In spring training in 2008, I was playing centerfield for the Rays against the Yankees at Legends field, which has terrible visibility in the late afternoon. The hitter, catcher, and umpire, were all in the shade, making them seem more distant than usual. I was trying to see the ball come off the bat against a background of mostly white fans mostly wearing white tee shirts and sunglasses and watches and other shiny things flashing in the sunlight.
A promising rookie named David Price was on the mound, facing Derek Jeter. Though David made Derek look foolish that day, they were only just beginning a fascinating competitive relationship. David was only slated to throw an inning, so he was throwing 100 mph fastballs, half of which I could see from centerfield weren't straight, some moving several inches right to left. My anxiety level in the outfield subsided a bit as it became increasingly clear that inning that the Yankees couldn't touch anything David threw, let alone actually hit anything to us in the outfield with any authority. The opposing Yankee fans gave David a healthy round of applause as we ran off the field.
Unlike regular season bullpen scenarios, bullpen appearances during spring training are typically scheduled. These short scheduled appearances give a pitcher an extraordinary opportunity to showcase electric stuff. David Price is a frontline starter like Clayton Kershaw or Zack Greinke who could probably put up Aroldis Chapman or Wade Davis numbers out of the bullpen. Frontline starters are talented enough to frequently overmatch hitters while pacing themselves to throw 6 to 9 innings, but they are rare. Rotations are rounded out by more hittable pitchers who are prized more for their durability and tact in avoiding big innings and terrible outings.
Watching Price that day made me question the way we think about pitching staffs and how they are constructed. After all, movement notwithstanding, 97 mph is better than 93 mph. And these days, baseball decision makers have begun to ask some of those same questions—hence phenomena like the rise of the super bullpen in Kansas City, a team that just won the World Series despite their starters throwing less innings than any other team's starters in the American League.
Of course the construction of a major league pitching staff begins with a series of minor league decisions: for one, organizations must decide who is going to the bullpen or who is going to be a starter. Sometimes it's a simple choice. Many pitchers don't have the stuff needed to consistently retire good hitters while pacing themselves to throw several innings. In the large array of data we have of pitchers pitching out of the bullpen compared to pitching as starters, we see their stuff declines the longer the appearance. But what if starters had shorter leashes? Could their stuff play up for five innings instead of seven or eight?
The anchor of that Royals bullpen, Wade Davis, is a former starter who was moved to the bullpen and became the most dominant reliever in baseball. For the sake of thoroughness, I asked Wade about the idea of developing a pitching staff where starters were asked to do less, and relievers were asked to do more.
"Last year in our bullpen we basically had 5 or 6 guys who were former starters ready to go one or two innings," he said of the Royals.
But what if instead of being a 1-2 inning reliever, he was a closely monitored starter? As the line between starter and reliever gets blurred, it's worth considering whether a pitcher with the skillset of a Wade Davis—"top of the rotation stuff"—might be able to do more for his team if he were given more innings to pitch. My instinct is that while he may not be a David Price if moved back to the rotation, Wade could definitely be a James Shields.
For what it's worth, Wade told me the difference between his results starting and relieving might have less to do with the amount of innings thrown or the psychology of the role, than the way he prepared for it.
"I don't know if it was a mental thing or a physical thing. As far as what I would be now if I was still a starter—I don't know. Really when I got moved to the bullpen I started taking my life seriously, and I got a trainer, and I started lifting weights, and I just stopped being lazy."
This aligns up with my memory of Davis as a teammate—he wasn't a gym rat, he was just a pitcher. More of a baseball player than an athlete who happens to play baseball.
"I met this trainer in New York right when I got traded to the Royals, when he started training me he started pointing out huge flaws in the way my body functioned and how I moved. We started doing heavy weights, Olympic lifts, crossfit type stuff and within a year I lost body fat, gained weight, and jumped up in velocity and confidence."
