For someone who considers Felix Hernandez his favorite all-time Mariner, I’ve been pretty hard on him this year. I was cautiously optimistic, but not sold by his very good Cleveland start. San Francisco was a disaster. His outing against Kansas City was fairly poor when considering the quality of competition and the environment. But today, in a 2-1 loss against a fairly good hitting Oakland Athletics team, Felix gave me reasons to be optimistic that he might be able to be something more than a back of the rotation pitcher.

Obviously, it’s a small sample size. I’m not willing to say he’s fixed. We have over two years of data now that suggest that Felix just isn’t very good anymore. This start could be the aberration, and he could go back to being well below average like he has been for a long time. Or, it could be the start of a new approach to pitching that lends itself to a strong second act in a Hall of Fame career.

Using data from Brooks Baseball, I took a look at the King’s pitch selection in 2018. I grouped four seam, sinkers, and cutters together—all variations of fastballs—because Felix generally tries to throw those to get ahead in counts early, or induce contact. Entering today’s game, Felix had thrown some variation of a fastball roughly 51% of the time, compared to 20% for his changeup 20%, 22% curves, and 6% sliders. This pattern is somewhat more fastball-heavy than Felix’s usage patterns from 2015-2017, when he threw a fastball around 45% of the time. It seems clear that Felix entered 2018 believing that establishing the fastball was a key to success.

Today, Felix threw that playbook out the window—a very high window, so hopefully it didn’t hit an unsuspecting pedestrian underneath the window. The King threw 41 fastballs out of 101 pitches (40.5%) while throwing his changeup 22.7% of the time, his slider 15.8% of the time, and his curve 20.7% of the time. He recorded five swinging strikes on the changeup, two on the curve, one on the slider, one on the sinker, and none on the standard four-seam fastball. He didn’t throw any cutters today (and honestly, it’s debatable if he does at all—it’s more likely that Brooks just misidentified a few four seamers).

This new pitch mixing added up to 6.1 innings of five-hit, two-run ball with seven strikeouts and no walks. And against a pretty good offense, too! Small sample size applies this early in the season, but Oakland’s 116 wRC+ as a team is tied with the Mariners for fifth best in baseball. This isn’t like giving up three runs to the hopeless Royals. This is getting burned by one mistake—a two-run homer to Jed Lowrie in the first inning—and pretty much dominating the rest of the way.

Staying away from the fastball is probably a good decision. Felix had thrown 134 fastball-type pitches in 2018 going into the game, and batters had feasted on them. In plate appearances ending in a fastball, Felix’s opponents were 9-for-27 with a double, two homers, and six walks against just four strikeouts. That’s a .333/.455/.593 slash line. Not exactly the type of pitch you want to try to establish. On the other hand, all of his softer stuff—the change, curve, and slider—have been far more effective, holding opposing batters to just 5-for-28 with a double, two homers, and four strikeouts against two walks, or a .179/.233/.429 slash line. So the softer stuff can still be punished when a hitter squares it up as evidenced by the .250 ISO against, but it’s clearly much rarer for batters to square up said soft stuff.

Felix just needs to accept that he’s a junkballer now. And that’s fine! Batters are probably still going to hit his fastball hard, but when they’re less certain it’s coming, he should be able to sneak one by from time-to-time…something he can’t do when he’s throwing around, or more than 50% fastballs. By pitching backwards today, he was able to keep Oakland hitters off balance, and post a vintage Felix type of box score while utilizing his repertoire in a completely different way. Time will tell if this is a new strategy for Felix, or if it was just a one-day gameplan against Oakland. I believe if he’s going to have any chance of being an above-average pitcher again, he’s going to have to pitch more like he did today than he has in his other starts this season. Let’s hope this is the start of a trend rather than a one-outing outlier.

Unfortunately for the Mariners, it was a vintage Felix outing in another way as well—the offense couldn’t get a damn thing going against pretty-good-but-not-great lefty Sean Manaea, who looked very much great today! As much as I’ve given Ryon Healy a hard time, it would be nice to have him—and a fully healthy Mike Zunino—in the lineup against tough lefties. Sure, Taylor Motter went deep for the Mariners’ only run today as their first baseman, but in general, you’re going to get better results from Healy. Nelson Cruz also doesn’t quite look like himself right now, so games against lefties may be a bit of a struggle until the roster is back at full strength.

Unfortunately for the M’s, they face another tough lefty tomorrow night when the Houston Astros come to town and throw Dallas Keuchel against James Paxton tomorrow night at 7:10. While Keuchel hasn’t been outstanding this year, he’s always given the Mariners fits, and they’ll hope to reverse their fortune tomorrow night to open the four-game series.