Sunday, April 26, 2009

“Theme” My Bippy

When Matt Cain left today’s game with a 4-1 lead, broadcasters Mike Krukow and Duane Kuiper were aglow about how, for a change, this year’s “theme” was, “Score Runs for Cain.” The Giants had a number of opportunities to score more than four, but they went into the ninth with that three-run lead, which Brian Wilson coughed up in a heartbeat, leading to extra innings. At this point, hey, the Giants could still win, but they’ve already burned all their best relievers, having used Jeremy Affeldt, Bobby Howry, and Wilson. Who’s left? Brandon Medders, Justin Miller, and Merkin Valdez. Of these, the best appears to be Valdez (and he’d better be), but he generally can’t throw more than an inning. I’m not entirely convinced that Medders and Miller can either. So I can’t say I have a good feeling about the rest of this game.

Now, Wilson came in with that three-run lead. That’s SOP for ballclubs these days: If it’s a save situation, bring in the closer, without fail. A save situation, as you probably are aware, is one in which the reliever enters the game and meets the following conditions (according to Rule 10.19 of the official baseball rules): 1) he is the finishing pitcher in a game won by his team, 2) he is not the winning pitcher, and he either (enters the game with a lead of no more than three runs and pitches for at least one inning; enters the game, regardless of the count, with the potential tying run either on base, at bat or on deck; or pitches for at least three innings, no matter how big the lead is.

Most often you’ll see your closer on the mound at the beginning of what his team hopes will be their opponents’ last offensive inning. Those are the easy saves. The pitcher’s coming in with a built-in lead, and all you have to do is shut down the other team for an inning. A one-run lead, certainly, is precarious, and this is where closers to be at their most intense. The bigger the lead, though, the less they concentrate, to the point where they seem to dread coming in with a three-run lead. Heck, I don’t know, maybe Wilson was exactly as intense and focused as usual, but I betcha... not. He got battered around, finally giving up the tying runs on a two-run homer by Juston Upton of the Diamondbacks, who seems to hit only Giants pitchers particularly well.

Now, this happens to every closer at some point. He’s not always gonna shut ’em down. In fact, closers tend to succeed about 80 to 90 percent of the time, which I don’t think is so great given the advantage they have when they enter the game. Granted, not all saves are easy. Closers aren’t going to throw three innings, so you won’t see Wilson enter the game with a 16-3 lead in the seventh, so get that out of your head right now. (Like the Giants are ever gonna score 16 runs in a game this year.)

And when I say “easy,” I mean in a relative sense. Starting the ninth with a one-run lead doesn’t seem so easy if, for instance, you’re facing three straight hitters who routinely knock baseballs out of parks. That one-run lead can disappear rather quickly. Plus there are those save situations where a pitcher might enter the game with a one-run lead, no outs, and the bases loaded. One sacrifice fly later: blown save.

In essence, my standard rant about saves is that saves are a gaudy statistic. Francisco Rodriguez saved over 60 games last year, smashing the existing record. And it’s not as though that’s not impressive. But you can’t equate it to, say, a starting pitcher winning 20 games—I’m not even sure you could equate it to 15 games—because success is built into the statistic. Usually you have to really screw up if you want to blow a save.

All that said, it’s still a better-defined stat than wins for a pitcher. Wins seem far more arbitrary, and on some occasions, they can be awarded on the whim of an official scorer. Yesterday’s victory over Arizona would be a good example, because Randy Johnson came out in the fourth inning, having walked seven hitters, but the Giants won anyway. Johnson’s immediate replacement, Justin Miller, got the win, giving up a run in an inning and two-thirds. Now, had the official scorer judged that Miller had pitched either “briefly” or “ineffectively,” he could’ve given the win to any of the four subsequent pitchers. Not that he should have.

But what do you do in a game where one team had an 8-0 lead and kept scoring, while the other team scored in bunches of four or five or eight but still couldn’t catch up? Let’s say the starter left in the first, and seven relievers all got lit up? Hard to know who gets the win, really. Let’s say that last pitcher doesn’t get lit up, and puts down the opposition in order in the ninth, with a one-run lead. Well, he really should get the save, but since all his teammates had pitched either briefly, ineffectively, or both, he ends up getting the win, which probably doesn’t help him when it comes to contract negotiations.

So neither wins nor saves (albeit clearly defined) is a great statistic. The Giants have shown us over the last few years, thanks to pitchers such as Matt Herges, Tyler Walker, and Brad Hennessey, that anybody can rack up plenty of saves. A guy could save a good 40 games despite an ERA over six, but Lord, what if he leads the league in saves? He’s your closer again next year. Yeccchhh.

