Tuesday, 18 June 2013

Esmil Rogers: Hidden Treasure or Fool's Gold?

I don't know what your feelings were when, back in November, the Blue Jays traded utility infielders Mike Aviles and Yan Gomes to the Cleveland Indians for right handed relief pitcher Esmil Rogers, but mine could probably best be described as optimistically indifferent.  I found it pretty hard to get too worked up about the trade one way or another.  At the time, it fit in with an apparent organizational focus on acquiring big armed relievers, and was notable only because it fit that trend.  The players sent to Cleveland were really just bit parts.  Yan Gomes had been a bit of fun in 2012, but was pretty terrible overall.  Mike Aviles was, infamously, the piece that came back from Boston in the John Farrell trade and had been tentatively penciled in at second base by fans since at the time there weren't really any other options.  To some it seemed curious that the Jays would trade their only second base option (Izturis hadn't been signed yet, but would be about a week later) for yet another hard throwing reliever.  While it was a little odd, we were still talking about utility infielders and low leverage relievers so there were only so many cares I could be bothered to give.

Now, I keep referring to Rogers as a relief pitcher, but that wasn't always the case.  The Rockies had used him as a starter in 2010 and 2011, giving him a combined 21 starts over that time, but the results were, well... pretty terrible.  He made 13 of those starts in 2011, posting an ugly 6.66 ERA, although playing in Colorado was likely something of a factor as his xFIP was a better but still not great 4.91.  His biggest problem was his control.  That year as a starter he walked 11% of the batters he faced, or 4.77 per nine innings.  Home runs were also a problem (but when aren't they in Colorado?) as he was giving them up at a rate of 1.63 per nine innings.

In 2012, however, he pitched exclusively as a reliever and had significantly better results.  After a  mediocre start to the season with Colorado, he was traded to Cleveland in June and became a whole new guy.  Both his walk and home run rates got under control, and he pitched to a second half ERA of 3.40 and FIP of 2.84.  Given his poor results as a starter, and his relative success in relief, nobody would have been surprised if the book on Rogers' starting career had been closed.  The Blue Jays certainly seemed determined to keep him as a reliever, opting to give starts to a revolving door of nobodies before Rogers got his shot May 29th against Atlanta, and did well enough to earn himself two more starts since.  What should we expect from him going forward?

He had an excellent start Thursday against Texas in which he went 7 innings, allowing 1 run on 5 hits and a walk, with the lone run being allowed on a solo home run to Nelson Cruz in the second inning.  So far as a starter in 2013, he has pitched to a 1.26 ERA, having allowed just 2 earned runs over his three starts.  So is this just a run of good luck for Rogers, or his he doing something differently than he has in the past that might give us hope that his success is something more than a mirage?  For starters, we can all agree that Rogers' ERA is due to regress.  His FIP as a starter is 3.48, so that's already an indicator, despite the tiny sample size, that he's been having some good fortune so far.  However, even if his ERA should regress to match his FIP, it would still be a significant improvement on his previous turns as a starter.  There are some signs that he's pitching differently than he was in Colorado, and better results following a change in approach are generally more encouraging than better results that just seem to come out of nowhere.  He hasn't been using his repertoire in the same way we've seen in the past, and so far it's been working for him.

The first change is the difference in his fastball usage.  According to Brooks Baseball, in 2011 he threw his four seam fastball 56% of the time to both right handed and left handed hitters.  His sinker, or two seam fastball, was only an occasional offering with Rogers throwing it just 4% of the time.  In 2013, he has been throwing that sinker far more.  He is still throwing the same number of fastballs, about 59% in 2013, but now a lot more of them are of the two seam variety.  Of his total pitches thrown in 2013, only about 45% of them have been four seam fastballs, and 14% have been two seamers.  Rogers' sinker gets more swings, whiffs, and ground balls than his four seamer does, so it's possible that mixing his sinker in more has allowed him to continue throwing mostly fastballs without just relying on blowing everyone away with his impressive velocity.

The increase in his sinker usage so far is most notable when he is ahead against left handed hitters, where he throws his two seamer 29% of the time, second to his four seamer at 38%.  In 2011, he still threw 39% four seamers in these situations, but following in frequency were sliders and curve balls at 26% and 20% respectively.  The sinker was almost non-existent. 

