UPDATE (12/12/2015): Upon being officially announced, the parameters of the Ken Giles trade are different than expected. In return for Giles and Minor League infielder Jonathan Arauz, the Phillies received Vincent Velasquez, Brett Oberholtzer, Thomas Eshelman, pitching prospect Harold Arauz, and 2013 first-overall pick Mark Appel.
The rebuilding Philadelphia Phillies traded Ken Giles, their young flame-throwing closer, to the Houston Astros for a package of four young players headlined by starting pitcher Vincent Velasquez. It's yet another seemingly steep price for an analytically-savvy team to pay for a closer after the Red Sox traded for Craig Kimbrel and the Dodgers attempted to trade for Aroldis Chapman.
The package - starting pitchers Vincent Velasquez and Brett Oberholtzer, outfield prospect Derek Fisher, and pitching prospect Thomas Eshelman - is reminiscent of the large haul returned for the more established Kimbrel. Certainly, relievers are risky propositions, but competitive teams seem to be spending more than in past years for players whose value maxes out around 2.0-2.5 WAR.
Take the Giles deal for instance - over the last two seasons, he's been about as good as they come (by both fWAR and rWAR, he's roughly equivalent to Kimbrel). His contract situation, with five years of team control remaining and two at a pre-arbitration rate, is without a doubt the most appealing of any elite closer in baseball. Even still, his total value has a lower ceiling due to his limited role.
But in trading him, the Phillies are receiving Velazquez, who entered the 2015 season as a top-100 prospect, an established back-end starter in Oberholtzer, a borderline top-100 type in Fisher, and a 2015 draftee in Eshelman. A two-win player making league minimum has a sizable amount of surplus value, but is it possible that these kinds of elite relievers are being undervalued in the public?
I'd hypothesize that the true dollar value of a player is more dynamic than a broad $/WAR model. Maybe a player is more accurately valued at the lowest cost of acquiring the same marginal WAR improvement that this player provides, but over one or more other roster spots.
To illustrate with an example, suppose there is a projected 3.0 WAR first baseman available in free agency, and an interested team currently has a 2.0 WAR first baseman. That free agent would not add three wins to the team's projected total; instead he would add only the one win over the player being replaced.
However, there's almost certainly another team with a replacement level first baseman currently penciled into the lineup. The free agent's marginal value to this second team is his full three wins, and they should have more urgency in signing him. The first team might be more wise to look around the diamond at other positions to see if the same marginal value could be added with a less expensive player.
For instance, they may have a 1.0 WAR shortstop. In signing a 2.0 WAR player at that position, they've added the same win to their projected total, but at the salary expected by a 2.0 WAR player, not a 3.0 WAR one. They could look at all available options at every position and perform a similar calculation, finding the least expensive method of adding one win of marginal value to the roster.
This might be why ace starting pitchers and closers seem to have heightened trade value over their traditional surplus value - they replace a 5th starter or 7th reliever, replacement-level players on most teams. These are the cases in which 100 percent (or more) of the player's value is marginal improvement. In every case, they're directly replacing a 0.0 WAR player with 2.0 WAR (or far higher, if an ace starting pitcher is involved).
This is why teams like the Red Sox, Dodgers, and Astros are chasing after relievers - they have well-rounded teams with positive contributors at most positions. It would cost the Red Sox far more to improve upon their weakest position player by the same amount as adding Craig Kimbrel over a AAAA arm.
Additionally, while a starter or position player distribute their value among all situations, all of the closer's value comes in a high-leverage environment. This "leverage value" component should be more accurately compared against the second-best reliever in the bullpen, the one this new acquisition is replacing in those situations.
These teams could also just acquire that massively-expensive position player and move the player they are replacing for marginal value at another position - however, a competitive team might not want to exchange that kind of player for prospects, and select teams are moving Major League pieces for other Major League pieces. On top of that, the team would need to ensure they're exchanging this player to fill other marginal holes.
Another team would have to be a match on several levels: they need to want the player being replaced, be willing to return other Major League pieces for him, have players that add marginal value to the original team, and can afford to assume the player's contract. It might be far less work to just acquire the reliever.
So, when Ken Giles and Craig Kimbrel are moved for what traditional surplus value models would declare an overpay, it may be because more of the value of a 2.0 WAR closer is directly added to the win total. This change in philosophy does seem abrupt, but the likely reason it's being seen only in the trade market is that there are no consistently elite, free agent closers available this offseason.
Instead, we're left with the sticker shock of trades involving established pitchers and the rising prices of free agent starters (David Price, Zack Greinke, et al.) as teams reevaluate how much these kind of players actually add to the team.
. . .
Spencer Bingol is a Featured Writer at Beyond the Box Score. You can follow him on Twitter at @SpencerBingol.