Current conventional wisdom has a major league manager putting his best reliever into the closer role, which means that sometimes the games he enters aren’t on the line. How wise is that? I mean, there’s nothing wrong with bringing him in with that one-run lead, but shouldn’t your best reliever pitch in situations that are more crucial than, say, a three-run lead? Bill James espoused this, and I think he has a point. For example, assuming Brian Wilson really is the Giants’ best reliever, I’d sure rather see him in the game right now, with the game tied in the bottom of the eleventh, in an effort to keep the game exactly where it is. And what if the Giants score a run in the twelfth and are poised to score more? Wouldn’t that be an opportune time for Bob Melvin to bring in Chad Qualls, the reliever most likely to douse a rally? In fact, Qualls pitched an inning earlier, with the game tied and no save to be had. In fact, that was a fine time for Qualls to pitch. But in extra innings, you’d be reasonable to expect the home team’s closer to pitch with the game still tied.

You don’t see a closer entering a game in the eighth to preserve a tie, or to keep the opposition down to a one-run lead. You sure don’t see him in the third inning, trying to get through a bases-loaded, nobody-out situation. But what if you did? Would this put a permanent rend in the fabric of baseball space-time?

Yeah, pretty much. It has a lot to do with people’s comfort in pigeonholing other people. Pitchers like to have a role, even if it’s just “You’re the guy who comes in when we’re down 10 runs.” If you’re the closer, you’re the guy who comes in in a save situation—indeed, if you pitch in a non-save situation, it’s either an extra-inning scenario or you’re just in the game to “get some work.” You aren’t going to prepare to come into the game before—at the very earliest—the eighth. You certainly won’t even consider the possibility of being brought in with the bases loaded and nobody out in the third, even if the next three scheduled hitters are a combined 0-for-60 against you and have slider bat speed at best.

This is due to the “closer mentality,” which just means the mindset that major league managers use a hundred percent of the time. I’m not sure I’d know how to do things any better—I fall into the same pattern when playing tabletop or computer baseball games—but the closer mentality has to do some harm. For some teams, it has to be far more harmful than it’s worth. Look at the 1987 Giants, when closer Scott Garrelts blew 10 saves in 22 opportunities. Maybe he just shouldn’t have been pitching in save situations. Maybe he would’ve thrived in middle relief, or as a starter, as he did later. But no. He’s the closer, so until the manager changes his mind, he’s always going to pitch in save situations. Not tie games on the road. Never to preserve a one-run deficit.

Baseball hasn’t done a great job in evaluating relief pitchers. Even blown saves and holds aren’t yet official statistics. But on a team when the best reliever pitches in a variety of situations, when managers use their relief staff based on matchups, who’s pitching well and who isn’t, etc., his baseball card might well show a good ERA, a healthy WHIP (that’s walks-plus-hits per inning pitched), and an incredible strikeout ratio, but it won’t show a lot of saves, maybe not even holds. This’ll cost a guy money.

It might be worth trying out such an approach in a tabletop game or something, but it won’t be happening in major league baseball anytime soon.

(Nor will the Giants be winning today’s game, as Connor Jackson just whapped a drive that bounced on the warning track with the bases loaded in the twelfth. Not a sniff of Merkin Valdez was to be had, either.)


  1. All of this makes me think about something we'd spoken about a while ago. Why don't teams have 14 pitchers and just make them all relievers, with each pitcher throwing 2, 3 innings tops in a game. Or even one inning per pitcher, nine pitchers total, and do this every game.

    Much of the answer is precisely what you said here--that there's a "tradition." "Guys have gotta know their role." Bunkum.

    Really, why doesn't a team do this? If you're an NL team maybe you have 13 pitchers, and among the remaining 4 non-starters all you need is another catcher, two great hitting pinch-hitters to come up in every game when the pitcher's spot comes up, and one utility guy in case someone gets injured.

    Genius! That's why the idea is waaaaay down here deep in the dark misty nether crevices of a blog comment.

  2. I'm on the other side of this one. (Yes, I still exist.) Pick your nine best pitchers and let 'em get all the innings they can handle.
    Carry three catchers, two guys who can hit coming off the bench, and three utility men.

  3. Hi, David—great to hear from you.

    I was gonna comment on your post here, but I thought I’d make an actual blog entry out of it.