Steven Wright as a Joe Kelly Alternative

Today, Joe Kelly took the mound in Ft. Myers, and it didn’t go well; he gave up seven hits in 2 2/3 innings before leaving with an injury currently being called biceps tightness

Kelly said he had biceps tightness that limited him to fastballs; 'not concerned.' Says he's had it before. 'Usually doesn't last too long'

— Alex Speier (@alexspeier) March 16, 2015

Today’s performance continued a trend from his two previous spring training starts, as he’s now allowed 17 hits in 7 2/3 innings this spring; his spring training ERA currently stands at 11.05. On their own, the results aren’t a big deal, but when you combine it with a potential injury, it might be fair to say that Kelly’s spot could be up for grabs at this point.

Of course, any time the Red Sox rotation is mentioned, people will invariably bring up Cole Hamels, but Boston continues to seem uninterested in meeting the Phillies price, and they do have some interesting alternatives in-house. While youngsters Henry Owens and Eduardo Rodriguez are the two best long-term prospects and might be ready for an audition, John Farrell has made sure the beat writers keep Steven Wright’s name in the picture as well. Per Alex Speier, from this morning:

The Sox would not be opposed to adding a veteran starter to their Triple A rotation. But Farrell again expressed confidence in Steven Wright as a depth starter. “He gives us a lot of comfort. As that knuckleball has come along, he’s throwing a lot more strikes,” Farrell said. “He’s got the ability to give a contrast of style.”

Projecting the performance of knuckleballers is not quite impossible, but is something kind of close to it. Wright could be very good — as he was in a September trial last year — or he could be terrible, with pretty much any result in between seemingly equally likely. Knuckleballers are lottery tickets, especially ones with spotty track records and iffy command. Dickey’s success came after he cut his walk rate in half, which Wright appears to have done in Triple-A last year, but we essentially have half a minor league season and a month of big league action where Wright showed a consistent ability to throw strikes. If he’s not throwing strikes, he’s not any good, and it’s not a big surprise that the Red Sox wouldn’t want to count on his 2014 command improvement carrying over when trying to win in 2015.

But there are a couple of reasons why I think it might make sense for the Red Sox to give Wright a shot, especially if Kelly needs replacing on the Opening Day roster: his Z-Contact% and the knuckleball carryover effect.

We’ll start with former. While Wright’s big league career data includes just 590 pitches to 145 batters and all the results need to be taken with large grains of salt, Z-Contact% is one of the metrics that tends to stabilize fairly quickly , and can serve as a pretty decent proxy for quality of stuff. If you’re throwing pitches in the zone that guys still can’t hit, you’re doing at least one thing right.

Over the last three years, the starting pitcher with the lowest Z-Contact%: Max Scherzer, at 80.5%. Right behind him, at 80.6%? R.A. Dickey. No other qualified starting pitcher is under 83.5%, so these two are the extreme tail for in-zone contact allowed by a guy who has taken the hill every five days of late.

Steven Wright — again, in 34 career innings, and most of those coming out of the bullpen, so caveat emptor — has allowed an 80.0% Z-Contact% in the big leagues. If you change the filters to include all pitchers who have thrown 30+ innings over the last two seasons so that Wright can appear on the leaderboards, he has the 25th lowest Z-Contact% of any pitcher in the majors over the last two years; 25th out 562 pitchers. The guys ahead of him are almost all flame-throwing relievers who just blow away opponents with 100 mph gas. Of course, there are some mediocre names on the list as well, as the presence of Michael Kohn, Burch Smith, and Josh Outman should be a reminder that a low Z-Contact% doesn’t automatically correlate to big league dominance, but guys who can post Z-Contact rates this low are generally pretty good.

So, that’s one very small sample data point in Wright’s favor. It’s something to at least make him a little more interesting than just some random knuckleballer. But even without that, random knuckleballers maybe deserve a little more of a shot than they often get, because there is some evidence that guys who give that kind of “contrast in style” actually do elevate the performance of their teammates.

A little over a year ago, Chris Carruthers published a Community Blog post here looking at the results of pitchers who pitched in relief of R.A. Dickey, and the starters who followed him in the rotation and pitched the next day. The results were pretty stunning, especially when it came to the same-game relievers, as they experienced huge improvements in their performances when relieving Dickey as opposed to relieving other pitchers.

He also applied this same analysis to pitchers following Tim Wakefield, and again found a strong uptick in performance when pitching behind a knuckleballer. There appears to be some real evidence that adjusting to these contrasting styles can have a negative impact on the opposing hitters, and if you follow a knuckleball starter with some flame-throwing relievers, the combination might perform better than their individual projections would suggest.

So, even if Wright himself isn’t great, it’s possible that a lingering effect of his style of pitching could lead to reasonable run prevention outings from all pitchers on days he starts. He can’t be horrible, of course, but a mediocre Wright plus some bump to his relievers could potentially be as effective as a decent fifth starter with normal reliever support. And given that Wright has at least one indicator suggesting that he might be a reasonable-ish big league pitcher, giving him a shot seems like a potentially reasonable option for the Red Sox.

Or, if they wanted to get particularly outside of the box, perhaps the Red Sox could use a tandem combination of Wright and either Owens or Rodriguez, both of whom are high velocity left-handers. By using both for three to four innings at a time, the team could limit their exposure to big league hitters early in their careers and manage the workloads of the young arms while still allowing them to gain big league experience. There’s a downside to using two roster spots for one starting job, but as a temporary solution while the team tries to evaluate what they have in Wright, it might not be a crazy experiment.