Gleyber Torres Should Be Hitting for More Power Than This
After hitting 62 home runs over his first two seasons in the big leagues — his age-21 and age-22 seasons, no less — Gleyber Torres was hailed as baseball’s next superstar. Even in lineups featuring Aaron Judge, Giancarlo Stanton, and Gary Sánchez, no Yankees player hit more homers over the 2018 and ’19 seasons than Torres. And while subpar defense limited his overall ceiling, his 5.6 WAR over that stretch still represented an impressive output for a player that young.
In 2020, however, the power that defined Torres’ offensive production all but disappeared. Torres, who posted a .235 ISO in 1,088 plate appearances over his first two seasons, slashed just .243/.356/.368 last season, good for a .125 ISO. He homered just three times in 160 plate appearances, but improved plate discipline — his walk rate jumped 5.5 points compared to his career averages, while his strikeout rate fell 5.6 — drove what remained above-average offensive output. Torres still posted a 106 wRC+, but he couldn’t outhit his defense like he had the two years prior and posted just 0.2 WAR.
| Year | PA | HR | AVG | OBP | SLG | wOBA | wRC+ | WAR |
|---|---|---|---|---|---|---|---|---|
| 2018-19 | 1088 | 62 | .275 | .338 | .511 | .354 | 123 | 5.6 |
| 2020 | 160 | 3 | .243 | .356 | .368 | .326 | 106 | 0.2 |
It was fairly easy to dismiss Torres’ 2020 performance as a shortened-season outlier, and while we could go back and raise eyebrows at his markedly worse batted ball quality, it’s important to remember that expected stats are meant to be descriptive rather than predictive. It was entirely possible that the COVID season was the reason for Torres’ dip in both batted ball quality and top-line production. The pre-season projection systems would mostly agree with that sentiment. Both ZiPS and Steamer projected a career-high in wRC+ and a 30-homer season, likely due to the fact that Torres is a young player with an excellent track record, one who should only be improving with more time in the big leagues.
Unfortunately, this season has brought more of the 2020 same from the infielder, who has struggled mightily with the power, producing even worse results than he did during his mini-power outage last year. Through games on Saturday, he’s slashing .242/.325/.317. His .075 ISO is the seventh-lowest in baseball. He’s homered just three times in 272 plate appearances. It’s hard to believe that this is the same player who led all Yankees in homers from 2018-19, a time when the Yankees hit more home runs than any other team in baseball.
What’s even more concerning is the fact that we now have over 400 plate appearances of low-power Torres, a sample that represents nearly 40% of his pre-2020 data. Here’s the same chart as above, but with last season and this season combined:
| Year | PA | HR | AVG | OBP | SLG | wOBA | wRC+ | WAR |
|---|---|---|---|---|---|---|---|---|
| 2018-19 | 1088 | 62 | .275 | .338 | .511 | .354 | 123 | 5.6 |
| 2020-21 | 432 | 6 | .242 | .336 | .335 | .304 | 92 | 1.0 |
That isn’t ideal, and it makes one wonder how exactly Torres’ power has fallen off of a cliff and whether or not it can get back to where it once was. To be sure, Torres benefitted greatly from playing in Yankee Stadium. Even when he hit close to 40 regular season round-trippers in 2019, he was one of the players who most outperformed their “expected home run” total, which is tracked by Statcast and probabilistically considers how many batted balls should have left the yard based on their launch angles and exit velocities. Including the postseason, Torres homered 41 times in 2019, but by expected home run, he should have only hit 32.4, a difference of 8.6:
| Player | Team | Actual HR | xHR | HR-xHR |
|---|---|---|---|---|
| Alex Bregman | Astros | 45 | 31.7 | 13.3 |
| Eugenio Suárez | Reds | 49 | 38.1 | 10.9 |
| Kyle Seager | Mariners | 23 | 13.0 | 10.0 |
| DJ LeMahieu | Yankees | 29 | 19.9 | 9.1 |
| Brett Gardner | Yankees | 29 | 20.4 | 8.6 |
| Gleyber Torres | Yankees | 41 | 32.4 | 8.6 |
| Jesse Winker | Reds | 16 | 7.6 | 8.4 |
| Jason Heyward | Cubs | 21 | 13.1 | 7.9 |
| Yuli Gurriel | Astros | 33 | 25.8 | 7.2 |
As you’ll notice, there are three Yankees, two Reds, and two Astros in the top-10 — a lot of that is due to ballpark effects. Yankee Stadium is home run friendly, and though it does seem as if Torres benefitted greatly from his home park, he’d still be quite the power threat with 32 homers rather than 41. Plus, it’s not as if Torres has changed home ballparks since 2019; he still plays for the Yankees. And it’s hard to even consider the changing baseball as the reason why he’s not the same player. This problem started in 2020, when the ball was still very much juiced. Things seem to have changed for Torres himself.