The truth is, in many cases, a starter is in the eye of the beholder. There is no scientific formula to determine whether a guy belongs in the bullpen or the rotation.
"Sometimes they pitch their way into the bullpen," says R.J. Harrison, the Tampa Rays' senior advisor for scouting and baseball operations. "They're not able to command the fastball well enough. They're not able to develop that third pitch—they're not able to soften the ball. Sometimes their stuff is better than their ability to pitch. With Jake McGee, we made that decision after Tommy John, which made it easier. Asking him to quote unquote 'pitch' didn't make sense...his value was in utilizing his power in short spurts.
"The radar gun doesn't always tell the story. (McGee's) fastball is probably as good a stand alone fastball in baseball. I mean, this guy throws fastballs by major league hitters. They actually swing and miss at it. I think fastball command is the most important indicator—a guy with a big, lively fastball doesn't have to command it as well in just an inning of work."
When starters with plus fastballs never develop command, they increasingly end up in the bullpen. But at the same time, the most lively fastballs are hard to command. Developing a live fastball may be counterproductive to developing command, but they both have value. As a hitter, something I noticed over time en route to the majors was the increase in fastballs with movement, whether on purpose or not. Straight fastballs became increasingly rare.
Let's take this experiment out to its extreme. If a pitching staff were to consist of thirteen bullpen pitchers, could we really expect them to throw about an inning and a third every other day? That works out to about 112, high-intensity bullpen innings a year, which seems unsustainable for the arm. Most teams rarely let a reliever throw more than 90 innings. But, as Wade points out, a considerable degree of the abuse the reliever's arm takes is due to the sporadic, inconsistent nature of their use. There's no reason not to believe some pitchers couldn't throw 100-120 innings out of the bullpen if they had a much clearer sense of when they're going to pitch, and if they were never asked to throw on back-to-back days. Though 13 closers may not be the most effective way to build a staff, the role of the starting pitcher is another traditional baseball convention that may change. It might be changing already.
The predicament of what to do with a young power arm was a hot topic for a moment as a result of Brian Cashman's handling of Joba Chamberlain and Phil Hughes as rookies. It isn't rare to use rookie starters in the bullpen, but this predicament seemed to linger longer—impetuous fans thought Joba should be able to throw the 96 MPH fastball and the dizzying slider he featured in the bullpen for 9 innings every fifth day as a starter. Most arguments of that time seemed to miss the point that pitchers' stuff depends on how much stuff they are actually tasked to throw.
With Joba's data muffled by his freak injury, Phil Hughes is a better comparison with Wade Davis—both were pitchers whose longtime roles were undetermined as rookies, and whose stuff is comparable, as I can say first hand after having faced both. Though Wade has become the best reliever in baseball, it's still quite unclear whether or not he ended up in the right role. Wade credits his turnaround not to his new role as a reliever, but to his aforementioned new trainer, Niyan Oladipo, and his new attitude, which happened to coincide with his move to the bullpen.
Hughes has become a solid innings-eating starter, and perhaps Davis could have too. In 2013, Davis' last year as a starter, Davis posted a 5.32 ERA in 135.1 IP. That year, Hughes posted a 5.19 ERA in 145.2 IP. Per Fangraphs, Davis' average fastball as a starter was about 92.5 MPH while Hughes' was 92 MPH.
Wade Davis' numbers out of the bullpen are so astronomically good that it's clear he is a more valuable asset than Phil Hughes. But it is important to remember we're talking about guys who were, at one time, very similar pitchers. Davis has far exceeded the Royals' expectations, or even best hopes, for him. Could Phil Hughes have been a dominant closer? Could Wade Davis have become a dominant starter?
It's relatively impossible to forecast a Jake Arrieta-like evolution for Davis—Arrieta's velocity somehow climbed during his evolution. But Arrieta, like Madison Bumgarner early in his career, suffered from being coaxed by player development staff into normalizing his natural mechanical tendency of throwing across his body. There is also the real possibility that Davis may not have improved at all as a starter if he stayed in that role. And now he's so dominant as a reliever that there's no point in going back.
