Why Relievers Are Not Worth The Risk

There are a lot of baseball fans and pundits alike who are curious as to why the number one relief pitcher on the free agent market, Rafael Soriano, has not yet found a new team for next season. Optimistically, I’d like to believe it is because teams have learned that long contracts for relievers are bad investments that rarely work out as intended, complicated by the fact that the relievers seeking such deals often require draft pick compensation, as Soriano does.

While a number of teams have proven my optimism misguided by doling out two and three year deals like candy (Benoit, Downs, Feliciano, Crain, Putz, etc.), teams have yet to seriously bite on the big fish in Soriano. Whether that is because they draw the line at four or five years but not three, or because they have reservations about the amount of money he is likely to cost, it is the right decision to avoid him. Why? Simply put, it’s because relievers are generally unpredictable and of limited value at best.

Their unpredictability is clear; in a study of the 26 relievers who pitched at least 50 innings every season from 2007-10, their WAR dropped, on average, 36.25% from season to season. In the same timeframe, position players (min.500 PA) saw their WAR increase by about 15% on average, and starting pitchers (min.160 IP) saw it increase by 7.98%.

Now to look at the actual value of the relievers just furthers the point. The 70^th highest WAR from a position player in 2010 was from Hunter Pence, who posted a 3.1 WAR. The 47^th highest WAR from a starting pitcher was by Dallas Braden, who put up a solid 3.0 WAR. The highest WAR from a reliever was from Carlos Marmol, and he pitched to a 3.1 WAR. The reason that relievers aren’t valuable like other pitchers has nothing to do with how talented they are. Many relievers have top notched stuff, control, or both (see Rivera, Mariano). The problem with relievers is that they do not pitch often enough to be really valuable. A top starter will pitch in excess of 200 innings in a single season, and a top reliever might pitch 80 or 90 innings. On a smaller scale, it’s like comparing a regular pinch hitter to a starting position player. The value is limited because of the lack of opportunity, and the smaller sample size leads to greater variability in performance.

In order for a reliever to be really worth the money, years and draft picks they sometimes require, he would need to be able to sustain effectiveness while frequently pitching two or even three innings at a time for a full season. Since the structure of bullpens today dictates that the best pitcher acts as the closer, only pitching the 9^th inning and rarely more than one inning at a time anyway, the notion of a star reliever being worth close to as much as a star position player or starter is somewhat of a pipe dream.