July 19, 2011
Kimbrel and Venters: Ticking Time Bombs?
One of the surprise stories in the National League this season has been the success of the Atlanta Braves. They have the third best record in all of baseball and the second best record in the National League despite the struggles of Dan Uggla and Jason Heyward and the injuries to key players such as Martin Prado.
A big part of that success can be credited to the pitching staff that currently has the highest team VORP in the league—a full 19 points over the San Francisco Giants—and the second-highest team FRA—just 0.1 points behind the vaunted Philadelphia Phillies pitching staff. Their actual 3.12 ERA also trails the Phillies, but plate appearances against Braves pitchers end in strikeouts 21 percent of the time, which is tied with the Giants for the best in baseball. As good as Jair Jurrjens, Tommy Hanson, Tim Hudson, and Brandon Beachy have been, the team’s bullpen has been incredible. Lefties Eric O’Flaherty and George Sherrill have helped bridge the gap between the starting rotation and their thunder-and-lightning combination of Craig Kimbrel and Jonny Venters at the back end of the bullpen.
That duo had fewer than 105 innings of major league experience heading into this season, and most of that belonged to Venters since he spent the majority of 2010 with the Braves. So far this season, Kimbrel has a 1.93 SIERA while Venters sits at 2.38, and the duo is arguably the deadliest one-two punch in baseball. That said, they are also one of the most over-worked duos in baseball. Coming out of the break, Venters has pitched in 51 games so far in 2011 while Kimbrel has thrown in 47 contests, giving the due a combined 98 games pitched. If they both continue on their current pace, Kimbrel will appear in 83 games while Venters will appear in 89. Manager Freddi Gonzalez has talked about scaling back the work for the dastardly duo, but in the heat of a playoff race, it is tough to imagine him living up to those words.
Not only are both pitchers on pace to appear in more than 80 games each, both are on pace to work more than 80 innings this season. Kimbrel is on pace for 81 innings of work while Venters is on pace for a whopping 97 innings of work. In today’s day and age of bullpen usage, that is not a common feat. In fact, nobody has done it in the past two seasons, but Venters (surprise!) fell one outing shy of doing so last season.
Many people will point to Joe Torre for his usage of Scott Proctor in 2006 and 2007 as bad usage of a reliever. Torre used him in 115 games before he was shipped off to the Dodgers, where Grady Little used Proctor 31 times and to give him back-to-back seasons of 83 appearances. Another example was Tom Kelly’s fondness for “Everyday” Eddie Guardado, who was used 83 times in a 78-win season in the first of an eight-season run of at least 60 appearances. Every pitcher is different, though. After all, Joe Maddon utilized Randy Choate in 85 contests last season, but Choate has shown no ill effects this season as he has been a very dominating lefty specialist for the Marlins.
The effect of appearances and innings is up for debate, but how does a double dose of those two factors affect pitchers the following season? Since we have two pitchers on pace to hit those measures on the same team—both of which have solid-to-excellent value in both redraft and keeper leagues—let’s take a look at how pitchers that go 80x80 do the season following such arduous work.
Over the past 20 seasons, 42 pitchers have broken the 80x80 barrier in a single season, starting with Juan Agosto in 1990 and ending with Carlos Marmol in 2008. I focus on the last 20 seasons since bullpen usage has become more specialized over the past twenty or so years compared to the early 80’s and prior, when relievers would routinely go two to three innings. The sample size is admittedly small due to those constraints, but even expanding the date range back to post-WWII only increases the sample size by another 30 pitcher seasons.
The graphs below show how the 80x80 members performed, as a whole, the season in which they made the 80x80 club along with how they performed the season following that accomplishment. The graphs also show a control group of relievers with “normal” workloads. I’ve defined “normal” as pitchers who threw between 50 and 70 innings out of the bullpen in a given season.
By Games Pitched
Only three pitchers—Steve Kline (2000), Paul Quantrill (2001), and Jon Rauch (2006)—saw their appearances go up the following season. 16 other pitchers saw their appearances decline by at least 20 games or more with Peter Moylan and Oscar Villareal missing significant chunks of the following season with elbow injuries. However, while the 80x80 group saw an 18 percent reduction in their appearances the following season, the Normal group saw a 20 percent reduction in appearances.
By Innings Pitched
Only Salomon Torres (2004) saw an increase in innings after his 80x80 season and, at that, it was an increase of just two and two-thirds innings. Torres was unable to repeat the feat in 2007, however, as he threw 41 fewer innings that year following his final 80x80 season, joining Billy Koch, Greg H. Harris, Proctor, Villareal, and Moylan as pitchers who saw their workload drop by at least 50 percent due to injury or ineffectiveness. In all, 17 different pitchers saw their innings total drop by at least 24 innings the season after working over 80 innings. On the whole, the 80x80 group saw a 24 percent reduction in innings the following season while the normal group saw just an 18 percent reduction in workload.
Obviously, there are many influences on a pitcher’s ERA from one season to the next, and our sample size may not be large enough to drown out all the noise. Brad Clontz actually saw his ERA improve by nearly two full runs, one of 16 different pitchers who saw their ERA improve by at least one full run after going 80x80. Villareal, Harris, and J.C. Romero, however, all saw their ERA’s worsen by three runs or more. 80x80 pitchers, as a group, saw their ERA increase by 89 percent while the Normal group increased by just 73 percent.
Kline, Swindell, Capps, Rauch, and Turk Wendell are the only five pitchers who saw an improvement in their WHIP a season after going 80x80. Stan Belinda, Quantrill, and Torres (twice) saw no change, while everyone else suffered a drop-off with Romero, Marmol, and Aaron Heilman suffering some of the worst damage. The difference in group WHIP changes is much more noticeable than group ERA changes as the 80x80 group saw a 16 percent spike in their WHIP compared to just a 7 percent spike for the Normal group.
By Strikeout to Walk Ratio
14 pitchers here actually improved the season following the high workloads with Chad Qualls and Jon Rauch showing the most improvement. 28 pitchers, however, showed a decline in their command of strike zone (as measured by K/BB) led by Quantrill (who went from a 4.8 ratio in 2001 to a 2.1 ratio in 2002). The Normal group was able to nearly repeat their strikeout to walk ratio in the following season while the 80x80 team was much less successful
By Strikeout Rate
Surprisingly, 19 pitchers saw an improvement in their strikeout rate following an 80x80 season with Proctor leading the way; his strikeout rate jumped four full points from 2007 to 2008 despite the rest of his line declining badly. 23 pitchers saw their rates decline with Belinda, Rod Beck, Mike Timlin, and Felix Rodriguez all suffering a drop-off of two or more strikeouts per nine innings. Interestingly, the 80x80 group suffered a drop-off only half as severe as the control group in this metric.