May 1, 2003
The Red Sox Bullpen
I was on the couch last night around 8:30, and I mentioned to Sophia that I really didn't feel like writing my column. She offered, sweetly, to write it for me, so I asked her what she could say about the Red Sox bullpen.
"Um, nice butts?"
I just smiled, and a couple of seconds later she said: "How about nice arms?"
I replied: "You're probably better off writing about the butts."
With Ramiro Mendoza and Brandon Lyon combining to allow two runs Wednesday, the Red Sox bullpen has now given up 16 runs in 23 9th innings. They've used seven different pitchers to protect 9th-inning leads (plus Jason Shiell to hold a 14th-inning bulge), and the question of who will pitch the ninth is a question that every member of Red Sox Nation has been obsessing about since long before Opening Day.
Much of this stems from a misunderstanding of the "no-closer" bullpen that the Red Sox assembled this winter. While the strategy downplays the save statistic, it's not about random roles for pitchers, which is how many people have interpreted it. Roughly speaking, the idea is that the bullpen's best pitcher will throw the highest-leverage innings, rather than being used solely in save situations. This is done in part by trading off low-leverage save situations--three-run leads in the ninth inning that can be protected by lesser pitchers--for higher-leverage situations earlier in the game.
There are other principles in play as well, such as minimizing the use of one-batter specialists, and allowing effective relievers to pitch multiple innings, but the basic idea is to allow game situations, not Jerome Holtzman's legacy, to dictate how the bullpen's best pitcher is used.
Little has done a fairly good job with the secondary characteristics. He's used his relievers for extended outings, and in part because he had no lefty reliever for a while, hasn't obsessed over platoon matchups. But Little never seemed to buy into the idea that he has a bullpen assembled without regard to making one guy the closer, and has been trying to brand someone with the Scarlet C since day one.
The constant search for that closer has led him to jump from pitcher to pitcher based on whose last 30 deliveries looked best. His crowning of journeyman Brandon Lyon after a good outing in Tampa Bay during the season's first week was just the most egregious example. Even now, Little is pointing to Robert Person, soon to return from rehab and two years removed from effectiveness, as someone who might be the guy for the Sox in the ninth inning. Little doesn't seem to know who his best relievers are, or how to determine this, and he seems almost diametrically opposed to not having a traditional closer.
I shouldn't blame Little for this entirely; the problem is that he doesn't have a clear-cut "best" reliever, but an assortment of guys who are fairly good and who have had stretches of effectiveness, even dominance, in their past. I love Alan Embree, and I think the Red Sox got him pretty cheaply, but he's not Arthur Rhodes. Chad Fox has been very good when healthy, but he's not been healthy often, nor is he a guy you can have get six outs on a regular basis. Mike Timlin, Ramiro Mendoza...good pitchers, but not an A-level reliever among them.
I didn't see this coming. I was so enamored of the idea that the Red Sox were abandoning the concept of the high-save closer that I ignored that fact that they were missing a key element: The ace reliever who you want to funnel 100 high-leverage innings to, a guy who can handle the workload. Don't get me wrong: Ugueth Urbina wasn't going to be that pitcher, and the decision to let him sign with the Rangers was the right one. But in collecting moderate-cost talent this winter, the Sox neglected to get the one thing that makes any bullpen-closer-free or otherwise-work: a great reliever.
I'm quite annoyed with myself, because in praising the Sox moves this winter I did the thing statheads are always being accused of: letting my bias affect my analysis. Because I approved of what the Red Sox were doing, I didn't look closely enough at how they were doing it, and I missed the hole in the plan.
The Red Sox have two problems right now, buy-in and personnel, and neither is going to be easy to solve. With Little clearly wanting a traditional closer and not selling the front office plan to his players, he's a barrier to implementation of the plan. His players, having seen the bullpen struggle and their manager's resistance, have lined up against the idea of not having a nominal closer. At some point, the Sox front office is either going to have to let Little do it his way or get someone else in place.
The bigger problem, though, the that the Red Sox lack the talent needed to make this work. There's speculation that they'll trade for Scott Williamson, who is just the type of pitcher that they need: dominant and capable of going multiple innings a couple of times a week. Williamson did very well in Jack McKeon's version of this bullpen in 1999, and would be a great addition for the Sox, immediately becoming their ace reliever. If they don't get Williamson, it's not inconceivable that Person, who's had stretches of dominance in the past, could end up being the guy. It'd be nice to see him get a major-league hitter out before he gets handed a high-leverage role, though.
The Red Sox had a great idea this winter, but they executed it poorly. While they're on the right track, between the lack of dugout commitment to the plan and a lack of personnel to execute it, it may be time to concede the point and let Little run a more traditional bullpen. Being right is nice, but it's not worth punting a season over.