There's been a bit of talk recently about bullpen roles, what with Jake McGee's recent dominance and the game tying home run just given up by Joel Peralta. A few weeks ago, Bill Baer on Crashburn Alley had a fantastic little breakdown of Charlie Manual's bullpen usage by leverage, so I thought now would be a good time to do the same for the Rays. I'm merely copying his template.
Leverage Index (LI) is a measure of how important a specific moment of a game is. It shows how the win expectancy can swing based on the outcome of a single plate appearance. Average leverage is one, with numbers less than one meaning below average importance, and numbers greater than one meaning above average importance. Over the course of a season, a starting pitcher's leverage Index will be close to average, but a relief pitchers's Leverage Index can vary much more. It's up to the manager to figure out how to get his best relievers into the game at the most important moments.
Because it's easy to mislead when using percentages, I'm including a table of the raw numbers. These are the number of plate appearances each pitcher has had in each LI range. Feel free to skip this table. More colorful ones are below the jump. Click on any table to make it bigger.
Now here's how many plate appearance each reliever has pitched in each LI range as a percentage of his total plate appearances.
I see three groups here. Rodney, Peralta, Gomes (who's not here anymore, but had high leverage appearances when he was), and McGee form the high leverage group and make appearances in situations with below avearge importance less than 60% of the time. Davis, Howell, and Badenhop form a middle group. Joe Maddon will allow them to pitch in important spots if the matchup and situation dictate, otherwise they have less put on their shoulders. Ramos, Lueke, and De La Rosa are purely for mop up duty right now.
Now here is how each range of Leverage Index is being split up among the Rays' relievers.
I know that having a closer in Fernando Rodney (even if he's only the de facto closer) isn't the optimal way to manage a bullpen in terms of splitting up PAs, but it's working so far. Rodney has taken the majority of plate appearances with a leverage index over 4. Peralta has gotten most of the rest, in his job as eighth inning guy and occasional fireman.
I can see the argument for giving McGee more of Peralta's high leverage spots (and this will all change when Farnsworth returns), but I don't think the system as it sits now is broken. And there's probably some truth to relievers performing better when they have a role that they can mentally embrace (studies have shown that closers perform worse in non-save situations). By not naming a closer or other set roles, Maddon gives himself the flexibility to use his bullpen well, but by sticking to those unnamed roles when he can, he may improve the mental state of his relievers.
All data is from Fangraphs.