Gerrit Cole’s Wild Card Dud Was the End of a Longer Slide

When the Yankees signed Gerrit Cole to a nine-year, $324 million deal in December 2019, they envisioned him contending for Cy Young awards and pitching do-or-die games in October. They likely didn’t imagine those would be Wild Card games, however, nor did they foresee the ace right-hander taking an early exit before things really got out of hand, but that’s just what happened on Tuesday night in Fenway Park. On the heels of a strong but uneven season that may yet garner him a Cy Young award, Cole fizzled, surrendering a pair of homers and retiring just six of the 12 Red Sox he faced before departing with a 3-0 deficit. The Yankees’ offense was kept at bay by opposite number Nathan Eovaldi and the four relievers that followed, and the Red Sox advanced with a 6-2 victory.

Cole pitched about as well as any American League starter this year, posting the highest strikeout rate (33.5%) and strikeout-to-walk differential (27.8%) among ERA qualifiers, and finishing second in both flavors of WAR, trailing only Eovaldi in FanGraphs’ version, 5.6 to 5.3, and the Blue Jays’ Robby Ray in Baseball Reference’s version, 6.7 to 5.6. His 2.92 FIP ranked second behind Eovaldi’s 2.79, while his 3.23 ERA placed third behind Ray’s 2.84 and Lance McCullers‘ 3.16. Cole also had the lowest xERA of any AL qualifier (3.15).

Even so, the 31-year-old righty entered Tuesday night with at least some cause for concern. He left his September 7 start after 3.2 innings due to tightness in his left hamstring, and while he was solid enough in his return a week later — five innings, seven strikeouts, and one run on 108 pitches against the Orioles — he was cuffed for five homers and 15 runs in 17.2 innings over his final three starts. One of those was actually solid enough; he threw five shutout innings against the Red Sox in Fenway Park on September 24 as the Yankees built a 7-0 lead, then allowed a three-run homer to Rafael Devers in his final inning of work.

The late-season funk was deep enough to generate questions about which version of Cole would show up on Tuesday night, but unfortunately for the Yankees, the question was settled early. Kyle Schwarber sent a 103-mph rocket to center field on Cole’s fifth pitch, a 98.2-mph four-seam fastball that was more or less in the middle of the plate. Brett Gardner caught it for an out, but the missed location and the quality of contact didn’t bode well for the pitcher. After getting Enrique Hernández to pop out, Cole avoided the strike zone altogether against Devers, and walked him on six pitches. He then fell behind Xander Bogaerts, 2-1, before leaving a changeup in the middle of the zone; the slugging shortstop hammered the ball 427 feet to dead center for a two-run homer:

Per Baseball Savant, it was just the third time in his career that Cole served up a home run to a right-handed hitter on a changeup, and the first time in over four years. His first two came while pitching for the Pirates, first on May 5, 2014 against the Nationals’ Ian Desmond, and then on September 17, 2017 against the Reds’ Eugenio Suárez.

Pitch choice and location aside, it was an all-too-familiar spot for Cole to wind up in. During the first inning of his starts this year, batters hit .265/.317/.521 for a .354 wOBA against him, accounting for seven of the 24 homers he served up. His ERA in the first inning was 4.80, compared to 2.91 in all other frames. Tuesday was a bad time to hold form in that regard.

In the second inning, Cole had to work around a one-out double by Kevin Plawecki, who hit a 105-mph rocket off the center field wall on a 98.5 mph fastball that again caught too much of the zone. Cole recovered to strike out both Bobby Dalbec and Christian Arroyo, the former looking at a 3-2 slider that was actually off the outside corner, and the latter swinging at high cheese. To start the third, he got ahead of Schwarber 0-2, but after missing way outside with a changeup, he came back with a 97.4 mph four-seamer above the zone. The Boston slugger went and got the cheese, schwarbing it to right field at a 110.3-mph clip for a solo homer.

