Projecting rest-of-season saves

Oh saves, you fickle, fickle statistic.

All this week -- previously in "60 Feet 6 Inches" on Tuesday and "Hit Parade" on Wednesday -- I've been refreshing our projections, offering forecasts for more than 300 players from June 1 forward. But when it comes to the week's final projections group, the closers, things naturally get a little bit hairier.

Saves drive this position. Period. Lose 'em, and any hope of value is lost with them.

Think of it this way: Had I projected relief pitchers from, say, May 26 forward, and published them a week ago, Joakim Soria would have a whole different forecast. At the time, he was the Kansas City Royals' closer, one who had struggled initially this year but had at least a chance of righting himself while still getting the lion's share of the team's save opportunities. A week ago -- and this is merely a guess -- Soria might have been projected for 18 more saves.

Today, he's projected for nine, and that's actually a generous total for the game's newest ex-closer. But that nine-save drop is significant; it dropped Soria 20 spots in this week's rankings, and the only thing even keeping him in my top 40 is the prospect that he'll quickly turn it around for a team that has nothing but young, inexperienced alternatives. That's the way it goes when sniffing the saves market: The loss of those precious save chances can mean as much as a 25-spot (or more) decline from one week to the next, in almost every instance more than you'll ever see for any player whose drop wasn't a direct result of a disabled-list stint.

Fantasy owners are probably about as anti-Soria as they could be today. They'll probably call a No. 33 ranking "crazy," and feel that his replacement, Aaron Crow, isn't ranked high enough (he's 31st). When it comes to projections, they might be right; give five of those projected Soria saves to Crow and Crow probably rises 10 spots, while Soria drops another 10 or more. That's why, of any position, it's closer at which we're most often guessing. ERA, WHIP, strikeouts -- we can project those with at least a decent degree of accuracy.

The saves? OK, I'm confident today. But ask me again in a week.

Back to the Royals for a second, before getting to the projections: I'm not so certain that if you strip Soria of any of those nine saves, that those wouldn't all go to Louis Coleman, not Crow, simply because Coleman would be the one getting them later in the year. Just something to think about, if you're a Soria owner in the position to speculate: Crow might be off to a fantastic start, but his minor league track record was poor, littered with ERAs north of five. Coleman, meanwhile, was a lights-out minor league reliever who had a 2.16 ERA, 0.90 WHIP, 10.52 K's-per-nine and 4.27 K's-per-walk ratios in 62 career appearances. And while I'm a believer that minor league relief prospects often disappoint, they don't all disappoint -- look at Chris Perez. In time, Coleman could have that kind of impact.

As with the week's previous two columns, let's precede the projections with a few notes: These are my projections alone, not ESPN's, using my projections method. And as they've been formulated during the past several days, understand that even I've learned some things from the exercise; you might find some of this week's ranking shifts puzzling, and the simplest explanation might be that the projected numbers showed me something I hadn't considered before.

In addition, these rankings follow roughly the formula we use in our Player Rater, though they're not exact. Upside, injury risk, role and team factors come into play, particularly in lower tiers, and while a certain projection might seem worthy of a higher or lower ranking, consider that any might have impacted the specific rank.


Note: Tristan H. Cockcroft's top 75 relief pitchers are ranked for their expected performance -- and projections -- from this point forward, not for statistics that have already been accrued.

Tristan H. Cockcroft is a fantasy baseball analyst for ESPN.com and a two-time champion of the League of Alternative Baseball Reality (LABR) experts league. You can e-mail him here, or follow him on Twitter @SultanofStat.