One very interesting point made by Derek Albin at Views from 314 Ft. and then furthered by Al Melchior at The Athletic was that Torres has just been far too late over the last two seasons and not able to catch up to the fastball. Torres is pulling the ball significantly less than he did his first two seasons, particularly on fly balls, and that is likely a main contributor to his bleak power numbers. After pulling the ball roughly 44% of the time from 2018 to ’20, he’s done so just 39% of the time since. The drop-off is more significant among fly balls and line drives: he went pull-side 32% of the time from 2018 to ’19 and just 24% of the time from ’20 since. This has all happened amidst a notable increase in his groundball rate, which ultimately gives us a full set of his troublesome changes:
| Year | GB% | LD% | FB% | Pull% | Pull%, FB/LDs |
|---|---|---|---|---|---|
| 2018-19 | 35.3% | 22.4% | 42.3% | 43.5% | 31.9% |
| 2020-21 | 43.4% | 20.3% | 36.3% | 38.5% | 24.0% |
Why does pulling your line drives and fly balls matter? Melchoir touched on this with a couple of case studies, but I wanted to look at something a bit broader. Using our splits leaderboards, I pulled every player-season from 2017-19 with at least 90 fly balls and line drives. (That was just the auto playing time suggested for the output.) I then compared each of these figures to the player’s isolated power that season. Pulling the ball on line drives and fly balls does have a statistically significant (p-value of 0 for the coefficient) relationship with a player’s power production. Here’s what the association looks like visually:
There is definitely something there, which shouldn’t be surprising: The pull side is where hitters typically generate their power. That was certainly true for Torres, who, from 2018-19, posted a 91.2 mph average exit velocity when pulling the ball, good for 67th percentile pull pop. When he went the opposite way, however, he posted just a 85.6 mph average exit velocity, putting him in the 52nd percentile. The 5.6-mph difference didn’t rank near the top of the league, but it was the 65th-largest of the 204 hitters with at least 500 batted balls from 2018-19.
Torres has completely flipped the script, and not in a good way. He’s pulling less, and even when he does, he isn’t making good contact. His exit velocity on opposite field batted balls has actually increased since 2020, up to 88.0 mph, but his exit velocity on pulled balls has dropped all the way down to 86.5 mph. Torres used to have 67th percentile pull-side pop. Now? He’s in the ninth percentile. Among the 145 players with at least 500 batted balls from 2018-19 and at least 200 from 2020-21, Torres’ drop in pull-side exit velocity is the worst in baseball, and it’s especially puzzling given that he is still just 24 years old.
