with one truly dominant arm in the 'pen, and that person is almost always the closer (except in the case of teams like the Indians, Tigers and Braves), who will almost always be deployed according to the following lightning-etched decrees:

1) Thou shalt not use thy closer prior to the eighth inning, and almost always in the ninth alone;

2) Thou shalt not use thy closer in games where ye are not to the fore; equalized games and one-run deficits are the realm of lesser men;

3) Thou shalt not use thy closer in games where ye are to the fore by more than three runs.

Conveniently, almost all situations within these parameters constitute high-leverage situations, with the exception of three-run leads against non-juggernaut teams. They also encompass all situations in which a "save" can be awarded. Which, if you're a fantasy leaguer, is about the only thing one can expect of value from a bullpen arm. But like most pitching stats that are part of the 5x5 roto structure, saves are bad at assessing the value of a reliever, not to mention closer. For the sake of focus, I'll leave that for another day.

What is important to note is that just because almost all save situations are high-leverage situations as well, not all high-leverage situations are save situations. Often, games hinge on events that take place in the seventh and eighth innings, situations in which teams are often willing to either let a tiring starter continue to work or bring in a reliever who is not the best available arm. This is almost always a bad strategy, particularly when the situation is charged with high-leverage variables.

For example:

It's the seventh-inning, and your team currently leads by one run. There is one out, but runners are on second and third and Albert Pujols is coming to the plate (which means that this isn't a situation where a LOOGY can be of use). In your bullpen, you have two relievers to decide between:

Reliever A: .30 DERA, 2.8 K/BB, 1.265 WXRL
Reliever B: 5.97 DERA, 0.5 K/BB, .118 WXRL

Easy decision, right? But the problem is that Reliever A is Jonathan Papelbon and B is Mike Timlin. While Tito's proven to be one of the more progressive managers in the majors, it's a reach to think he'd put Papes out there in the seventh, even though this is very clearly the turning point of the game. And if Tito wouldn't do it, sure as shit there ain't another manager in the majors who would.

The point of this exercise wasn't so much a rant about bullpen misuse; it's within this matrix that I think the Phillies can benefit. Tom Gordon is the team's "closer," even though he's probably the second-best reliever in the 'pen. I think this is a massively liberating situation for the Phillies, since they can now utilize Myers (presumably the bullpen's ace) in the highest-leverage situations possible, no matter what inning it comes in. If the Phillies have a two-run lead in the ninth against the Fish, let Gordon pitch: the odds of the Phillies winning in that situation are fairly staggering no matter who's on the bump. If Myers isn't penciled in for that ninth-inning-only role, he can then be deployed in those situations where his team most needs a high-strikeout bullpen ace. This kind of use -- not shifting him to closer -- would most likely give the Phillies the best possible increase in overall win expectancy, not to mention freedom to deploy the bullpen in more creative methods.

Unfortunately, all of this is strictly hypothetical, since there's no way the chimps in charge of the Phillies will ever utilize this kind of gambit with Myers, as evidenced by how they've used him so far. But if we could somehow get Charlie Manuel to flip spots with Manny Acta, and then buy Acta a BP Premium membership, this could be really interesting.


St said...

An interesting take. A couple of non-snarky questions:

1. Isn't it impossible to know which situations are highest-leverage until the game is over? Meaning, if, say, Charlie throws Myers out there in the aforementioned situation (btw, who doesn't just walk Pujols there?) in the 7th inning. Myers gets the out. But then in the eighth the bases are loaded with no outs and Scott Rolen's at the plate (I think that's impossible, given the Cards' batting order, but go with me here.) Wouldn't the second situation be higher-leverage? In other words, isn't that one reason why closers are saved for the ninth, because doing so gives them the best chance of pitching in the highest-leverage situation?

2. I think the atrociousness of the Phillies bullpen tempers this discussion a bit. Meaning, the gap between Brett Myers and, say, Alfonseca or Geary (the two best relievers thus far) -- much less Flash, who's probably third or fourth on that list -- is prodigious, and so I don't think anybody wants Myers coming in to get a crucial out in the 7th and then leaving the team to rely on Flash or Rosario or Madson to close it out. Because, as we've seen repeatedly this year, that's not likely to happen.

It'll be interesting to see what happens as Myers acclimates to the bullpen, and as Charlie acclimates to having him there. He went to the 'pen right before the Phils started winning games by six runs (or losing them by two, as was the case last night), so he hasn't had a whole lot of opportunities to affect games.

I see you point, but am still inclined to think Myers should be the closer, and will be by the break.

(On a side note, a new wrinkle emerged when the Phils finally demoted Matt Smith yesterday and called up lefty Fabio Castro, probably their best relief prospect. Matt Smith was the best player we got in the Bobby Abreu deal. Seriously, was that trade great or what?)

Diesel said...

Hmmm ....

1. While it's impossible to predict if any given situation will definitely be the highest-leverage one of a game, you can usually make educated guesses. Removing the hypothetical I presented (you're right that walking Pujols is the only move there, if a high-leverage situation presents itself in the seventh or eighth, doesn't it make more sense to worry about the "bird in hand" than it does to worry about what might happen? Particularly in later-game situations, there's a decent chance you won't be going through the order again, so what we're talking about is a guy to dispatch the other team's best hitter(s) in situations where not preventing runs at that time will seriously hinder your win expectancy. More precisely: Unless you're able to lock up that tough spot in the seventh, you're probably fucked in the ninth anyway. At any point in the last three innings, a team is going to encounter its highest-leverage situation. When you make someone a "closer," you put yourself in a spot where you don't have that same flexibility to use your best reliever when you can fairly well assume he's most needed.

It should also be noted that the only hangup with the title "closer" is the expectation it brings with it. It's entirely conceivable that teams will quit being slaves to the save and not make the closer title so loaded with strict role definitions. But that's not the case with the Phillies, or any other teams, right now.

2. I think it's a safe bet to say that Flash is probably still the best non-Myers alternative in the bullpen, but it doesn't change how I see it, much. First, there's nothing to say you couldn't use a former starter to work more than one inning; conceivably, Myers could work the seventh and eighth there.

Of course, what I'm suggesting is a pretty big leap for most teams, and more importantly the fan base/media. For my plan to work, it would require a team have the intestinal fortitude to ride it out and not listen to what the scribes say when Myers gets that crucial out, and then Geary or Flash get shelled an inning later and lose the game. The bottom line is that it really doesn't matter what happens afterward, because without that crucial out it's usually academic anyway. Does that make sense? It's harder for me to be clear in comments sections, I've found.

Anyway, I see your point as well. But I think that there's some sound logic behind what I'm proposing, and I really think it could make a killing on the big stage.

AnEasyMark said...

"When you have to go to your 'pen, you lose and when you want to go to your 'pen you win."
Orel Hershiser's Big Book of conventional baseball wisdom

One of the places that I would like to look at, regarding 7th, 8th, inning scoring, is the teams that are 'leveraging' their way out of facing the opponents #1 reliever.

Diesel said...

One of the places that I would like to look at, regarding 7th, 8th, inning scoring, is the teams that are 'leveraging' their way out of facing the opponents #1 reliever.

I don't understand what this means.