Gary Sanchez Keeps Getting Worse
I just released my two simple projections and Gary Sánchez could not be any more polarizing. Using just last season’s stats, he projects to post a replacement-level .656 OPS. And when I use multiple years of StatCast data, he projects for a .889 OPS, the 4th best overall value. The difference comes down to BABIP regression.
To say Sánchez had a disappointing 2020 is a huge understatement. For the hitters with 170 or more plate appearances, his .618 OPS ranked 13th worse. His .147 AVG was the worst. It was fueled by a combination .159 BABIP (2nd worst only to Edwin Encarnacion) and 36% K% (6th worst). The single-season projection’s regression raises his BABIP up to .223 and his AVG to .193. They are better but still far from rosterable.
Transitioning to the optimistic StatCast data, I’ll keep it simple. Here are his 2020 league ranks on several StatCast metrics
The values are almost all at the extreme ends of the spectrum. My StatCast projections like that he hits the ball hard and in the air (Barrell%). After rooting around through the projections’ code, the disconnect between the two projections is with his BABIP.
First, here are historical StatCast BABIP projections:
| Season | BABIP |
| 2017 | .313 |
| 2018 | .293 |
| 2019 | .283 |
| 2020 | .280 |
| 2021 | .275 |
While his projected BABIP is on the decline, it’s still reasonable compared to his .159 BABIP this last season. Compared to other hitters, his 17.4% Barrell% should lead to more hits. Here are the six hitters closest to his Barrell% and their BABIP.
| Name | Barrell% | BABIP |
|---|---|---|
| Matt Chapman | 18% | .291 |
| Teoscar Hernández | 18% | .348 |
| Brandon Lowe | 18% | .309 |
| Bryce Harper | 17% | .279 |
| Brandon Belt | 17% | .356 |
| Eloy Jiménez | 17% | .340 |
| Average | .321 | |
| Gary Sanchez | 17% | .159 |
The other six average a .321 BABIP, almost twice what Sánchez posted. Digging even a bit deeper, Sanchez is getting eaten up by the shift.
While teams are reluctant to shift pull-happy right-handed hitters, they shift him. Since Sánchez is one of the league’s slowest players, he’s almost a guaranteed out when he puts the ball on the ground. Here is a short history of how the league has adjusted to a slowing player who pulled his batted balls at a career-high rate last year.
| Season | GB% BABIP | Shift% | Pull% | Sprint Speed | League Rank |
|---|---|---|---|---|---|
| 2016 | .250 | 37% | 54% | 26.2 | 38% |
| 2017 | .240 | 38% | 52% | 26.1 | 33% |
| 2018 | .091 | 53% | 51% | 25.9 | 28% |
| 2019 | .149 | 74% | 50% | 25.5 | 23% |
| 2020 | .086 | 89% | 58% | 24.9 | 11% |
The shift is just devouring all his groundballs as seen by the .086 BABIP on them. With so few hits, his career-high 36% strikeout rate destroyed any remaining hope for a decent batting average. I don’t see the strikeouts improving much since his strikeout rate has increased for three straight seasons (23% to 25% to 28% to 36%).
Overall, Gary Sánchez needs one or more of the following factors to improve while none of the others get any worse.
- Quit pulling the ball (possible)
- Make more contact (maybe)
- Run faster (yea right)
I’m not betting on any of them. I think the best scenario is that he doesn’t get any worse. I think he’s degraded into an average catcher who has a .220 AVG and hits 15 to 25 HR. He needs to halt years of degrading talent just to stay even, let alone improve.
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
Sanchez just seems to have never got his timing down this year. At all. It may just be me, but it sure looks like his bat is more waggly to start his swing and he can’t hit most hard stuff well. He almost looks like late career Pujols where he is either having a hard time with the fastball or completely guessing wrong on breaking stuff. Might make a swing change/simplification to get him back on track.
Great stuff here, love the progression of the storytelling. I think the league BABIP decreases and shift increases specific to Sanchez are exactly the culprit and the reason we can’t blindly take the statcast projos. Could you do this for other low BABIP players that saw increases in shifts? I think there may be something larger here since shifting has become so extreme. BABIPs may not rebound for all players.