| Name | Pull EV, 2018-19 | Pull EV, 2020-21 | Difference |
|---|---|---|---|
| Gleyber Torres | 91.2 | 86.5 | -4.7 |
| Andrelton Simmons | 89.1 | 84.5 | -4.6 |
| Manuel Margot | 90.3 | 86.3 | -4.0 |
| Corey Dickerson | 89.9 | 86.3 | -3.6 |
| Trevor Story | 93.2 | 89.7 | -3.5 |
| Trey Mancini | 90.4 | 87.0 | -3.4 |
| Jonathan Villar | 89.2 | 86.0 | -3.2 |
| Eugenio Suárez | 92.6 | 89.4 | -3.2 |
| Starling Marte | 89.7 | 86.6 | -3.1 |
| Yoán Moncada | 94.0 | 91.0 | -3.0 |
As others have noted, as a result of this huge drop in power, pitchers are willing to challenge Torres more. Though his discipline has only improved since entering the big leagues, he is not as scary as he once was. He’s getting more fastballs, and he’s not doing any level of damage. He’s hitting just .142 with a .315 slugging percentage on 85 plate appearances ending with a four-seamer, a far cry from his .303 average and .703 (!) slug against the pitch in 2019. Perhaps those who surmise that Torres is just late on everything are right. It would explain why he’s pulling the ball significantly less than he used to, and why he has posted such a low exit velocity on the batted balls he does pull.
Torres is still just 24. He has plenty of time to figure things out, and has a track record of being a productive major leaguer. The Yankees would be wise to continue to give him a long leash to allow him to figure things out. Perhaps it’s mechanical, or perhaps it’s just timing. But it’s obvious that Torres isn’t the power hitter he used to be, and while time is definitely on his side, it seems as if tweaks must be made in order for him to return to being the power threat he once was.
Devan Fink is a Contributor at FanGraphs. You can follow him on Twitter @DevanFink.
Wait when did he hit 41 HR?
2019. 38 in the regular season and another 3 in the postseason.
Like a lot of home run hitters in the modern era, he has an open-palm swing with choppy action that produces a lot of underspin (hence his overperforming his xHR numbers). When you swing like that, if you’re a fraction of a second late, you don’t pop it up (as you would if you had a traditional HR hitter uppercut), you hit the ball straight into the ground, often up the middle or oppo. And so you try and speed up, but that just produces more roll overs and off balance can-of-corn fly balls. He’s clearly lost his timing for whatever reason. The weird thing is this is usually a sign of aging, of very slightly declining reaction time. That’s clearly not the case with him, and is more likely a mental block, of thinking too much about being quick (which slows you down). I suspect he will find his timing very suddenly one day, and not look back.
On the YES broadcast, O’Neill and Cone agree that too much movement has seeped into his mechanics. I don’t have anything to back that up; take the opinion of two former players for what it’s worth.
Thanks Devan, great analysis here!
I had a feeling that Torres’s drop-off was a combination of a combination of luck and juiced ball in his first 2 seasons, followed by a combo of bad luck and “deader” ball the last two years. But you underscored what would be the most worrisome item for Torres going forward – if MLB pitchers have found an exploitable weakness and he’s not been able to fully counteract how he’s now being pitched. (This makes me appreciate guys like Trout even more, for how quickly they can regain the upper hand.)
Maybe as simple as reducing bat weight, lenghtening the arc
Not to oversimplify things because Devan laid it all out very nicely, however could the chief culprit be that he’s just not being aggressive enough at the plate? He’s been swinging roughly 5% less from ’20 onward and when he does swing, he’s more frequently late on pitches. Perhaps he’s just trying to take that extra split second to decide to swing or not, which has been good for his plate discipline numbers (K’s are down, walks are up), but is causing him to be late on the ball which is hurting his quality of contact.
I agree there might be something there. His plate discipline seems to be almost elite now, but just from watching him, it looks like he’s not taking many “athletic” or “explosive” swings. It looks like he’s trying to not strike out and therefore taking less aggressive cuts to improve his contact. This is extremely non-fangraphs type analysis but it’s the main thing I can figure from putting together his raw numbers with the eyeball looks
What’s bizarre is that Gleyber’s 2020 regular season was bad, but he was a terror in the ’20 post-season: .435/.567/.696 in 30 plate appearances. Small-sample size, sure, but worth pointing out.
