Building A Better Bullpen
Daniel McCutchen was allowed to labor through 92 pitches while the Pirates' best reliever sat alone in the bullpen and umpire Meals broke the spirit of Buccos fans with his infamous safe call.
The defeat was one of three extra-innings losses the Pirates suffered on that road trip, and Hanrahan did not pitch in any of them. The Hammer was left in the toolbox for save opportunities that never arrived. Then the Bucs plummeted from division contenders to an 18-41 finish. Hanrahan’s continued usage primarily in those save spots is an example of an in-game tactical mistake which could be costing the team wins.
Dan Fox thinks the save stat is an inefficiency to exploit, meaning other teams "play to the save" too often. He should know. As the Pirates' director of baseball systems development and a former writer for the stats-minded Baseball Prospectus, Fox is one of the game's most respected numbers gurus.
"Teams don't always utilize their best pitchers in the highest-leverage situations," Fox said. "Probably as an industry, there is room for improvement."
Fox knows the ins and outs of Leverage Index, which quantifies each situation of a game as it affects the game’s outcome. He and other stats authorities have proven, for example, that pitching in the 8th inning with runners on and your team leading by only one run is a more pressing situation than starting the 9th inning with a two-run lead (a typical save situation). After all, a hitter with runners on in the 8th can put his team ahead with one swing.
But under manager Clint Hurdle, Hanrahan has pitched in 58 home games and only 41 away games; closers on the road are often kept in the bullpen until a save opportunity arises. Worse, last year McCutchen, Veras and Resop all made more appearances and pitched more innings than Hanrahan. In what other sport are lesser players utilized more often than better players at the same position?
"There's always a question of when do you use [Hanrahan] in extra innings. That's a work in progress," Hurdle said. "The one thing statistical analysis never can take into play is the human element."
Ah yes, the human element. Hurdle seems to subscribe to "closer mentality," the thought that some pitchers don't perform well outside a defined role. Fox agrees to an extent, but believes pitchers can change their perception of roles.
"If the pitchers themselves think the last three outs are more difficult…then it actually will be," Fox said. "It's going to be a self-fulfilling prophecy.”
One of the most vocal critics of the save-centric bullpen model is Joe Sheehan, a baseball writer for Sports Illustrated and former Prospectus colleague of Fox.
"The state of managing from 7 p.m. to 10 p.m. is as bad as it has ever been," Sheehan said. "There's very little original thought. There's a desperate attempt to win the press conference as opposed to winning the game…It is a fear of 'Oh my God, we're going to take a lead and not have our closer.'"
Sheehan denounces Hurdle as a "fairly conservative, routine, I-run-my-bullpen-like-everyone-else" manager. But Hurdle does receive background statistical information from Fox. The second-year Pirates skipper also recognizes sabermetric analysis is important, even talking about leverage as well as the statheads do.
"You want guys with certain skill sets to pitch to the middle of the order," Hurdle said about his tactics. "And sometimes it's not just a hard thrower, sometimes it's a guy with a good sinker... watching hitters' swings and matching it up with the skill sets of the pitchers."
Yet missteps like deploying Hanrahan less often than Resop remain. Hurdle freely admits his focus is on cementing the 8th and 9th innings. This season, his bullpen is a talented group, and has delivered one of the best earned run averages in baseball. But Sheehan believes most teams (including the Pirates) still aren’t optimizing reliever usage, thus leaving wins on the table.
"It's all high-leverage stuff late in the game," Sheehan said. "If I correctly use my bullpen, I can swing a game from a loss to a win.”
And how much would correct usage be worth to a team?
“Probably worth 3 to 7 wins per year."
That is a huge number. Analysts value an All-Star player at about 5 wins better than a typical Triple-A callup. And after four teams saw the fates of their 2011 seasons sealed only on the season's final day, fans know how precious each win is.
So how does a manager subvert save culture and closer mentality, and make decisions to help his team as much as an All-Star player does? Sheehan emphasizes two particular points that run contrary to modern bullpen strategy.
1. The 9th inning is not always the most important inning.
In 2006, Tom Tango literally wrote "The Book" on leverage. One of his conclusions: a mediocre reliever will only blow about one more three-run-save opportunity over a full season than a lights-out closer would.
"You don't actually need the closer to be able to finish the game," Fox said. "All they need to do is extinguish the rally, stabilize the leverage, and then you can go to a lesser pitcher to pick up the cheap save."
2. Great relievers can pitch more than one inning per game.
This seems simple, but it’s not widely implemented. Craig Kimbrel, Jonathan Papelbon, Hanrahan, Mariano Rivera and Ryan Madson were the top five closers in baseball last season. One other feature they have in common? They can all count on one hand the number of times they pitched more than one inning in 2011.
If these pitchers are the best of the bullpen best, they can surely shut down the opposing lineup for two innings, not just one.
Let me add one more to Sheehan's points:
3. Trust your pitchers -- they can handle new roles.
I asked a couple Pirates relievers, do you prepare differently in the bullpen based on the in-game situation?
Tony Watson: "No, it all comes back to staying aggressive… in the strike zone and getting ahead of the hitters."
And the closer himself, Hanrahan: "No, you have to go out there with the same mindset of putting up a zero. Any time you come into a game, that's your goal."
Most importantly, each pitcher said his main focus is on winning baseball games.
"Personal numbers are a side number," Jared Hughes said. "That's not what matters. What matters is the team winning."
For players, a brave new bullpen would require reassurance in their own abilities. Young pitchers like Hughes have the confidence to collect big outs in the 7th. The 9th is not too different, and such a mindset can start in the minor leagues.
For managers, bullpen optimization will require identifying assets such as the best pitchers to strand runners on base or relievers to build defensive shifts around. And as Sheehan hammers home, if Hanrahan and other closers possess the mental toughness to complete the final three outs, why not completing other outs?
Optimal reliever usage could easily be worth an extra two or three wins for the Pirates this summer, enough to shift a tightly-contested NL Central race. And it can be done without spending a dime, a paramount goal for small-market teams.
Fox has supplied Hurdle with the proper data, the numbers that would allow the manager to play the bullpen chess game like a grandmaster. If he does, it would give Buccos fans a better chance to replace nightmarish Jerry Meals memories with the excitement of a pennant race.