Barring a late-spring signing of Jose Valverde, the Tigers will be turning to a pitcher new to the closer’s role (or, in the case of Octavio Dotel, many years removed from his last closing opportunity). Things have been rough in the bullpen in spring training, particularly for the assumed front-runner Bruce Rondon, who has allowed five hits (including a home run) and five walks in just four appearances to date.
The competition appears now to be a bit more wide open. Joaquin Benoit, Al Alburquerque, Phil Coke and Octavio Dotel join Rondon as options for the ninth inning in Detroit. Who fits best? To help answer that question, I took a look at what characterized the most successful pitchers to move from a setup role (or other bullpen role) into a closer role the next season.
In a research piece for FanGraphs+, I found that relievers who typically make a 20+ save jump from the previous year exhibit two characteristics in particular: a high strikeout percentage and high fastball velocity. I don’t think this should be a terribly shocking discovery; managers have always had a preference for gas in the ninth inning since the advent of the closer. The piece also made a few other conclusions, most notably that walk rate was a poor predictor of who picked up saves (perhaps a good sign for Rondon or Alburquerque, both owners of walk rates over 15 percent last season).
But this was a fantasy-slanted piece, and so it was less concerned with the effectiveness of these relievers and more concerned with simply whether or not they picked up saves. After all, there have been 40 seasons in which a pitcher has recorded at least 15 saves with a 5.00 ERA or worse; save totals don’t necessarily reflect effective pitching.
For this exercise, I looked at the 48 pitchers to go from five or fewer saves in at least 50 innings pitched one year to 15 or more saves the next from 2003 to 2012, and I looked at K%, BB% and fastball velocity as potentially predictive statistics. Were they?
Observe, those 48 newly-minted closers ordered by BB%:
The above chart shows frequency of previous year walk rate among these 48; three pitchers had a walk rate between four and six percent; 10 had a walk rate between six and eight percent, et cetera. The shades in the bar represent the ERA- (top) or FIP- of the pitcher. This can be viewed as a heatmap of sorts; if one side of the graph exhibits particularly dark shades, it suggests more successful pitchers come from that extreme.
In the case of walk rate, there isn’t much differentiation. The bins from a four percent walk rate through a 12 percent walk rate all have an average ERA- between 63 and 81; the average FIP- marks fall between 73 and 88. Some of the best pitchers — such as Aroldis Chapman, who walked over 20% of batters the year before he ascended to Cincinnati’s closer role — have high walk rates. I would assume there is some selection bias in here, as a manager won’t put a high-walk pitcher in a high leverage role (or possibly even on the team) if he doesn’t have a blazing fastball.
That’s good news for Rondon and Alburquerque, who both fit the poor control but great stuff model that has a decent history of success as closer.
Strikeout rate is clearly one of the most important selection criteria for closers — a large majority of new closers had a strikeout rate over 21 percent the year before assuming ninth-inning duties. Phil Coke is the only one of the Tigers’ potential closers to post a strikeout rate under 21 percent in 2012.
Unsurprisingly, the high-strikeout pitchers have been significantly better as closers. The dark shades cluster towards the right of the chart. Not a single bin above 24 percent strikeouts averaged an ERA- over 73, whereas three of the four under 24 percent did (a similar but less extreme pattern holds for FIP-). This is the same reason Carlos Marmol was able to have a relatively successful run as closer despite having no idea where his pitches were going — even once he puts runners on base, the opposition eventually has to make contact to move them around.
Alburquerque owns a tremendous 36.2 percent career strikeout rate over parts of two seasons. Benoit struck out over 29 percent of hitters in 2012. Rondon struck out over 24 percent of hitters at every level in the minor leagues. All four of the non-Coke options should be above-average relievers in terms of whiffs next season.
We already knew a pitcher almost always needs plus velocity to earn a shot at the closer’s role — just two of the 48 pitchers in our sample had a fastball velocity under 90 MPH the previous year. But does truly high-end velocity help?
It appears it does. Although the samples are small at the extremes, looking at the four most common velocities suggests more speed means more outs. The 91-92 MPH velocity pitchers averaged an 80 ERA- while the 93-94 MPH velocity pitchers averaged a far superior 65 ERA-.
Again, Rondon (no hard data, as he didn’t pitch in the majors, but he can touch the upper 90s) and Alburquerque (94.7 FBv), despite their lack of control, exhibit great velocity and that could be enough to make them effective closers. Benoit also brings heat at 93.7 MPH on average. Coke and Dotel bring up the rear at 93.1 and 92.8 MPH respectively.
Rondon, Albuquerque and Benoit fit the new closer bill exceptionally well in terms of velocity and strikeout rates, with Benoit and Albuquerque ahead of the pack. Although it’s entirely possible one of the three (or five) fails upon promotion to the ninth inning, it seems unlikely all three would collapse under the bright lights of the closer’s role.
So do the Tigers need another elite reliever? It always depends on the cost, but the club has at least two and possibly three relievers who have a historically successful profile for moving from setting up to closing. If the Tigers can get them leads — and their roster looks like it should have no problem doing so — Jim Leyland should be confident his relievers can save games.