The closer position in baseball is a relatively recent development, one which has only come about as a full-time position with specific duties in the past twenty years. There were closers before then, guys who came into the game in tight situations, but they might appear in the sixth or seventh inning as the ninth, and (at least early in this century) might be starters pitching on their off-day.
But as pitching has become more specialized—due both to the expansion of leagues (and hence the thinning of talent) and the improvement in baseball hitting—so have positions within the pitching roster. Nowadays, we have starters and relievers, but it’s rare for a pitcher to do both, and when he does, it’s often because he’s been “demoted” to the bullpen.
Starters are used every fifth day or so, with rigorously strict workout and rest days in between. Managers would only consider bringing them out of the pen in important and tight situations, as Lou Pinella famously did with Randy Johnson in 1995, bringing him on for three innings in Game Five on only one game’s rest, and advancing them past the Yankees for the Mariners’ first playoff series victory. But these moments are famous in part because they rarely occur.
Relievers, on the other hand, have a little fluidity, though they are generally divided into long relievers (who may also be spot starters in a pinch), short relievers, lefty specialists (sometimes called LOOGYs, or Lefty One Out Guys), setup men and closers. In a ideally managed game these days, the starter will go seven innings, the setup man will pitch the eighth, and the closer the ninth, assuming the game is tight.
This increased, almost anal-retentive, specialization among most managers, coupled with an attentiveness to statistics, has meant not only the rise of the full-time closer, but also his accumulation of saves. The “save” is an often-derided statistic, dependent upon situation and management decisions (as, it should be added, are classic stats like RBI or ERA), and confined to a narrow set of conditions when the game is on the line. The closer must enter the game with no more than a three-run lead, record at least one out, and finish the game. Plus, he must pitch three or more innings (this rarely becomes a save), and enter the game with the tying run on deck, at bat, or on base.
So it’s easy to see how situational the saves statistic is, and it’s led directly to the usage of the closer in recent management theory. No more proof of this need be offered other than the all-time single-season saves leader list, all of whom have pitched since 1990: Bobby Thigpen of the 1990 Chicago White Sox currently holds the record with 57, while Mariano Rivera is tenth with his 50-save performance of 2001. Eric Gagne and John Smoltz have both threatened to unseat Thigpen, each coming within two saves in 2003 and 2002, respectively.
The latest guy to challenge Thigpen is the Anaheim Angels’ Francisco “K-Rod” Rodriguez, who has been their closer since taking over in 2004 from Troy Percival. Since then, he’s notched 178 saves for Anaheim, including a 47-save 2006 and a 45-save 2005, and has averaged 32 saves throughout his career (including about two seasons as Percival’s setup man).
This year, however, he’s broken his average halfway through the season—K-Rod racked up his 34th save last night against Oakland, tying him with Smoltz’s 2003 record for most saves by the All-Star Break. At this pace, he should break Thigpen’s record, but there’s a whole lot riding on that, not all of which Rodriguez can control.
For one rather obvious thing, saves are dependent upon both team victories and victory margin. Margin of victory is the more important factor—it’s not at all unusual for a closer on a losing team to rack up his share (Brian Wilson of the 37-48 Giants currently leads the NL with 23 saves). In some ways, it’s easier for a losing team to get more saves, since their win margins tend to be tighter. A real juggernaut will tend to blow out other teams, leaving no room for saves. Thigpen recorded 57 in large part because he was only one of five closers in baseball history with 57 save opportunities.
The Angels seem well set-up for this, as Scoscia is an excellent manager, meaning he tends to keep his team in games, and is good at making the decisions that give them a narrow edge of victory. But K-Rod can’t control this. He also can’t control his own health. He’s avoided the DL so far in his career, but his violent delivery has repeatedly made him a candidate for injury, and he lost four games to recurring ankle problems in April.
One thing he can control is his control, something that faded down the stretch in 2007. In the second half, his ERA jumped more than a point as he struck out fewer batters per inning and walked more; he went from 24 saves before the break to 16 afterwards. His worst month was July, when his ERA ballooned to 3.86 and he only notched two saves.
None of those numbers are catastrophic, but any month like that will likely blow K-Rod’s chances at saves immortality. Whether or not he does this may not reflect on his performance or that of his team—as discussed earlier, a power surge by the Angels may actually have the opposite effect on his saves chase. And that’s why so many people deride this statistic, for its emphasis on individual achievement in a narrow set of circumstances, one that doesn’t necessarily coincide with your team’s performance. A save means a win for your team, yes, but a win doesn’t always mean a save.
Those of us who play fantasy baseball are famliar with that, the letdown when your closer’s team scores another run in the eighth or ninth inning, putting that elusive S out of reach. Bill James and other sabermetricians argue against the closer’s sudden importance for this and other reasons: sometimes, a game is decided in the sixth or seventh inning, but no self-respecting manager today would “waste” his closer to “save” the game at that point—and the saves stat wouldn’t recognize it if he did.
Baseball went more than a hundred and fifty years without the necessity of a last-inning man, so the argument goes, so why do we need it now? As closers have come and gone, managers often assign a particular mystique to the position, giving the closer spot to “proven” closers with poor stats, instead of promising young guys with better stuff.
This enrages methodical and eminently logical sabermetricians like James, and with good reason, even if the “proven” moniker is often slapped on managers and players alike. Old candidates are resurrected for the same managerial positions, and aging vets given the nod in the field over untried rookies. Psychologists have proven that people prefer the risk they know over the one they don’t, so this tendency is hardly surprising.
But it does show the subjectivity of the choice and usage of closer, and this—along with the narrow set of circumstances required for the save—means that we shouldn’t assign the same weight to K-Rod’s pursuit of history as we did to the undeniable authority of the home runs (whether chemically aided or not) of Maris, Sosa, McGwire and Bonds. No matter what other circumstances might have been aiding or abetting these sluggers, the home run is a singularly individual act, one which only requires putting a good swing on the right pitch, and such things as batting order and game score only bear marginally on their ability to do so.
So let us offer measured praise to K-Rod, recognizing that regardless of circumstance and usage, he’s still a guy opposing batters hate to face, and someone who’s made the best of what chances he’s been given. He may have 34 saves because he’s been given 36 chances, but he’s also got a success rate in those circumstances of 94% and—in baseball or life—that’s an excellent rate of return.