After a soft infield hit to the third base side by Hernández, and then a six-pitch walk to Devers, Cole’s night was done; manager Aaron Boone pulled the plug before the Yankees’ inconsistent offense, which managed just six runs during the team’s final three games of the regular season as it squandered home-field advantage for this game, had to dig out of a bigger hole. Clay Holmes extricated the Yankees from the jam with a strikeout of Bogaerts and a double play ball off the bat of Alex Verdugo. The Yankees made a game of it, trimming the lead to 3-1 in the sixth, but a bad send by third base coach Phil Nevin and a great throw by Bogaerts left Aaron Judge hung out to dry at home plate on the second of Giancarlo Stanton’s two long singles off the Green Monster. The Red Sox never let them get any closer.

It wasn’t that Cole lacked his typical velocity; his 97.8 mph average four-seamer was a whisker ahead of his season average. His 18% swinging strike rate and 34% CSW rate were both above his season averages as well. Yet his command was lacking; he threw far too many noncompetitive pitches, particularly fastballs:

Worse, when Cole got to two strikes, he couldn’t close the deal:

Ouch. Three of the four batted balls of 100 mph or higher that he allowed came with two strikes.

Afterwards, Cole refused to blame his hamstring or his bout of COVID-19 for his late-season woes, telling reporters, “At the end of the season, we are all going through and wearing whatever we’ve had to overcome to get to this point. You know, the other team is dealing with the same kind of situation.” He noted that it wasn’t so much that his fastball command was unreliable, as he generated three popups with it; that Schwarber had to expand his zone to reach the homer; and that his changeup got hit hard. “When it’s all said and done, there wasn’t one pitch that was good enough because we didn’t get the job done,” he said.

Asked whether he could put his finger on what happened over the last month, Cole didn’t offer a blanket explanation, saying “Evaluate each game individually… It just wasn’t the same answer every time.”

In terms of batters faced, Cole’s start was the sixth-shortest in Wild Card game history:

Shortest Wild Card Game Starts by Batters Faced
Pitcher Tm Opp Year IP H R BB SO HR BF
Liam Hendriks OAK NYY 2018 1.0 1 2 1 1 1 5
Luis Severino NYY MIN 2017 0.1 4 3 1 0 2 6
Sean Manaea OAK TBR 2019 2.0 4 4 0 5 3 10
Jon Gray COL ARI 2017 1.1 7 4 0 2 1 11
Ervin Santana MIN NYY 2017 2.0 3 4 2 0 2 11
Gerrit Cole NYY BOS 2021 2.0 4 3 2 3 2 12
SOURCE: Baseball-Reference

Four of those six starts involved the Yankees, who have made too much a habit of traveling this route into October by playing in four of the last six AL Wild Card games. Two of the above starts came in the same game in 2017; after the Twins chased Severino, the Yankees chased Santana and rallied to win while the bullpen held Minnesota to a single run over 8.2 innings.

That list is no place to be, and a start like Cole’s will be a tough one to live down given that it occurred within one of the game’s most heated rivalries, and against the backdrop of the Yankees’ relative lack of postseason success in recent years. They’ve exited before the ALCS in seven of the past nine seasons, and haven’t won the World Series since 2009, an eternity by the franchise’s standards.

Cole’s start will be cited as evidence of his fall from grace after a strong first two months to 2021, though the reality is a bit more complicated. Undoubtedly, his season was a clear step down from 2019, when he delivered a 2.50 ERA, 2.64 FIP, 39.9% strikeout rate, and 7.5 WAR for an Astros team that came within one win of a championship. But judged by everything except his 2.84 ERA, Cole’s 2021 season was a step up from last year’s shortened campaign; driven by a home run rate that ballooned to 1.73 per nine between seasons of 1.2 per nine on either side, his FIP rose to 3.89 in 2020 while his strikeout rate dipped to 32.6%, still good for third in the AL.

Cole’s 2021 season had two obvious points of inflection that conveniently segment his body of work into thirds, more or less: the crackdown on pitchers’ use of foreign substances, which was first reported on June 3, though enforcement didn’t begin until a few weeks later, and the pitcher’s positive test for COVID-19, a breakthrough infection that was reported on August 3 and that sidelined him for over two weeks.