A lot of 2020 seems like small sample size. Although his K% did go up the rest of his peripherals such as swinging strike rate, fb%, and overall swing percent stayed pretty consistent. His babip is almost comically low even for a person w/his batted ball profile so a regression to the mean should be expected. Overall I would expect a return to 2019 with a .232 average and 30+ HRs. That’s a somewhat valuable player at the catcher position but it all depends on cost to roster him in your league
I’m not sure he gets back to 30+ homers without a juiced baseball. I see gleyber torres in a similar boat as sanchez. A happy funball and weak competition made him look like an early round pick, and he had a below average 2020 (SSS of course)
I think you’re right about his average rebounding, but the rising strikeouts are concerning. Thats not SSS noise, that’s a trend over a good chunk of PAs , pitchers have exposed a weakness
Gary Sanchez’s power is most defiantly NOT the product of the happy fun ball. He has consistently ranked in top 10% of EV, barrels, and hard hit percentage for many years and as Jeff pointed out he did so again this year. He also hits a decent amount of fly balls too. As Jeff stated earlier in the article, the issue with Sanchez is his K% and the fact that he gets killed by the shift. He also plays half his games in Yankee stadium which should also help his hrs. I agree that his average will never be respectable until he figures out how to go the opposite way against the shift and strikeout less but the power is absolutely legit. There are reasons he might not get to 30hr next year but they are playing time based. His defense is terrible and if he’s just going to provide and empty 30hrs w/out getting on base then I don’t know how long the Yankees will continue to play him. He also consistently misses time with injuries and being a catcher this is not likely to change next year
How is his defense terrible? Has he ever posted a negative defense number in his career?
He’s led the league in passed balls three out of the last four seasons.
Are passed balls not included in catcher defense?
2020 season only, 100+ innings:
Tied with Sandy Leon and Yadier Molina for most errors by a catcher (5);
Behind only Luis Torrens (6) on passed balls (5);
Most Wild Pitches Allowed by 2 (19 over Tony Wolters 17);
DRS score of -4 (55th of 64)
Framing score of -0.8 (46th of 64)
So, yeah actually, LOTS of negative defensive numbers.
Interesting that the statcast profile is pretty similar to Luis Robert’s besides an extreme difference in foot speed. Robert doesn’t have nearly as shiftable a profile, though.
I’d pay good money to watch Gary try to drop a bunt against the shift and still get beat to the bag.
Well, righties bunting against the shift doesn’t make sense in the first place. They’d be bunting the ball towards first base instead of away from it, and the first baseman always has to play close enough to the bag to get there on a normal ground ball, anyway, so he could easily field it and make a short throw to the pitcher covering the bag. Unless the defense is playing in, they’d actually be better off trying to bunt for a hit by bunting it into the shift.
Really seems like one of those guys who goes to a different team, and then hits 30 HR.
I have to think the yankees are on a very short lease with at him this point, especially considering their preference for Higgy in the playoffs. Really can’t see him getting an extension at this point and wonder if he gets traded this offseason. JT Realmuto would be a nice fit for NYY.
Similarly, gary not in YS3/that lineup could get even worse
A shining example of why Statcast’s x-stats need both a speed component and a pull/oppo component (as well as a ballpark adjustment). I genuinely don’t see him being able to post a BA over .200 on a regular basis. He’s a dead pull hitter with no speed. That’s the Pujols formula for posting horrible BABIPs year after year after year. And unlike Pujols, Sanchez strikes out 25-30% of the time
(Side note, but I’ve wondered many times over the last few years just how different Pujols’ career would have been if he had come along 10 years later when the shifting revolution started. He wasn’t quite as dead pull as he is now, but he was still pretty strongly a pull hitter early in his career and he’s always been a balanced hitter, hitting just a shade more GB than FB while never really being a huge LD guy either. I have zero doubt he wouldn’t be as great as he was, but how much worse would he have been?)
Also, it’s barrel, not barrell. Was willing to let it go the 1st time as a typo but it’s misspelled 3 times.
Their xStats do have a speed component: https://technology.mlblogs.com/augmenting-statcast-expected-batting-average-with-sprint-speed-6be7f60770d2?gi=d7dc736afb8f
I do wish they had pull/oppo adjusted xStats. Andrew Perpetua had a site with xStats that incorporated spray angle and it was interesting to see who were the biggest differences between Savant’s xStats and his. Unfortunately he was hired by a team and is no longer able to maintain the site.
That adjustment is only on weakly hit or topped balls, the theory being that faster players are more likely to beat out those types of hits which is absolutely true. However, y contention is that it matters on hard hit balls as well. The defense gets to play Pujols 20 feet behind the infield dirt (and more than that to LH hitters) where they have to play Billy Hamilton on the infield dirt, That means less reaction time for the defender which means more ground balls get through.
That speed adjustment doesn’t go far enough
Where is the shift data coming from? Unless I’m reading it wrong, his Statcast says he only saw the shift 55% of the time.
https://baseballsavant.mlb.com/visuals/batter-positioning?playerId=596142&teamId=&opponent=&firstBase=0&shift=1&season=2020&attempts=10&batSide=R&gb=1&fb=0