There’s no getting around it - Felix Hernandez had about as nightmarish a 2017 as he could have had. Plagued by injuries throughout the year, Felix posted his fewest innings pitched since his half-season rookie year in 2005. Even when he did pitch, he was bad. He posted an fWAR of 0.4, which barely would have touched the still-not-good 1.0 figure he posted in 2016 had he had a full gauntlet of innings.
Barring another major pitching acquisition (and there are still some good starters on the free agent market), the Mariners seem content to go into 2017 with Felix Hernandez as their third starter, behind James Paxton and Mike Leake (yeah). Is there hope for Felix to return to anything resembling the flamethrower of yesteryear? Because if there isn’t, then the Mariners might be screwed.
Let’s start by looking at why Felix was so bad in 2017. It wasn’t really the strikeouts or the walks. His BB% was 7.0%—about par for the course for his career. His K% was 21.2%, which was actually much better than his number for his 2016 campaign.
No, the reason Felix was so bad last year was the dingers. He posted a HR/9 of 1.77. For comparison, his next highest total from his career was 1.12, in 2016. These dingers weren’t really the product of the fly ball revolution. Felix didn’t give up many more flyballs than in years past. So why so many dingers? Look no further than the weird little outlier below.
Felix Hernandez’s Splits
Had Felix pitched the entire season, that 22.4% figure would have qualified as the highest in all of baseball. Not only that—that 22.4% would have been the highest HR/FB by any qualified starting pitcher in all of baseball history. Ever. And it wasn’t just that dudes were shelling him - his Hard Contact % was 30.7%, which was above his career mark of 27.0%, but by no means a career-worst. No, on the surface, it looks like Felix just got super unlucky.
Fangraphs Projections seem to agree. Steamer has Felix pegged for a much-better-but-still-not-good 1.19 HR/9, and about 2.0 expected fWAR. And that’s with the strikeouts going down and the walks going up. It all seems quite reasonable, and the historic HR/FB mark is certainly unsustainable.
Does that mean Felix will get better? Even if “better” just means hitting 2.0 fWAR? 2.0 fWAR seems feasible. The glory days, not so much.
First of all, there’s the well-documented and consistent drop in Felix’s velocity.
You don’t need to be a data scientist to know that that is bad. It’s gotta level off somewhere... right? Right? Then, there’s the almost-definitely-related problem of his induced whiff rate.
One good way to keep people from hitting dingers: don’t let them hit the ball. Felix has clearly gotten worse at not letting them hit the ball, which is bad.
How do you get better at making people swing and miss? One way is to increase velocity. Let’s assume for a moment that increasing velocity is off the table. Yes, Felix made a jump in velocity in 2014. That might happen again, but it doesn’t seem super likely considering that it just fell right back down again. One might think that the other way would be to increase spin rate.
The correlation isn’t fantastic, but it’s there. If that correlation is unconvincing, here’s a fantastic article that outlines the relationship between whiff rate and spin rate. There are some outliers on either end of the spectrum - Carter Capps is an example of a pitcher with an extremely low spin rate but a very high whiff rate. As you would expect, with increasing spin rate also comes a lower xWOBA (expected Weighted On Base Average, which takes Statcast stats like launch angle and exit velocity into account).
Yes, the correlation is there, but it’s weak at best. Still, in general, it appears from this that higher spin rates do result in worse outcomes for the hitter.
Could Felix do that?
It looks like he already has. And it hasn’t been working.
Statcast data for spin rate is only available going back the last three years. Still, the data is pretty clear for Felix. His spin rate has increased across the board, regardless of pitch selection.
So even though higher spin rates seem to generally correlate with more successful pitchers, the opposite has proven true for Felix. Why could that be? Driveline Baseball suggests that it’s not just high spin rates that are effective, it’s spin rates that are far from the league average. How do Felix’s pitches over the last few years compare to the league average?
Spin rate generally increases with velocity, so for these averages I took spin rates of pitches within 1 MPH of Felix’s pitches. The clear trend is evident in each type of pitch: Felix has always been a low spin-rate guy (with the exception of his curveball), and is getting closer to the league average, limiting his effectiveness.
There are other ways to get better besides making dudes whiff—Felix could learn to induce weaker contact, or more ground balls, or just hope that Dee Gordon is really good at Center Field. Still, Felix has always been a pitcher who gets strikeouts, and if he can’t do that, his ceiling is unquestionably lower.
Aside from that awful HR/FB, the one thing that seems somewhat likely to improve is Felix’s health. Until last year, Felix had a longstanding clean bill of health. Last year’s injury woes may seem like a sign of things to come, but I remain optimistic. I see evidence that Dr. Lorena Martin, the Mariners’ new Director of High Performance, may be influencing Felix’s offseason routine for the better.
But Zach... Felix has thrown soooo many innings! This has got to be the beginning of the end, right?
Well, maybe. But it turns out sheer innings thrown isn’t actually predictive of injury likelihood. If you can bear to get through it, I recommend giving this research paper a read. If you can’t, here’s a highlight.
One prior study has evaluated cumulative work as a predictor for injury in MLB pitchers. While there were several issues with the study methodology, the authors found no correlation between a MLB pitcher’s cumulative work and risk for injury.
So if cumulative work isn’t necessarily a predictor, what could it be? The paper goes into specific types of injuries. Felix’s main issue last year was with his shoulder, so what predicts shoulder injuries?
While total rotation deficit, GIRD, and flexion deficit had no relation to shoulder injury or surgery, pitchers with <5° greater external rotation in the throwing shoulder compared to the non-throwing shoulder were more than 2 times more likely to be placed on the DL for a shoulder injury (P = .014) and were 4 times more likely to require shoulder surgery (P = .009).
In layman’s terms, pitchers with less flexible throwing shoulders are way more likely to suffer a shoulder injury, and it isn’t even close. So what’s Felix focusing on during this offseason, coming off of an injury-plagued 2017? Flexibility!
If Felix actually can get more flexible, it seems quite likely that he’ll be able to stay healthy. An increase in velocity seems unlikely, and an increase in spin rate would seem ineffective and a decrease unlikely. The Steamer projection does seem to be the most likely outcome for Felix—a full season that’s also better than 2016 and 2017, but nowhere near the glory days.