Why Roenis Elias can't steal Joe Kelly's rotation spot

Red Sox fans have little reason to be impressed by Joe Kelly. Since his arrival in Boston, he has pitched to a 4.60 ERA, showing serious difficulties in pitching into even the middle of games, dealing with control issues, and generally failing to look like a player who pitched to a 2.69 ERA in 2013 with the Cardinals.

Red Sox fans also haven't seen anything impressive from Roenis Elias in Boston, but that's because he hasn't pitched (outside of his spring training debut) since being traded. And a brief look at his baseball card stats should bring the thought to mind: "isn't this guy just...better?" Since coming to Boston, Kelly has had a higher walk rate, lower strikeout rate, and significantly higher ERA than Elias, who comes in at a 3.97 mark on his short career. If anyone should be the fifth starter, isn't it Elias, not Kelly?

It's an argument I made a few times myself, in fact. And with Elias getting off to a good start this spring with three scoreless innings on Monday, it might seem like the pressure is on for Kelly as he makes his debut against the Pirates. But I've come around. It shouldn't be, and it probably isn't, and here's why...

As easy as it may be for oft-burned Red Sox fans to forget, Joe Kelly does have real value. He isn't due to become a free agent until after 2019, and it's always worth remembering that we live in a world where players like Ian Kennedy get paid. In a big way.

"Well wait," you might say, "Ian Kennedy isn't great, but he's better than Kelly was last year." It does look that way, but it's not actually so clear-cut, and this is where Elias' numbers come in too. Kennedy was pitching in San Diego last year, Elias has spent his career in Seattle, and Kelly was in Boston. One of these things is not like the other. OK, well, they're all different in different ways. But clearly Fenway Park is the only environment of the three which is downright hostile to pitchers.

Just to put some numbers on it, Kelly has pitched to a 91 ERA+ (park/league-adjusted) since joining the Red Sox (Kennedy was 85 last year, for the record). That's not good. It is, by definition, below-average, and the sort of thing the Red Sox would ideally upgrade. But as it happens, Roenis Elias? He hasn't really been much better. His ERA+ for his career--which has been about as long as Kelly's time in Boston--is just a 93. Yes, 93 is higher than 91, and that's good for Elias, but it's not really a significant enough difference to be going on unless we're in a completely even vacuum otherwise.

And we're not! The important context lies in options and, for Kelly, the uncertainty of optional assignment waivers. Both Kelly and Elias have options. Either one could be placed in Triple-A without being designated for assignment. But Kelly has been up for too long now--if he goes down, other teams get a shot at him.

Traditionally, we've heard that there's a "gentleman's agreement" to ignore said waivers among clubs. We even saw it in 2014 when the Sox optioned Stephen Drew to the minors after signing him to get him into game shape after his long layoff. Technically any team could have claimed Drew. They would have effectively been signing him to a one-year, $10 million deal while circumventing the qualifying offer that Boston extended to him after 2015. Or, in the most likely case, a team like the Yankees could've simply put in a claim and then stuck out their tongue as the Red Sox were forced to pull him back, throwing him right onto their 25-man roster without any sort of warmup.

Nobody did, but that situation is a little different. For one, the Red Sox were basically using the waivers to get a player some much-needed spring training after an awkward offseason. They weren't so much trying to cheat roster limitations, but being prevented from doing something by the rules that was not really outside the spirit of said rules. With Kelly? They just have too many bodies, and would be trying to stash away an asset that, by rights, should go to whichever team is willing to give him a spot in the major leagues right away.

Still, teams have looked the other way on such moves before, but Kelly is...really worth picking up. Again, we Red Sox fans haven't necessarily seen his best side, but at the end of the day a sub-average starting pitcher with upside and three cheap(ish) years left is quite the commodity in Major League Baseball. It doesn't hurt that if other teams held it against them in any concrete way, the claiming team could rope the MLBPA into things in a hurry, since "gentleman's agreement" is basically the nicest possible way of saying "collusion."

Could the Red Sox just pull him back? Sure. But then we'd just be left with one too many men on the roster again, and while the Sox can stash either Elias or Kelly away in the bullpen as a perfectly reasonable swingman option, they're actually overfull there as well. Hell, maybe the claiming team would actually be after one of those potential reliever DFAs. On Boston's 40-man roster, Tommy Layne, Edwin Escobar, and Steven Wright are all out of options. Meanwhile, Koji Uehara, Junichi Tazawa, Craig Kimbrel, and Carson Smith are locks for four spots, with Robbie Ross very likely to occupy a fifth. Already it looks like the Red Sox will be losing a pitcher to waivers. Putting Kelly or Elias in the pen would have them down another.

Those don't seem like big names to lose, but they're valuable depth in roles with awfully low reliability and high turnover. Meanwhile, moving Kelly or Elias into the bullpen would put them in a position to make a short spot start or two, but make them less capable of filling a long-term opening in the rotation compared to if they were making regular starts in Pawtucket. Given that either one would likely be seen as one of the first lines of defense otherwise, that's no small consideration. In the end, the Sox would probably just be sending Elias to the minors anyways.

What is a small consideration, though, is the damage either Kelly or Elias can do if the Red Sox let the wrong one pitch for a while to start the year. Or, at least, the difference in expected production is. Sure, if Kelly pitches to a 9.00 ERA where Elias would've been great, that's going to cost the team wins. But the actual disparity we can expect is likely much smaller. Two points of ERA+ just doesn't scream "disastrous mistake" to me. And if Kelly does pitch to a 9.00 ERA, well, he won't last long, and the Sox can quickly move on without sacrificing any depth elsewhere.