Rogers hasn't just been mixing up the types of fastballs he's been throwing.  He's also been using his repertoire of breaking and off speed pitches differently than we saw in 2011.  Most notably he is throwing a higher percentage of sliders than we've seen before.  The increase is against both handed batters but, as expected, more pronounced against right handed batters.  The additional slider usage seems to have come at the expense of his curveball and change up.  In 2013 he is throwing his slider 30.5% of the time, his curve ball 8% of the time, and just 1.6% of the pitches he's thrown have been change ups.  Contrast that with 2011 when those numbers were 20% sliders, 11% curve balls and 9% change ups.  Clearly it is the change up that has mostly fallen by the wayside.  He's only thrown a few of them to left handed batters this year, and none at all to righties.

So I think it's pretty safe to say that he's pitching differently in 2013 than he did in 2011, the last year in which he made a start, and also the year in which he made the most appearances as a starter.  It appears as though the elimination of his change up started in 2012, as he hardly threw it at all in relief that year.  However, he also didn't throw his sinker very much.  Instead we saw the appearance of a cutter that he threw about 8% of the time.  So clearly Rogers has been tinkering with his repertoire these last two seasons, but the most dramatic changes definitely took effect in 2013.  So should the results of those changes give us hope for the future or is he just riding a run of good luck since he stepped into the rotation?

Well that's where it gets a bit trickier.  Of the numbers that are most influenced by luck and randomness, BABIP, strand rate (LOB%), and HR/FB rate we see warning signs.  His BABIP is lower than league average at .256, but his strand rate is astronomically high at 96% (70-72% league average) and his HR/FB rate is a bit higher than average.  He's been striking out more batters than he had been in relief, and as a starter in 2013 both his K% and BB% are sitting a bit above average, even after striking out just one batter in the last of his three starts.  As a starter in 2013 those numbers are at 19.6% and 7.1% respectively, compared to a pretty ugly looking 15.5% and 11.1% in 2011.

So I don't want to be the guy that rains on Esmil Rogers' parade as there are certainly some encouraging signs.  He's made some changes to his repertoire and performed very well so far as a starter in 2013 and, as mentioned, better results that follow a change in approach are more encouraging than ones that just seem to come from nowhere.  He's also become a bit of a fan favourite by helping his own cause with the bat in Atlanta in his first surprising start.  That's all very well and good, but it certainly looks like his ERA is better than a deeper look suggests that it should be, although I can't imagine anyone expects it to remain at 1.26.   For one thing, his sample size as a starter this year is still so small that one bad outing could blow all his numbers up real good.  Hell, he starts tonight against Colorado, so maybe by the time you're reading this it will all have come crashing down and this entire exercise will seem like it was a total waste of your time and mine.  I sure hope not!  Or maybe he'll have another great start and we can all start arguing about what happens when Morrow and Happ come back because, with apologies to Gregg Zaun and his latest idiotic idea, it won't be Morrow heading to the pen.

Most troubling for Rogers and expectations for his future performance are his lower than average BABIP and his way too high to be sustainable strand rate.  These numbers suggest that he's been getting a bit lucky with balls in play not going for hits, but has been extremely lucky with the timing of those hits.  Men are getting on, but they aren't scoring, so Rogers has been scattering his hits well and not allowing the opposition to string them together and push runs across.  Unfortunately this tends to be more luck based than a skill he can control, although pitchers who can strike guys out are sometimes better able to sustain an above average strand rate than guys who depend on defense for their outs.  Either way, a LOB% of 96% has nowhere to go but down, so Rogers will, with absolute certainty, start allowing more runs than he has so far.  However, even if his ERA were to get more in line with his FIP of 3.48 it would be a huge improvement over his past results as a starter, not to mention better than Dickey (4.90 ERA, 4.88 FIP), Morrow (5.63 ERA, 5.43 FIP), Buehrle (4.66 ERA, 4.48 FIP) and Johnson (4.38 ERA, 3.96 FIP).

Again, all standard small sample size disclaimers apply, but Rogers has always had a live arm and if he's figured out how best to use his repertoire of good stuff without walking the ball yard, he could be a valuable piece for this team.  If he hasn't stolen himself a rotation spot, at the very least he might have leap frogged a few names on the depth chart to become the team's first choice for long man/6th starter, perhaps filling the void that had been so ably filled these past few years by Carlos Villanueva who departed for the Cubs as a free agent this off season amidst much clamouring for the Blue Jays to resign him.  Then again, maybe he'll just turn right back into a pumpkin and spend the rest of his days toiling as a low leverage reliever.  It's certainly too early to say, but there are signs that Rogers is doing things differently than he was in his previous turns as a starter and I, for one, am going to keep hoping the results continue to be better, even if not as good as they have been, until he shows otherwise.

I just hope it's not tonight against Colorado.  That would be embarrassing.

No comments:

Post a Comment