Last season, my friend who's a Mets fan was talking to be about how the Jerry Manuel was using Pedro Feliciano way too much. It got me thinking. What are the effects of overworking a reliever? Do they flame out within a few years? How many seasons can they last while appearing in so many games? Let's find out.

The easiest way to find overworked relievers is to look at the single-season records for most appearances by a pitcher in one season (I used baseball-reference here). Feliciano is unsurprisingly number 4 on that list. Numbers 1-3 are Mike Marshall, Kent Tekulve, and Salomon Torres. Almost shockingly, all 3 of them are righties.

Marshall was a righty closer primarily in the 70's for 10 different teams. He debuted as a closer for the Montreal Expos in 1971, appearing in 66 games with a 5-8 record, saving 23 and striking out 85 in 111.1 IP. He had appeared in just 28 games the year before (5 starts). You would think that he would have worn down as the season went on, but the exact opposite occurred. He went 1-5 with a 6.30 ERA in the first half, and 4-3 with a 2.64 ERA in the second half. In 1972, he appeared in 65 games, but he did much better, going 14-8 out of the bullpen with a 1.78 ERA, 95 K's, and 18 saves in 116 IP. He was so good that the Expos put him in for 92 games in 1973, and he did not disappoint, going 14-11 while posting a 2.66 ERA, 124 K's and 31 saves (tops in the league) in 170 IP. After the season, the Expos traded him Marshall to the Dodgers. They must have thought that he would flame out after appearing in 92 games. At least in the short-term, they were wrong. In 1974 for the Dodgers, he set the record by appearing in an incredible 106 games, going 15-12 with a 2.42 ERA, 143 K's, and 21 saves (surprisingly tops in the league) in 208.1 IP. Yeah, over 200 IP out of the bullpen. In the seasons following 1974, Marshall averaged just 44 appearances per season, although he did appear in 90 games in 1979 at age 36. From the beginning of his career in 1967 to 1974, Marshall posted a 3.01 ERA. From 1975 until the end of his career in 1981, Marshall posted a 3.34 ERA. But, you can divide that 3.34 ERA into a 3.81 ERA in the 4 seasons after 1974 and a 2.95 ERA after that. Marshall apparently shows that a reliever can't keep up the 90+ games per season rate for too long and without being less durable and less effective for at least a few years after.

Tekulve was a righty closer for the Pirates, Phillies, and Reds from the mid-70's to the late 80's. After posting a 2.65 ERA and an average of 57 appearances per season in his first 3 major league seasons, Tekulve appeared in 91 games for the Pirates in 1978, going 8-7 with a 2.33 ERA, 77 K's, and 31 saves in 135.1 IP. He then appeared in 94 games for Pittsburgh in 1979, going 10-8 with a 2.75 ERA, 75 K's, and 31 saves again in 134.1 IP. After 1979, in the 10 seasons left in his career, Tekulve posted a 2.97 ERA and averaged 69 appearances per season. He even appeared in 85 games in '82 and 90 games in '87. He posted a 2.65 ERA from the beginning of his career to 1978. Tekulve didn't really wear down or become ineffective after appearing in 90 games in consecutive seasons. But is he the exception that proves the rule?

Torres was a righty middle reliever for 5 teams in the 90's and 00's (the decade of 2001 to 2010). He only saved 57 games his entire career. He posted a 4.48 ERA and an average of 31 games per season from 1993 from 2005 (he didn't appear in any games in the majors from 1998 to 2001 because he retired). In 2006, Torres went 3-6 with a 3.28 ERA, 72 K's, and 12 K's in 94 relief appearances and 93.1 IP. He posted a 5.47 ERA in 56 games in '07, and a 3.49 ERA in 71 games in 2008, his final season. He definitely lacked durability and effectiveness in '07, but he had his finest season in '08, saving 28 games in 71 appearances. Torres is in a different category than Marshall and Tekulve because he averaged many less innings per appearance than them primarily because he played more recently. Still, Torres does correspond to the hypothesis that appearing in 90 games causes relievers to be less durable and less effective, albeit to a lesser extent.

Other notable pitchers up there on the list include Paul Quantrill, Jon Rauch, Matt Capps, Scott Proctor, Heath Bell, and Mike Gonzalez.

Quantrill , another righty, is tied for 10th with 89 appearances in 2003 in addition to being tied for 22nd (two separate seasons), 61st, 96th, and 191st from other seasons) collapsed in 2005 at age 36 after 3 straight seasons with 86 or more appearances and 4 straight with 80.

Rauch, a righty reliever, is tied for 15th after appearing in 85 games in 2006 and 88 games in 2007 for the Nationals. He has averaged 69 appearances in the three seasons since, so he's been fine. His ERA after the two straight seasons with 85+ games has actually been better than before (3.66 to 3.74).

Capps, a righty closer, has been great despite making 85 appearances in his first full season at age 22 in '06, posting a 9-1 record with a 3.76 ERA. He has averaged 64 appearances, so he's certainly durable. He did post a 5.80 ERA in '09, but he's posted ERA's under 3.03 in the other three seasons including a 2.47 ERA and 42 saves in '10.

Bell, another righty closer, appeared in 81 games in '07, posting a 2.02 ERA. But, he posted a 2.77 ERA and an average of 70 appearances per season since. He certainly has been fine.

Gonzalez, a lefty closer, posted a great 2.42 ERA in 80 games in '09 for the Braves, but showed effects of the overuse in '10, suffering a shoulder injury in April, and posted a 4.01 ERA in 29 games overall.

Proctor, you may remember, was a righty middle reliever who the Yankees who was used by Joe Torre for 83 games in '06, posting a 3.52 ERA. Proctor was traded to the Dodgers at the trade deadline in '07 and posted a 3.65 ERA in 83 games overall. He has never been effective since, posting 6.09 ERA in 47 appearances since '07, including a 6.35 ERA in 6 relief appearances for Atlanta in '10. He missed all of '09 after undergoing Tommy John Surgery.

Now, let's talk about Feliciano. He has gone where few others have, pitching in 86 or more games the last 3 seasons, including 92 appearances in '10. Feliciano certainly a different case than Marshall, Tekulve, Torres, and the other pitchers listed above. Feliciano pitches for one or two batters at a a time while Marshall and Tekulve pitched often pitched 3 innings in an appearance, and Torres and the other active or recently active players pitched an inning per appearance. It has been about 50-50 whether these relievers have been able to be durable and effective after pitching in so many games. The data apparently shows that signing pitchers who have appeared in 80 more games in a season can be a risky proposition. But, I'll say it again, Feliciano is a different case. He may appear in a lot of games, but there's not nearly the stress on his arm that there was on the arms of Marshall, Tekulve, Torres, and all the other guys because he pitches so many less innings. Feliciano is not a risk to sign as a free agent. He does not have so much stress on his arm that he will fall apart any season now. He'll collapse eventually of course (not too many pitchers didn't collapse at the end of their careers), but when he's old (pending a serious injury). The Yankees and several other teams have been talking to Feliciano. If any of the teams can agree to terms with him, then they should by all means sign him.