The era of Aroldis Chapman closing for the Reds is here. Like all Reds fans, I thoroughly enjoyed watching Chapman secure the last three outs today against the Yankees. But I can’t abide by the discouraging decision that put him there.
It’s a short-sighted move and borne from disregard or ignorance of modern analytics.
1. Sean Marshall has been great this year (really). Sean Marshall is pitching more than well enough to keep his job as closer. He has the thirteenth highest strikeout rate (13.19 K/9) and the ninth best strikeout-to-walk ratio (7.0 K/BB) in all baseball. That’s this year and doesn’t take into account blowing away Robinson Cano on three pitches today. That’s elite stuff. His track record at getting important outs is longer than anyone else’s in the bullpen.
When you remove luck, Marshall and Chapman have pitched equally well this year. Chapman has a better K/9 but Marshall’s rate is also extraordinary. Marshall has a lower walk rate (1.88 vs. 2.95) and superior ground ball percentage (60% vs. 41%) than Chapman.
The main difference between the two has been luck. Advanced metrics offers statistics that measure luck – Batting Average on Balls in Play (BABIP), home runs as a percentage of fly balls (HR/FB%) and stranded runners (LOB%) are three. Marshall has the second highest BABIP of all 349 pitchers in the major leagues. His HR/FB% and LOB% are also both extremely high.
What’s important for fans (and organizations) to understand is that all three of these factors are largely out of the pitcher’s control.  One statistic that evaluates pitchers and removes these luck factors is called “expected fielding independent pitching” (xFIP) which is calibrated to the same scale as traditional ERA.
xFIP – Chapman (1.43), Marshall (1.53).
Sean Marshall’s xFIP is lower than any other current closer in the major leagues.
If an organization is strongly committed to incorporating modern analytics into their decision-making, they would evaluate pitchers based on the parts of their performance they can control, not luck.
2. Starters >> Closers. Starters pitch 180-200 innings per year instead of 60-70. Closers enter many games with two or three runs leads, sometimes against the weakest part of the other team’s order. In those situations, research shows almost any pitcher can be effective.
My favorite statistic that demonstrates the relative unimportance of closers concerns Mariano Rivera. It’s pretty much uncontested that Rivera is the greatest closer of all time. From 1997, when Rivera became the Yankees closer, to 2008, the Yankees won 97.3% (!) of all games they entered the ninth inning with a lead. So obviously, having a lock-down closer makes a huge difference. Right?
Consider this: From 1951-62 (another 11-year period of dominance for the Yankees, in the pre-closer era, when many pitchers finished games for teams), the Yankees winning percentage when they entered the ninth inning with a lead — 97.3%.
Surprising, but revealing.
The Reds have other pitchers besides Chapman who could be the closer. Chapman has pitched the best so far, but both Logan Ondrusek and Jose Arredondo have pitched brilliantly all year. They are also more accomplished than Chapman from the standpoint of pitching multiple days in a row.
Closers — at their best — simply don’t help a team nearly as much as a starting pitcher. In his 17 seasons with the Yankees, Rivera averaged 2.3 Wins Above Replacement (WAR). Drew Stubbs and Mat Latos have averaged more than that. Rivera has posted four seasons with 3 or more WAR. Mat Latos has already achieved that twice.
The move to the closer role dramatically reduces Chapman’s value to the team. He’ll pitch mostly when the team is already ahead, in situations where other pitchers could essentially fill the same role. Modern analytics again, make this clear.
3. Mismanaging Chapman, Part III. Moving Chapman to the closer role is probably the only decision that could actually worsen the organization’s mismanagement of Chapman’s talent.
As Dusty Baker’s closer, Chapman will pitch even fewer innings than he does now. So far, Sean Marshall has been used less than every other member of the bullpen, including Alfredo Simon. Last year, Coco Cordero threw 69 innings. Travis Wood, Edinson Volquez, even Dontrelle Willis threw more.
The move solidifies Chapman’s role in the bullpen, unfortunately moving him farther away from joining the starting rotation. At the start of next year he’ll be half way through his 6-year contract with the Reds having contributed a total of only 130 innings. To make matters worse, if the Reds do move him to the rotation in 2013, he’ll be on an innings limit.
—–
Moving Marshall out of the closer role after only fourteen, largely successful innings, gives off a whiff of panic that is surprising for Dusty Baker, who usually sticks with his closers through thick and thin. What happened to the manager who didn’t waver an inch in his support for Cordero even once in four years?
Maybe Baker is simply seizing a way to cement Chapman’s presence in the bullpen. That’s a struggle he’s waged with GM Walt Jocketty for at least a year.
Installing Aroldis Chapman as the closer is a rash decision based on obsolete metrics and flawed reasoning. It’s the opposite of smart, well-informed baseball decision-making. Use whatever adjectives you want for that.
My friend Mike Maffie, who helped me with this post, points out: As the thinking of baseball moves forward, the Reds move backward. Dustyball is the opposite of Moneyball.
The movie ‘Dustyball’ would undoubtedly be shot with a Super 8 camera on black-and-white film.