Okay, you’re right. Ryan Madson ought not to be closing games for the Phillies. However, it has nothing to do with mentality, or toughness, or any of the intangible reasons that have kept the Phillies’ best reliever out of the bullpen’s most glamorous spot for three years. Madson is the Phillies’ best pitcher out of the bullpen, by far, to the point where any argument to the contrary is really not worth discussing. So what should he do, if not close? Something far more important.
As an aside, this is hardly new ground being broken here; people smarter and more knowledgeable than I have been calling for the abolition of the closer role as such for more than a decade. This is just a different take on that argument.
Major league managers use their top relief pitchers in one of these three situations:
1) For the ninth inning, and the ninth inning only, with a lead of one to three runs.
2) For one inning, and one inning only, in the ninth or tenth inning of a tied game.
3) In a four-or-five-run lead, or down by one run, if he’s already warmed up and hasn’t pitched in a while (I came to know this as the “Ricky Bottalico situation”).
This usage is dictated almost entirely by the save rule, the benchmark by which relief pitchers are judged. A manager adjusting strategy to benefit individual statistics is a practice that cannot be condemned strongly enough–imagine if a football coach only called screen passes for no reason other than a fear of lowering his quarterback’s completion percentage.
I’ll once again quote Bill James: “If you use your relief ace to save a three-run lead in the ninth inning, you’ll win that game 99% of the time. If you don’t use your ace in that situation, you’ll win 98% of the time…an average team would win 97% of those games if they brought in Bryan Rekar in that situation.”
I bring this up because I was at a game three weeks ago in Pittsburgh that the Phillies lost, despite eight innings of one-hit, one-run ball from Cole Hamels, because they brought in Danys Baez in the 12th inning with Madson still out there in the bullpen. Why didn’t Madson pitch? Because managers adhere to the rule, almost as if it were codified somewhere, that you don’t use your closer in a tie game on the road.
So if closer use ought not to be tied to the save rule, what else should managers use? The answer is high-leverage situations. Leverage index is a statistic that measures how much impact a given situation will have on the game outcome. I think we can all agree that a bases-loaded, two-out situation in the ninth inning of a one-run game has more impact on whether a team wins or loses than a bases-empty, no-out situation in the early innings of an 11-1 game. Leverage index is an attempt to quantify that effect.
Because high-leverage situations are so important, you’d want your best pitcher to have a high average leverage index when he enters the game, or gmLI. And if you look at the gmLI leaderboard, closers dominate the list, because even though closers aren’t being used optimally, they still get into the game later in closer games than any other pitcher on the staff.
The best way to maximize the effectiveness of relievers–and if you had Ryan Madson, wouldn’t you rather he be pitching as many important innings as possible?–is not to be afraid of using them 1) before the ninth inning 2) for more than one inning and 3) in tie games. Addressing those points one-by-one: 1) if the Phillies are playing the Rockies, and the Phillies have a one-run lead in the eighth with the bases loaded and Tulowitzki and Gonzalez coming up, wouldn’t you rather bring in Madson, rather than Jose Contreras? Let the second-tier relievers come in later when no one’s on base and weaker hitters are up. You want your best pitcher facing the opponent’s best hitters in the most important situations, regardless of inning, or at least you should. 2) It’s not uncommon for good setup men to go multiple innings–Mike Stutes went an inning and a third in the game I referenced earlier–so why can’t they continue to do so when they become closers? 3) If a pitcher comes in in a save situation–up two runs in the ninth–and gives up a run, you still win. If you give up a run in a tie game, you’re losing. Bring in your best pitcher to lessen the possibility of giving up that one run.
Anyway, like I said, that’s hardly a new argument, and it’s been articulated better elsewhere, including in the Bill James essay I quoted. I just wanted y’all to think about it the next time the Phillies lose a close game without using Madson.
While we’re on the subject of pitcher usage, let’s touch on what’s been a controversial topic in Philly recently: pitch counts for starting pitchers. The Phillies are using their starting pitchers a lot. Right now, there’s a debate, with people whose opinions I respect on both sides of it, on whether the current usage rates for Cliff Lee, Roy Halladay, and Cole Hamels in particular will lead to fatigue or long-term health issues for one or more of the Phillies’ top three starters.
I don’t know for sure whether or not the Phillies are overworking their starting pitchers, because I don’t know with any certainty, how many pitches per start is too many. Given the fact that elbows and shoulders get blown out on major league mounds all the time, I suspect no one really knows for sure. Here’s what I do know:
1) There is a limit to how many pitches someone can throw without hurting himself, and it varies from pitcher to pitcher, and depends on any number of factors.
2) That limit is not 100 pitches. We think of 100 pitches as a limit because it’s in the ballpark of where most starters run out of gas, and it’s a round number.
3) Not only do the limits of endurance vary from pitcher to pitcher, they vary from start to start for a single pitcher, based on how hard he works. That depends on what kind of pitches he throws, how stressful the game, and individual innings, are, and even things like how hot it is outside or how hard the wind is blowing.
4) Overworking young pitchers tends to take a toll on them in terms of long-term health that older pitchers seem not to suffer. My choir director in college said something like that about forcing high notes.
5) I don’t think Halladay, Hamels, and Lee are anything approaching honest with Charlie Manuel and Rich Dubee regarding their own fatigue. I think all three want the ball all the time and wouldn’t admit to being tired within a game. With that said, I think a manager has a better handle on how tired his pitchers are than I do.
Henry Abbott of TrueHoop said that, in basketball, which is undergoing its own statistical revolution, the next great advance will not be in statistics, but in athletic training–we’ll collect, understand, and apply new volumes of biometric information to improve efficiency and endurance and prevent injury. In short, the academics who inspire the next Moneyball will not be statisticians and economists, but doctors and biomedical engineers.
If that’s true for baseball, as well, the next generation of scientific study of sports will likely yield the answer to this question of whether Cliff Lee throwing 113 or more pitches in 9 of his last 11 starts and 120 or more in three of his last eight starts will cause his arm to fall off. I certainly don’t have that answer now, but if anyone knows more, by all means, I’m eager to learn.