Subjectively, watching his at-bats this year, I’ll see him put a good swing on a great hitter’s pitch and then the ball just dies at about 300′. He’s gone through “hot” periods where he racks up the hits, but they’re all soft line-drives hit the opposite way.
Some online scouts think he’s become too contact-focused, and needs to be willing to strikeout some more in exchange for more power. But over the past few weeks, he’s certainly striking out more, but that’s all he’s done. I hope he figures it out because the game is better when we have more great hitters playing.
He just looks constantly off balance when he swings. He appears to be trying to uppercut and hit home runs but he rarely if ever barrels up balls lately. Gotta think its somewhat mental with him. Not sure what is going on in the Bronx but certainly a notable lack of confidence for pretty much that whole lineup since 2019.
I have always felt like there was a caveat to Gleyber Torres performance in 2019. He hit 13 home runs vs the Orioles which is in all-time record vs one opponent. If you take away the Orioles (you can’t, it still counts obviously) he hit .261/326/420. The average for non pitchers in 2019 was .257/328/444, so, without the Orioles his triple slash was slightly worse than the average non-pitcher. Not saying that those home runs don’t count..they certainly do, but to me anytime there is an outlier such as this it is always good to at least examine it. Now clearly he shouldn’t be *this* bad, but it wouldn’t be as surprising if we thought he was around a league average bat, rather than the perception of him prior.
Sure, it’s an outlier but why take away all 13 home runs? Why isn’t it reasonable that half of them should still count? How many of those specific13 were gone in any park? And even if you took away all of them he was still expected to have finished with 32 and we’d still be left wondering the same question. What happened to the power?
I literally said not to discount it, just look at it. Who knows. Aaron Judge for instance hit even better than Gleyber that year vs the Orioles (not hrs) maybe they had a read on pitchers, etc.
And what would be the point of just looking at it?
Every hitter in baseball – or at a minimum the AL East – got to face the same Orioles’ pitching in 2019, so should we question those other hitters’ numbers, too?
I still think it started intentionally. The plate discipline is my reason why, and the eye test as my anecdote. I watch the Yankees more than most teams and see Torres take 90 mph fastballs down the middle he used to crush on 2-1 counts.
It’s also worth noting that his defense has actually been pretty good by any metric you pick this season while only playing SS, and it’s his first time ever being a positive in base running value, which really does lead me to wonder – did he change what kind of player he wanted to be? Did the Yankees? And if it was intentional, is there time to change back?
Are base running metrics changed by your team? For example if you are the on the worst base running team in mlb history (2021 Yankees) are you somehow a positive by just not being as bad?
Interesting question, but I don’t think so. It’s part of WAR grading.
I was hoping the Yankees would trade him after last year. He owes most of his stellar 2019 to the Orioles (13 HR) and he’s not at all good at shortstop…. or second base. And like virtually the whole rest of the team, he is not a smart fundamentally sound ballplayer (and I lay much of the blame for that on Aaron Boone, who needs to go).
There is a lot of talk about the Yankees going after one of the big ticket shortstops this off-season. If they do sign one then Torres will be in a different uniform next season. Voit, LeMahieu and Urshela aren’t going anywhere and unless he regains some pop his numbers don’t fit in a major league outfield.
I feel like a lot of teams would be willing to buy low on him. Usually guys who were decent pre arb then sucked have suitors.
Probably a dumb question but do we know if he’s had his eyes checked?
In summary, eye glasses and a lighter bat. Get Marcus Thames on the horn. Done.
I think Thames is the Yankees’ hitting coach right now.
It seems like him and DJ LaMahieu have both lost their ability to hit for power. I think that teams are more hesitant to throw them away with the short porch in RF at Yankee stadium. Torres hit over .400 on fastballs on the outer half in 2019, and DJ was nearly as good. I think teams have responded to that and are throwing less fastballs away to Yankees right handed hitters.