The prevailing theory amongst player development executives is that a front end starter is more valuable than a dominant reliever. When David Price debuted in the 2008 playoffs as a reliever, the Rays didn't consider keeping him in that role. Aroldis Chapman is another enigmatic case study for converted relievers. Arguably the best closer in the game thereafter, the Reds initial decision to make him a reliever was rather whimsically based on the team's immediate needs en route to winning the NL Central in 2010. Just like the Rays handling of David Price, the Reds chose to try to develop Chapman as a starter though. For some reason they were impatient with his very minor struggles as a starter and chose to put him back in the bullpen during their 2012 division title run. Chapman posted a 1.51 era with 38 saves that year and was an All-Star, which essentially sealed their decision. Should it have?
Few relatively clear dividing lines in player development regarding pitchers exist—there are few starting pitchers or relievers that pitch in the major leagues throwing fastballs under 90 mph without having first established themselves in the league with higher velocity. Livan Hernandez, Mike Mussina and Bartolo Colon all used to reach the mid nineties with their fastballs. Often, teams will sign college starters who they believe will only have a chance of helping a major league club as relievers if they add some more velocity. Another clear line is when starters fail to develop more than two Major League quality pitches. Jake McGee was a starter for most of his minor league career but he never developed a third or fourth pitch—he has never started a major league game.
With that in mind, here are two pitchers who make compelling cases for bullpen conversion:
After two years of back-end-of-the-rotation numbers as a rookie and second-year pitcher, Nova showed promise in 2013 when he posted a 3.10 ERA in 139.1 IP. In his last two seasons he has underachieved while battling injuries. Nova's average fastball velocity of around 93 MPH is similar to Wade Davis' velocity as a starter. Nova could see the same rise in velocity as a reliever. The Yankees current need is starting pitching, so they'll likely try to keep Nova in the rotation.
Despite front-of-the-rotation stuff, Edwin Jackson has posted mid-to-back-of-the-rotation numbers throughout his career. In the last few years, he has failed to make the starting rotation, instead dwelling in the dreaded purgatory between relieving and starting. With an average fastball velocity of about 94 MPH, it's not crazy to think Edwin could hit 100 mph as a reliever, which would allow him to use more of the plate and potentially alleviate his command anxieties.
I asked Davis how the bullpen compared to starting in terms of the physical challenge—what might guys like Nova or Jackson expect were they to transition?
"Starting is more of a full-body challenge," he said. "There aren't that many guys who make 30 starts a year in the big leagues. It's just hard to do. In the bullpen your arm is wound tighter, your arm isn't loose all the time, you don't really get loose, you throw as hard as you can to get loose and then you go throw 15 pitches. When I was starting I'd throw 30 or 40 pitches to get loose and you have all day to get ready and stretched out."
Davis didn't think there was actually a major difference between starting and relieving from a mental standpoint—the notion that a personality plays better in the bullpen, he said, is overstated. You just have to learn to get comfortable living on the edge.
"It's definitely more intense in the bullpen," he said. "Not because it's the end of the game but because you aren't tired. You never get a chance to calm down at all. You're only on for a short amount of time so you don't get to work on anything. You're kinda white-knuckling the whole time.."
Did he miss starting?
"No. I think all I care about now, whether it's starting, relieving, playing shortstop, is making sure that I'm in good shape. Making sure I'm doing everything I can to be as productive and healthy as I can. I thought that I wanted to go back after 2014. I knew it wasn't going to be an option no matter how much I asked because the numbers weren't any good as a starter so the Royals would look terrible for trying it again. I kind of figured there was no reason to go back at 30, even now, at 32."
Would he have any particular advice for starters moving to the bullpen? In keeping with the ways of of the pen, he kept it simple.
"Don't give up runs."