A quick note: This article is about relievers and not (entirely) my high school sociology teacher, Mr. Span. I thought you’d want to know that.

 * * *

Back in high school, so the early Pleistocene era, I had a sociology teacher with a few stock phrases he enjoyed repeating. If you were in any of his classes you’d certainly hear him say, “Act like you know” more times than you could count. Although it was said like this, “[mumble mumble mumble] Act like you KNOW [mumble mumble mumble].” What did he mean? It could have been a lesson on the importance of confidence when speaking, but it could just as likely have been his retort to the giant pudding-eating, top-hat-wearing panda he saw over by the blackboard. He was old.

He had a few other gems that I’ve lost in a haze of college parties, baseball statistics, and screaming children, but there is one more that I remember. “Know your role.” Like his other phrases, it was most often recited while pacing the front of the classroom staring at the ground. It was meant as more of a… uhh… umm… well, now that I think about it, I don’t have a damn clue what he meant. But being high school kids we found it hilarious anyway. We’d mimic him in the hallways, pacing back and forth between our locker and the drinking fountain mumbling, staring at the ground, and shaking our head “[mumble mumble mumble] Act like you KNOW…” The other kids would laugh. Then you’d hit them with a quick, “Know your role!” and they’d howl with laughter.

We made fun and I couldn’t tell you a thing about sociology, but that teacher’s odd sayings have stuck with me.* The relevance of his sayings to what he was supposed to be teaching us was never clear, but the sayings themselves did resonate, eventually.

*As did the methods of my college statistics teacher who used to explain statistical theories with blue balls in a bowl. There is no actual problem with this beyond the fact that once you use the words “blue balls” to a classroom of college students you’re wasting your breath to go any further.

“Know your role” is the one I’ve thought most about, especially in relation to baseball.

How important is a player’s role? Is it something players should work within or fight to rise above? These are real questions for both life and baseball.

As baseball has advanced (or, depending on your point of view, moved forward through time) as a game, it has become more specialized. That goes triply for the bullpen. The modern bullpen is one that exists solely in a world of specialization. Every pitcher has an assigned role and when the roles are changed it’s big news, even if the end result is the same group of guys sitting on the same bench waiting for the same phone to ring.

It is against this perceived over-reverence for role, at least one form of which comes in converting relievers to starting pitchers, that modern sabermetrics has often railed. The argument in a nutshell: starters are more valuable because they throw more innings. The more innings thrown by a team’s good pitchers, the fewer innings must come from lesser pitchers.* The counterargument is that because starting is more difficult, in the end a team is sacrificing quality for quantity. At best. At worst, the role is so important that changing it can mess a pitcher up.

*The cost of a good starter on the free agent market compared to a good reliever plays into this as well.

Solving this dilemma is near impossible. There aren’t enough cases to conduct any kind of proper study and even if there were, the individual circumstances surrounding each instance would be enough to render the study worthless.

However, we don’t give up that easily. Just because definitive answers might not be in the offing doesn’t mean we can’t gain some clarity. This season offered the potential of some clarity as three high profile relievers in Daniel Bard of the Red Sox, Chris Sale of the White Sox, and Neftali Feliz of the Rangers all made the switch. This being just over a half-season’s worth of time, the results are incomplete, but as you might expect, the available information is interesting. And all over the board.

Sale was drafted in 2010 and threw 11 games combined in High-A and Triple-A, all of which came as a reliever. Then he was called up to the majors and put in the bullpen. By the end of 2010 he had thrown 23 1/3 innings of 1.93 ERA ball. He followed that up in 2011 by throwing 71 innings of 2.79 ERA ball. Both very good seasons, but neither suggested the kind of dominance he has displayed as a starter this season. Sale’s season (2.37 ERA in 117 innings) has maybe been a bit better than his raw numbers suggest, but, repeatable or not, he’s been one of the better starters in the American League. Given all that, it’s hard not to conclude that the change in role hasn’t affected him negatively. (If it has, imagine how good he’d be if it hadn’t!)

Farther south in Texas, Neftali Feliz made seven starts, compiling an ERA of 3.16 in 42 2/3 innings. All but one of those innings came as a starter. That was in line with conventional wisdom on the topic, as Feliz’s stats as a starter were a step down from those he posted as a reliever. But 41 innings isn’t much to go on. There would be more but then he got hurt. Now he’s rehabbing. Feliz and the Rangers get an incomplete, but this brings up questions. Would Feliz have been hurt if he had remained in the bullpen? There’s no way to know, though the conventional wisdom seems to be that pitchers are prone to injuries regardless of role. Or maybe that’s my own personal wisdom. In any case, it would be hard to say definitively that Feliz was damaged by the role change unless the increased density of innings was responsible somehow for his injury.

The last and perhaps most interesting of the three is Daniel Bard. Bard was the subject of much consternation during the offseason as the Red Sox made their plans to move him to the rotation known. That Bard himself was in favor of the idea combined with the fact that the Red Sox bullpen was full to the breaking point while the rotation was down a few men. The organization took the chance.

But what kind of chance were they really taking? Bard was a good pitcher. Was changing his role really going to ruin that? If he couldn’t hack it as a starter, the organization could move him right back into the bullpen and be no worse off for wear.

