Ken Giles: Jerome Holtzman’s Final Victim?

On Monday, Astros manager A.J. Hinch told reporters that Luke Gregerson would start the season as his team’s closer. Given Gregerson’s successful stint as closer in 2015 and his general track record of success in recent years, such an announcement might sound like a formality. After all, “Team’s Good Closer to Remain Closer” is not exactly newsworthy.

What made the announcement interesting is that over the winter the Astros traded Vincent Velasquez, Mark Appel, Thomas Eshelman, Harold Arauz, and Brett Oberholtzer to the Phillies for Ken Giles and Jonathan Arauz. Giles, as you likely know, had been extraordinary in relief over his first 115.2 innings in the majors and could easily be considered one of the best five or ten relievers in the game. Naming Gregerson the closer and Giles the setup man raised some eyebrows given the price the Astros paid to acquire Giles four months prior.

It doesn’t matter if you subscribe to the projections, recent performance, or a simple visual analysis of their stuff, Giles grades out better. We project he’ll beat Gregerson’s ERA and FIP by 0.30 to 0.40 runs this year and Gregerson has never had a season on par with Giles’ performance to date. Both generate lots of swinging strikes, but Giles has the velocity that appeals to scouts. Gregerson is a very good reliever, but there isn’t a plausible case to be made that he’s better than Giles. Yet when Hinch went to the bullpen on Tuesday, it was Giles in the eighth and Gregerson in the ninth.

The sabermetrician in me celebrates this as a watershed moment for optimal reliever usage. After all, we have been asking managers to avoid confining their best relievers to save situations for years and this move would seem to open up Giles for high-leverage moments in the sixth, seventh, and eighth innings. A win for modern thinking!

Giles is arguably one of the best relievers in the game and his manager has stated clearly that he will not be restricted to the save situation. If there are two men on and one out in the seventh inning, Hinch will be able to call on Giles without opening himself up to all sorts of second guessing and ruffled feathers. In a perfect world, Hinch would have named no closer at all and left himself open to using Gregerson with similar flexibility, but this certainly looks like a step in the right direction. Get ready for Ken Giles, Relief Ace.

Yet something about this decisions doesn’t sit right. While the Astros are certainly forward-thinking, the likely explanation for Hinch’s decision is that Gregerson has a guaranteed contract and Giles will soon be heading to arbitration and will make more money during that process if he accumulates saves. While everyone is slowly getting on board with the idea that saves don’t reflect reliever quality, the arbitration system, which is built on comparing players to other players, still does. And the system rewards relievers who earn saves.

In other words, the decision to use Giles earlier in games will save the Astros money and probably won’t cost them any wins. In fact, if we’re right about bullpen roles, it will actually help the team win by freeing Hinch to use Giles in a variety of situations and innings. This seems to be a case in which the economical choice for the Astros to do (depress Giles’ save total) actually lines up with the club’s short-term competitive incentives (use Giles in many different innings).

We’re accustomed to seeing teams trade a small amount of competitive value for long-term financial gains when it comes to service-time decisions. Teams regularly give away a few weeks or months of value to push back a player’s free agency or arbitration clock. As analysts, we accept that strategy as something that hurts the team (and player) in the short run but which also serves their long-term interests, because the millions saved overwhelm the small number of wins lost while an inferior player gets reps with the big-league club. But the Giles/Gregerson case and the lower profile ones that have come before indicate a legitimate flaw from the player perspective, as the team is not being asked to make a competitive trade-off in exchange for their financial gain.

In the case of Giles, the Astros don’t have to weigh the short-term competitive cost against the long-term gain, because there likely isn’t a short-term cost. Even if you think one exists, it would be tiny compared to the millions they will save by limiting Giles’ save total. In reality, the Astros are gaming the system by taking a player who should earn substantial arbitration salaries given his performance and limiting that future income that by using him in a role that arbitration does not reward.

Put another way: Giles has a financial incentive to be used in a way that is less beneficial to his team’s chances of winning. This is a perverse structure in need of realignment. We’ve long since known that saves are a worthless measure of reliever performance, but the arbitration system still pays relievers for accumulating them. Getting a save is better than blowing a save, of course, but there are lots of ways to get a save and then pitch worse than guys who didn’t get a save chance at all.

The Giles/Gregerson case highlights why the practice of rewarding saves in arbitration needs to change. It’s one thing to put management against labor in a battle for wealth, it’s another to put players in a position where they are trying to pitch their way out of a more important role to improve their earning potential.

There are essentially two ways to change this system. The first option is to wear it down piece by piece, with arbitrators slowly de-emphasizing saves until the pool of well-paid past awardees contains non-closers. To accomplish this, non-closers would ask for higher awards armed with more useful stats, but teams could also use this strategy to pay mediocre closers less. The second option is to collectively bargain saves out of the arbitration process. This seems less likely, given that many players and teams still believe in the value of saves, but it would be a more decisive blow to the current order.

It’s common practice in sabermetrically inclined writing to extol the virtues of teams who get more bang for their buck, because that means those teams can invest their savings in something else that will help them win. Depressing Giles’ long-term earnings is such a strategy, but it comes with a side of ethical ambiguity that isn’t as simple as management and labor fighting over a pot of money.

The problem is that it does actually make good baseball sense to keep the team’s best reliever away from the closer’s role. The Astros are getting the best of both worlds, and it’s a bit more difficult to argue on behalf of Giles in this case because this is what’s best for the 2016 Astros’ chances of winning a championship — a goal which he certainly shares. It’s clearly not fair to Giles, but it’s also difficult to criticize a team which is using its players in the most effective way possible.

Giles’ case has the potential to change things, because there is no pretense behind which to hide. Giles was an established closer and the team paid a small fortune to acquire him. He’s their best reliever and he won’t be asked to get saves (at least for the time being) to his financial detriment. This isn’t Giles’ fault and it’s not the Astros’ fault for stumbling into a win-win. The system is the problem and needs to be dismantled. Perhaps this shining example of its flaws will be the impetus for change and signal the beginning of the end for baseball’s worst statistic.