Marlins bullpen management in 2011

facebooktwitterreddit

Now that the Marlins have a bullpen that is set for the 2011 season, I wanted to take a quick look at the potential alignment that the team’s pen will go with. As you may recall, bullpen management was a bit of an issue in the past for the Marlins when under the former regime of Fredi Gonzalez. Will Edwin Rodriguez do any better?

The Cast

As of right now, there are eight Marlins bullpen pieces that are worthy of consistent innings. The four (mostly) major leaguers that the Marlins acquired (Ryan Webb, Edward Mujica, Mike Dunn, and Randy Choate) are undoubtedly going to get playing time. In addition to those four, the Marlins have four holdovers in Leo Nunez, Clay Hensley, Burke Badenhop, and Brian Sanches who have earned playing time as well. How will the team dole out these innings, and will they effectively leverage these pitchers given their true talent levels? Well, to find out, I created a rough estimate for expected LI for the eight guys based on their expected roles and their order on the official Florida Marlins website’s depth charts. The following table contains what I came up with along with each player’s projected ERA+ according to ZiPS.

The Performance

PitcherRoleExp LIProj ERA+
NunezCloser1.7115
HensleySetup/8th1.3115
Webb*7th/8th1.098
DunnLOOGY/7th0.992
ChoateLOOGY1.1106
Mujica*7th/8th0.9105
Sanches7th0.7122
Badenhop7th/Long Relief0.7109

*Webb’s and Mujica’s projections are park-adjusted but also based on projections for their performance in Petco Park.

Based on my guesses, you can already see some potential mistakes by the Marlins managerial staff. Webb, according to ZiPS projections, would be a bit worse than league average yet be garnering one of the highest leverage indices (and likely highest innings pitched totals) on the team. The same goes for Dunn, who has an electric fastball but has yet to prove he can control it. Meanwhile, based on the depth chart, the club would be wasting Sanches and Badenhop in lower-leverage roles in the 7th inning despite their skill.

If you plot a chart of each reliever’s expected LI and their ERA+ (I did ERA+ relative to the team average, calculated at 106), you would get a chart that looks like this:

The R-squared value listed there is at 0.04. In and of itself, that doesn’t mean much, but it does tell you that, given ZiPS’ projections, this is not a very efficient use of the team’s available pen. Hensley and Nunez appear to be used appropriately, but no one else looks particularly good. Ideally, the team would have a nice linear relationship, with the best pitchers pitching the highest leverage and the worst pitchers pitching the lowest available.

I rearranged the pen and the innings distribution to get a potentially better bullpen setup for the Fish given my own opinions on the roster. This is what I got:

It still isn’t the best thing in the world, but the R-squared of 0.36 probably indicates that this is a better fit given the talent involved. I moved Dunn and Webb into the lower leverage roles and moved Badenhop into the 8th inning and Sanches up a notch into the 0.9 leverage 7th/8th inning spot. This improved the fit, though it still showed Sanches as highly underutilized.

Peculiarities and Restrictions

There are some oddities and restrictions involved in this setup however. One is the possible necessity of a long reliever. The team only has one reliever who can be effectively stretched out to work 3+ inning stints, and that is unfortunately Badenhop, who appears to be one of the best relievers on the team. I have heard that a possible alternative would be to stretch out Sanches, who is likely a bit worse than Badenhop (career xFIP for Badenhop at 3.90 versus 4.86 for Sanches), but according to these projections, that would still keep one of the better pen arms away from higher leverage innings.

On the topic of Sanches, I feel as if his skill is being overstated here. An ERA+ of 122 seems awfully high for a guy who does not get the ball on the ground (career 33.4% GB%), isn’t overwhelming with his strikeouts (20.2% K% career) and walks an average number of guys (11.5% BB% career). All of Sanches’ numbers point to him being more of a journeyman than someone who should help anchor the bullpen. So far, Sanches appears to be skating by with absurdly low home run rates, but who knows how long those can last?

What about the wins?

Ultimately, the game is about wins, and the way we can estimate the win impact of this bullpen is with WAR. In order to get the team’s bullpen leverage as a whole at around 1.00, I forced in 50 IP of performance at a leverage of 0.5; I arbitrarily gave those 50 IP an ERA+ of 90, to simulate the fact that those pitchers would not be very good. Given the ERA+ here and the fact that ZiPS projecs a 4.14 ERA in the NL in 2011, we can calculate RA for each reliever and get WAR based on their innings and LI. The total the Marlins would run out there at the given innings is 4.0 WAR. As a comparison, Marlins relievers were worth 2.3 WAR last season pitching the same number of innings with a FIP similar to their reliever ERA. If you buy these ERA projections as true-talent estimates, then you might reasonably expect the Marlins to gain two wins after all of this. All in all, that would be a definite plus for the Fish.