NL Central closers: Concern is warranted

It's a mantra this time of year: Early-season action can lead to early-season overreactions, producing hastily conjured trends. Consider the lot of NL Central closers after their teams' first series:

March 31

  • Up 6-3, Brewer John Axford blows the ninth-inning save against the Reds by allowing four to score.

  • The Cards' Ryan Franklin blows his ninth-inning save opp up 3-2 against the Padres by giving up a tying homer to Cameron Maybin.

April 1

  • Astro Brandon Lyon blows a 4-2 lead over the Phillies by surrendering three runs and allowing hits to six of the seven batters he faced.

  • Joel Hanrahan of the Pirates notches a clean save to lock up a 6-3 win over the Cubs, making him the division's first closer to come through.

April 2

  • Although not a ninth-inning save opportunity, Bucs set-up man Evan Meek gives up five runs in the bottom of the eighth against the Cubs -- not a classic save opp, since all he'd get for his efforts if he did well is a hold, but it winds up being a bit of BS action nonetheless.

  • Carlos Marmol closes out the Cubs' 5-3 victory by striking out the side.

  • Francisco Cordero logs a save up 4-1 against the Brewers, but gives up a run.

April 3

  • Marmol blows a 4-3 lead in his ninth-inning save opportunity against the Pirates.

  • Hanrahan notches his glory stat in the bottom half of the inning, making him the division's picture of perfection as the only unscored-upon closer in the Central.

That's a total of four saves in eight classic "closer" opportunities in the ninth inning, and two of those were by Hanrahan, a man who had to pitch his way into the job over Meek's rival claim in front of new manager Clint Hurdle. Hanrahan's far from what you'd call an established closer, having never finished with a double-digit save season. Interpretive metrics like any of the flavors of Wins Above Replacement (whether FanGraphs' WAR or Baseball Prospectus' WARP) say nice things about Hanrahan's 2010, but the balance of his career suggests that, as with many relievers, what you get from him year to year is far from a sure thing.

Evaluating relief performance is easier than projecting it, of course, but because relievers face so few batters over a season, let alone in April, the results aren't nearly as cut and dried as you might wish. Take leverage-minded relief metrics and their potential problems: Referring to Win Probability Added, Axford's four-run disaster on Opening Day took a 91 percent shot at a win and converted it into 100 percent of a loss, but that goes far toward noting that leverage metrics only go so far to document value -- Axford was just a batter away from notching his glory stat, after all, and Ramon Hernandez's homer matters very little if Joey Votto doesn't work his way aboard for a walk, or if the umps rule that Brandon Phillips was out of the basepath while avoiding Casey McGehee's tag on Scott Rolen's grounder to third. That play created a bases-loaded situation that Axford had to pitch out of.

The full sequence of events, not all of which Axford had control over, define the value of the subsequent homer. But Axford's giving it up and Hernandez's hitting it winds up being the one play that accrues the most value, despite it's absolute dependence on everything that happened before to become possible. So Axford gets that big -0.96 WPA hit on Day 1 of the season.

Concern over overreacting to early-season performance, however, is merited in part, especially when we're talking about new managers or newly elevated or recently acquired closers. If Axford keeps getting lit up into April, that's not going to represent a statistically significant sample of what he's capable of, but it would also go pretty far toward Ron Roenicke subsequently deciding that enough's enough, and maybe it's time to hand save opportunities to Takashi Saito or somebody else.

The other thing to think about is that poor performance from a pen early in the season can change how a manager chooses to manage his bullpen, and that's what is at stake now, for the managers and for their relievers. For example, when Eric Wedge was managing the Indians in 2006, an early-season series of bad ballgames from his middle-relief crew led to a decision to favor letting his starters finish their own innings. As an adaptive strategy, it led to a league-leading tally of blown quality starts -- games when his starters had gone six innings and allowed three runs or less (I'm going with runs, not earned runs here), all because he'd acquired a strong distrust of the bullpen he had on hand.

In light of this "trend," you can sort of understand why Tony La Russa looked at Sunday's 2-0 lead over the Padres, with Jaime Garcia's pitch count at 95 through eight innings, and decided to leave well enough alone rather than risk another ninth-inning conflagration. We can't say which of these late-game disasters will herald a real trend in terms of how managers run their staffs, but just as a late-game blown save isn't always the product of a single independent event, these kinds of decisions will be the product of cumulative, mounting frustrations with bullpen performance, and these decisions, no matter how little evidence exists to support them either way, will get made just the same.

Christina Kahrl covers baseball for ESPN.com. She helped found Baseball Prospectus in 1996 and was voted into the Baseball Writers' Association of America in 2008 as one of its first four Internet columnists. You can find her ESPN archives here, and follow her on Twitter here.