Is Jeremy Jeffress Starter Material?

When the Blue Jays traded some treasured cash considerations to pry Jeremy Jeffress from the Kansas City Royals last November it didn't make too many waves. If Blue Jays fans thought anything about the transaction at all they probably thought something to the effect of, "here's a 25-year-old pitcher with a good arm who is out of options, I wonder where he'll fit?" Now that a year has passed fans have changed their tune. Now they are thinking, "here's a 26-year-old pitcher with a good arm who is out of options, I wonder where he'll fit?" Blue Jays fans are some of the most observant and numerate fans in the league, when someone's age changes by a year they almost never miss it. The point is that one year later it is still somewhat unclear where Jeffress fits with the Blue Jays despite his undeniable talents. Last week the picture became simultaneously clearer and foggier as rumors started to emerge that the Jays were considering using Jeffress as a starter in 2014.

My initial reaction to this rumor was to think it was ludicrous. Jeffress hasn't started since 2011, and has never done it at the major league level. He relies heavily on his unbelievable fastball (96.7 mph on average according to Fangraphs) which gives him more of a relief pitcher profile. He struggles with control and has a spotty history as a starter above A ball. All in all he doesn't seem like an ideal candidate. However, the Jeffress experiment makes more and more sense when you consider the construction of this roster and changes in Jeffress's repertoire in 2013.

First thing's first. Jeremy Jeffress is out of options. If he gets demoted to AAA he will have to go through waivers. He would almost certainly be claimed on waivers. The reason for that is twofold. One is his electric arm. Guys who can touch 100 mph don't grow on trees. If they become available on waivers they are pretty appealing. Jeffress is the type of player that a team that was rebuilding could easily stash in their bullpen and try to nurture into a top tier reliever.The second reason why he wouldn't get through waivers is because last year he took steps toward being said top tier reliever. Jeffress put up a 1.65 ERA and 2.58 FIP in Buffalo and a 0.87 ERA and 3.43 FIP in Toronto in limited work. Those numbers are rooted in small sample sizes but there is no doubt that they jump off the page. If Toronto cuts Jeffress loose, someone will take him.

Based on the premise that losing Jeffress would not be preferable and that he has to be put on the 25 man roster to be kept, the question is where to put him. There are seven slots in the bullpen (I belligerently refuse to believe that this team will open the season with an eight-man bullpen). The following slots are absolutely, positively taken: Brett Cecil, Aaron Loup, Sergio Santos, Steve Delabar, Casey Janssen and Dustin McGowan (if healthy). That leaves one spot, possibly two. There are three remaining relievers on this team that are out of options: Jeffress, Esmil Rogers and Luis Perez. Rogers showed his utility as a swingman last season and Perez is a lefty so it's hard to see Jeffress beating them out.

Where there might be a hole is the rotation. R.A. Dickey and Mark Buehrle are the only sure things in the group, although Brandon Morrow will be given every opportunity to hold down the third spot. That leaves two spots and even if one assumes there will be a free agent acquisition there is still one hole remaining. That hole could be filled by J.A. Happ, but there is enough uncertainty there that it is worth having as many challengers as possible.

It may seem odd to try and throw Jeffress in the rotation based on there not being enough room in the bullpen as one would think things would be the other way around. It is far more common for rotation rejects to head to the bullpen. However, it seems incredibly unlikely that there is an open position in the bullpen so there doesn't seem to be any harm in tossing another hat into the expansive and confusing ring that is the battle for spots in the Blue Jays rotation. There isn't a lot to lose here. If Jeffress doesn't win a spot, frankly the most likely scenario, then you can trade him for some magic beans just like you would have if he didn't make the bullpen. Other members of the bullpen that could conceivably be tried in the rotation (Cecil, McGowan, Rogers and Perez) have all had chances before and there isn't a compelling reason for giving any of them a shot. Putting McGowan in the running would be fun and interesting but it probably wouldn't be smart. It's worth trying something new on the off-chance you catch lightning in a bottle.

Interestingly, there might be some reason for optimism in terms of catching that lightning with Jeffress. The first thing he has in his favor is his velocity. When a pitcher transitions from the bullpen to the rotation normally they lose a couple of ticks. Jeffress has ticks to spare. Even if he lost 3 mph off his fastball (an aggressive estimate) he would have a fastball tied with Shelby Miller for ninth fastest among MLB starters last year. That kind of velocity can get you out of trouble. However, Jeffress has always had velocity and he has struggled as a starter in the minor leagues so the real question is "what has changed recently?" Unless there is something different about Jeffress, there is no logical reason to expect different results. In 2013 something did change. Jeremy Jeffress switched from a four-seam fastball to a two-seam fastball. The following Brooks Baseball chart shows his pitch usage over the past two years:

When Jeffress came back to the Blue Jays in September he came back throwing primary sinkers.This is fairly surprising as he had never thrown more than 13.6% sinkers in any of his previous MLB stints. Although PitchFX information from Buffalo is not publicly available, at least as far as I know, it seems he started leaning on it down there. Buffalo's pitching coach, Bob Stanley, was a sinker specialist in his playing days and while that link is far from a smoking gun there is also a fair amount of some circumstantial evidence. According to Brooks Baseball Jeffress did not throw a sinker in spring training or his first appearance with Toronto in April so the origin of this switch probably resides in Buffalo. The switch to using primarily a two-seam fastball could be a significant one, especially for a pitcher who would be pitching in a park where fly balls are so dangerous. The sample sizes are too small to say anything even close to definitive, but here is a comparison of the effectiveness of the two-seam and four-seam fastballs from Jeffress in his brief MLB career:

The difference between these two pitches is clearly exaggerated because the sample of pitches here is in the hundreds only. However, it might be fair to say that Jeffress seems more likely to get groundballs and throw strikes with this new pitch and those are two crucially important things. I'd probably go so far as to say that his two-seamer is a more effective pitch. He's tried something new and it's working. It may be adjusted to in short order, but for now it's working.

There are many things that are standing in the way of Jeremy Jeffress being an effective starter. His command is problematic to say the least. His changeup is a work in progress. In fact, Fangraphs doesn't even believe he has one. Brooks Baseball is a little more forgiving saying that he threw 12 of them in 2013. Either way it's likely that pitch will need to be added/refined (depending on who you believe) before he is ready to start. There is also the question of his stamina and durability as he hasn't started more than 15 games in a season since 2007. Despite all of this, Jeremy Jeffress is a pitcher with a golden arm and a developing sinker and for that reason alone he's got to be worth a shot. If he can't hack it all in spring training you ship him out of town before the season. If he breaks camp in the rotation and gets blown up you trade him, or maybe even sneak him through waivers if he looks bad enough. If he succeeds you have a starter making the league minimum. The field challenging for spots in this rotation is full of question marks. Why not add another one to mix?