Of course it hasn’t worked out. Bard was fine in the rotation, then bad, then a steaming bowl of mess. After a gruesome performance wherein he somehow managed to give up five runs in 1 2/3 innings on only one hit (six walks and two hit-by-pitches helped things along) he was mercifully demoted to Triple-A. You might think that would have solved things. Nope. Even after transferring back to the bullpen, Bard has 15 walks and 16 runs allowed in 18 1/3 innings for Pawtucket. It seems a statement from Bard follows each rough game saying he knows what he did wrong, it’s just a slight mechanical twinge is all, and things will be better the next time out. Sometimes things are better the next time out. If so, they are never better the time after that. Then comes another statement, rinse and repeat. Clearly there are many things wrong with Daniel Bard.

It all makes me wonder about the power of role. For Chris Sale, moving to the rotation wasn’t a problem. Feliz wasn’t having issues with it until he got hurt. C.J. Wilson and Matt Harrison have done fine with it. Further back, both Whitey Ford and Jim Palmer threw out of the bullpen during their rookie campaigns with varying degrees of success. They both went on to Hall of Fame careers. Changing roles didn’t ruin any of them, but then that’s a sample that I’ve selected because changing roles didn’t ruin them. Pretty sure that isn’t what the kids call “good science.” What it does show is that then, as now, changing a pitcher’s role isn’t necessarily a recipe for post-apocalyptical-Kevin-Coster-movie style disaster.

That being said, the potential to ruin a pitcher’s career exists. In theory. Right now Daniel Bard stands as the preeminent cautionary tale to that effect. At 27 years old, Bard’s career is in some sort off chaos. It’s hardly over, though I suppose it could be. It is possible he never rediscovers the mechanics/mojo/black magic that made him one of the premiere set-up men in baseball over the last few seasons. He could also remember that he just needs to turn his shoulders a certain way and then everything will be as it was. It’s hard to say. No, actually, it’s impossible to say.

But maybe the most pertinent question is this: Would Bard have imploded similarly if he had remained a reliever? Impossible to say for sure, but his 2011 might offer a clue.

In 2011 Bard posted a 2.03 ERA in 62 innings with 63 strikeouts to just 15 walks. That was from April through August. Pretty good, right? But, from September on Bard was a disaster. In 11 innings he gave up 13 earned runs (14 runs) while walking nine (to 11 strikeouts). That suggests that Bard’s problems might go further back than we thought. For what it’s worth, Bard posted a 6.57 ERA with 16 walks and 18 strikeouts in 24 2/3 innings in Spring Training.

So does all this tell us about role? Tough to say. Pitchers say it’s important to them. People in baseball run their teams a certain way which emphasizes it. That has to carry some weight. Yet there are numerous examples of pitchers successfully making the change like Chris Sale, just as there exists the cautionary tale of Daniel Bard. Ultimately it depends on the circumstances, the pitcher involved, his psyche, his throwing style, his health, the way the team handles it, and the vagaries of the winds. It’s a lousy answer, but in the end, we’re all just acting like we know anyway.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
No Jeff Samardzija? I thought there were like six relievers that were converted to starters this season (or was one Aroldis Chapman before Dusty left him in the pen). Maybe there are more samples than you think.
Samardzija was too hard to spell.

It seems to me that most of the pitchers chosen make the move successfully, but some smaller percentage don't (Phil Coke). Of course, the pitchers who are chosen are chosen because their teams think they can make the switch successfully.

In any case, Samardzija is a good example, you're right. I probably should have included a paragraph on him. But even if I did, do you think that makes it conclusive?
No, it wouldn't conclusive, but more data are better than less, right?

I'm actually surprised when relievers-to-starters works well; Ryan Dempster was a case where it was completely unexpected that he'd be a useful starting pitcher.

Dempster was a reliever from 2004-07 before being switched back to a starting pitcher. He then improved from a 4ERA, 1.4 WHIP guy into a 3.5 ERA, 1.3 WHIP guy over the past five years. Although most of the guys in your article also don't have much MLB experience, which Dempster did.

Everyone remembers the Ryan Madson on-again/off-again SP/RP switchover. That one was frustrating for fantasy owners.
More data is better. Agreed.

Papelbon was a starter for a while too, and I believe, in the minor leagues as well. There are lots of pitchers who switch back and forth in the minors. In fact, I'm probably not too wrong to say every relief pitcher was a starter at some point, though that's likely similar to the fact that just about every major leaguer was a shortstop at some point.
Come again? Sale's minor-league appearances were all in the bullpen, not as a starter. Both his player card here, his data sheet at baseball-reference, and his bio at concur in this.
This has now been fixed in the article.
It may be big news when a starter converts to a reliever or vice versa. It's small news when a starter pitches an inning of relief on his throw day to spell an overworked bullpen, but I sure would like to see more of that.

While we're at it, can we get some love for the underappreciated swingmen of the world? The '79 Pirates couldn't have made it to the postseason without Grant Jackson in a role that ex-Buc Tom Gorzelanny may get to reprise for the '12 Nationals.
Nolan Ryan also dabbled in the bullpen for the first four years of his career. Only about 60% of his appearances were starts before 1971.
Great point NYYanks.

Sidebar; Ryan never threw 200 innings until he was 25 (284). This fact always makes me laugh, whenever I hear an old cantankerous baseball writer talk about how Ryan's aproach with young pitchers (no pitch counts et al) is so "old School".