It doesn't seem to make a ton of sense.
Why would the spiritual heirs to new-age Moneyball be interested in a reliever? Didn't Billy Beane make a big deal about how closers are overvalued and should be moved to sucker teams? Didn't he make a big deal about how paying for closers doesn't make sense?
Well, if you noticed, the A's spent big for Jim Johnson last year. That failed pretty quickly, but the thinking behind it shows a shift in the market and a fundamental flaw in Moneyball critiques.
What Beane talked about in the Michael Lewis book was finding market inefficiencies. At the time, that means finding high on-base guys and not overpaying for closers.
Now, though, the pendulum has swung the other way. While virtually every team values OBP much more than they did 15 years ago, proven closers have lost market share. It's harder and harder for guys who have pitched in the ninth inning to get mega contracts.
Closers, though, are still very good pitchers. In most cases, they're the best reliever on a given team. If their price has been deflated a little by the market, their value might actually match up now with their cost.
But, we're getting ahead of ourselves. Why would Houston specifically be interested in David Robertson, the Yankees former closer who's on the free agent market? Does this team really need an expensive closer? Would he be worth the contract and the lost pick?
I talk a lot about how relievers are fungible. I don't think money can fix a bullpen. Buying the three most expensive guys on the market won't make Houston's bullpen unhittable, but finding an undervalued guy and paying him what he's worth could be a solution.
I'd posit that there is a small subset of relievers who defy volatility. The names are familiar. Guys like Mariano Rivera and Trevor Hoffman jump to mind. Joe Nathan was solid for many, many years. Going back a bit, Eddie Guardado put in the work for a long time at a high level. Billy Wagner did the same in Houston.
Most relievers are fungible, but an elite few may be able to be successful every year.
What defines them? Well, let's try strikeout rate. Since 2000, 22 relievers have posted a career strikeout rate above 30 percent. In that same time span, there have been 154 individual seasons by relievers with a K rate above 30 percent.
Yet, only 17 of the 92 total relievers in that group posted three or more seasons of 30 percent or better K rates. Robertson is one of those 17.
Eight players posted three seasons with five posting four and three posting five. Of course, that counts only qualified seasons. Robertson actually had a fourth season over 30 percent in strikeout rate but didn't throw 50 innings in it.
Most of the names on the list have been consistently good. K-Rod, Aroldis Chapman, Octavio Dotel, Wagner, Brad Lidge and Nathan all put together plenty of high-strikeout seasons. All 17 players also excelled at the highest level for an extended period of time.
To be elite isn't just about striking guys out. If the Astros are to invest in a big-ticket reliever, they also want someone who is durable. They do not someone who ends up on the disabled list immediately, which is the other reason why relievers are fungible. They get hurt an awful lot.
That's due to a lot of reasons. Take Brad Lidge. One of the reasons he was shunted into the bullpen was that he couldn't stay healthy as a starter. That wasn't a problem for Billy Wagner, but injuries did derail a promising career for Eric Gagne and B.J. Ryan.
Robertson, though, has remained very durable. Of those 22 players with K rates of 30 percent or better since 2000, he's got the fifth-most innings and the 10th-best K rate. Most of the guys above him on the innings list also sustained their success well into their 30's.
While we may not be able to predict how to get a pitcher to stay healthy and effective, there's a bit of self-selection at work. Every person is unique and some have super-strong UCLs or whatever. Maybe Robertson is one of them.
Even if he's not, it's highly probable that he puts up at least two seasons of similar effectiveness to what he's done in the past with the Yankees.
That's where we get to the crux of the issue. How much value does Robertson have to provide to justify the expense?
Well, if we trust FanGraphs' reliever valuations, Robertson has been worth anywhere from $7-$11 million in the past four seasons. Baseball-Reference has him worth slightly more per season while Baseball Prospectus is less bullish, putting his per-season price down around $6 million per season.
FanGraphs projects that Robertson will end up with a contract around three years and $32 million. That means he'd be overpaid, but not by a huge margin. If he produces two wins above average for the next three seasons, he easily justifies the contract.
There's also the matter of the free agent compensation. Houston would have to give up that competitive balance pick from the Marlins, which should fall in the mid-30's of the draft. New research, though, suggests the value of that pick is three times its slot value.
If we assume that draft pick will be worth around $1.5 million, that means all Robertson has to do to justify losing the asset is to produce one win above replacement in three seasons. Even the gold standard of reliever flops, Jonathan Papelbon, has done that with the Phillies.
Buying players in free agency will never be a good way to spend money. The players are older and rarely stay on the upper part of the aging curve with their new team. But, a short deal given to a reliever makes a lot of sense. The market may be overvalued for position players and starting pitchers. But, with a reliever tagged with a qualifying offer, the bidding may be so low that it brings his price down to manageable levels.
If that contract pays him about what he's worth as a player, it's a good deal. The Astros found a relative bargain in Scott Feldman like that last season. The very well could strike again with a reasonably priced pitcher in Robertson.