Here’s how his season looks by the basics:

Gerrit Cole’s 2021 Season in Segments
Date GS IP K% BB% HR/9 BABIP ERA FIP
Thru June 2 11 70.2 36.9% 3.4% 0.64 .298 1.78 1.77
June 3-July 29 10 59.2 31.7% 7.6% 1.81 .281 4.68 4.09
August 16 on 9 51.0 31.3% 6.1% 1.24 .341 3.53 3.15
After June 2 19 110.2 31.5% 6.9% 1.55 .309 4.15 3.31

One could argue that the hamstring issue marked another point of inflection, but I’m not sure how useful breaking the last part into segments of four and five starts — the first of which featured a lights-out 0.73 ERA and 1.02 FIP between his return from illness and his early exit — is in the grand scheme.

All of Cole’s numbers declined after the specter of sticky stuff enforcement reared its head, but as you can see, his ERA was half a run worse than his peripherals suggested, owing something to a lousy defense that ranked 27th in the majors in DRS (-63) and 25th in OAA runs prevented (-13); by the latter, Cole’s -3 runs placed him in just the 16th percentile among all pitchers.

Getting back to the leaderboard comparisons, Cole’s 31.5% strikeout rate and 24.6% strikeout-walk differential form June 3 onward both would have ranked third in the AL, his 3.66 FIP seventh, 16 percent better than league average, his 2.2 WAR tied for eighth. Relative rankings of that order would have registered as something of a disappointment had they been maintained over the course of a season, as the Yankees aren’t paying Cole to be merely a top-10 AL pitcher — and the team would have finished outside the playoff picture.

And now, a closer look at his Statcast numbers:

Gerrit Cole’s 2021 Season in Segments, Statcast Version
Date FFv Spin EV LA Barrel% HardHit% AVG xBA SLG xSLG wOBA xwOBA
Through June 2 97.2 2560 90.3 13.8 7.1% 41.7% .238 .195 .366 .347 .275 .252
June 3-July 29 98.1 2368 88.7 12.7 11.3% 37.7% .194 .239 .366 .425 .267 .316
August 16 on 97.8 2422 86.9 10.7 11.3% 36.1% .242 .229 .463 .472 .308 .313
After June 2 98.0 2393 87.9 11.8 11.3% 37.0% .218 .234 .415 .448 .287 .314
SOURCE: Baseball Savant

There’s no getting around the fact that Cole’s four-seam fastball spin rate dropped conspicuously after news of the crackdown emerged; his spin-to-velocity ratio (SVR) dropped from 26.3 to 24.4 in that same span. Even before the league began enforcing the ban on foreign substances via the umpire checks, he made headlines first for his awkward answer when asked directly whether he used Spider Tack on June 8, and then for his complaints about the difficulty of gripping the ball on June 16. Cole, who serves on the executive subcommittee of the Players Association and has spoken up about various labor matters such as service time manipulation and competitive balance, was hardly alone in discussing grip issues, or in calling for the league to incorporate player input into the new rules; among frontline hurlers, the likes of Tyler Glasnow, Max Scherzer, and (ugh) Trevor Bauer spoke up as well.

Among pitchers with at least 150 four-seam fastballs thrown before June 3 and 300 after that date, Cole had the 16th-largest spin rate drop, putting him in the 89th percentile of that group:

Fastball Velocity and Spin Rate, Before and After Crackdown
Pitcher Spin1 Spin2 Dif Velo1 Velo2 Dif SVR1 SVR2 Dif
Burch Smith 2521 2143 -378 93.2 93.6 0.4 27.0 22.9 -4.1
Madison Bumgarner 2494 2136 -358 91.1 89.7 -1.4 27.4 23.8 -3.6
James Karinchak 2452 2190 -262 95.9 95.9 0.0 25.6 22.8 -2.8
Walker Buehler 2630 2374 -256 95.3 95.4 0.1 27.6 24.9 -2.7
J.P. Feyereisen 2631 2405 -226 93.6 92.8 -0.8 28.1 25.9 -2.2
Richard Rodríguez 2583 2358 -225 92.9 93.2 0.3 27.8 25.3 -2.5
Garrett Richards 2586 2368 -218 94.1 94.4 0.3 27.5 25.1 -2.4
Shohei Ohtani 2345 2147 -198 95.6 95.6 0.0 24.5 22.5 -2.0
Tyler Anderson 2416 2222 -194 90.1 90.9 0.8 26.8 24.4 -2.4
Jake McGee 2261 2070 -191 94.4 95.1 0.7 24.0 21.8 -2.2
Tyler Mahle 2468 2278 -190 94.3 93.9 -0.4 26.2 24.3 -1.9
Spencer Howard 2279 2094 -185 94.7 94 -0.7 24.1 22.3 -1.8
Casey Mize 2257 2083 -174 94.4 93.5 -0.9 23.9 22.3 -1.6
Brad Boxberger 2506 2335 -171 93.7 93.4 -0.3 26.7 25.0 -1.7
Drew Smyly 2180 2012 -168 92.5 91.7 -0.8 23.6 21.9 -1.7
Gerrit Cole 2560 2393 -167 97.2 98 0.8 26.3 24.4 -1.9
James Kaprielian 2166 2005 -161 92.7 93.1 0.4 23.4 21.5 -1.9
Jordan Romano 2503 2346 -157 97 97.8 0.8 25.8 24.0 -1.8
Josh Sborz 2435 2282 -153 96.4 96.9 0.5 25.3 23.6 -1.7
SOURCE: Baseball Savant
Minimum 150 four-seam fastballs thrown before June 3 (set 1), and 300 thrown after that dat (set 2). SVR = spin-to-velocity ratio.

Cole’s drop in four-seam spin rate from June 3 onward was about 2.5 times the major league average of 68 rpm (from 2,319 rpm to 2,251). Similarly, his drop in SVR was about 2.6 times the major league average of 0.7 (from 24.7 to 24.0). Cole isn’t the only awards candidate on that leaderboard, as the names of Buehler and Ohtani stand out. As you can see, the changes affected pitchers with more fastball spin than Cole (who ranks in the 91st percentile in that category) and with much less. In terms of SVR — a category in which Cole placed in the 82nd percentile among qualifiers during the first part of the season — his drop was the 10th largest (93rd percentile).

Results-wise, the numbers in the table prior to that one show that Cole’s velocity actually increased beyond the first leg of his season, and his spin rate rebounded somewhat. Both his average exit velocity and hard-hit rate fell relative to that first leg, though his barrel rate did rise substantially. What’s more, where his actual batting average and slugging percentage were higher than his expected numbers during the first leg, they were lower during the other two legs save for a slightly higher batting average than expected in the third leg. Taken as a whole, his gaps between actual and expected stats were very small and right in line with his career norms, suggesting that at least some of the fluctuations had less to do with any particular changes than to sample size and randomness:

Gerrit Cole Actual vs. Expected Batting
Year AVG xBA dif SLG xSLG dif wOBA xwOBA dif
2015 .239 .239 .000 .336 .367 -.031 .274 .291 -.017
2016 .289 .263 .026 .410 .371 .039 .326 .304 .022
2017 .254 .254 .000 .434 .423 .011 .315 .314 .001
2018 .198 .197 .001 .332 .335 -.003 .265 .270 -.005
2019 .186 .180 .006 .343 .315 .028 .246 .237 .009
2020 .197 .196 .001 .405 .377 .028 .279 .271 .008
2021 .223 .207 .016 .372 .363 .009 .276 .272 .004
Total .226 .220 .006 .371 .363 .008 .281 .280 .001
SOURCE: Baseball Savant

Crackdown-wise, this is not meant to be an indictment of Cole given the number of pitchers who were apparently using foreign substances, or who experienced spin drop-offs from the first third of the season to the remainder, whether with regards to their fastballs or other pitches. He had points in his season following the ban where he pitched brilliantly, and he was hardly facing a cupcake schedule; by Baseball Reference’s RA9Opp calculations, which measure the park-adjusted scoring rates of the teams a pitcher faces in the service of that version of WAR, his opponents’ 4.96 runs per game was the highest among AL ERA qualifiers.

Still the performance numbers are what they are, and taken together with the disappointing end to Cole’s season, we — and he and the Yankees, more to the point — are left with more questions than answers as to what happened, both on Tuesday night and in the bigger picture, and where his performance goes from here. Like Clayton Kershaw, the only pitcher with four losses in potential elimination games (Cole is tied for second with three alongside eight other pitchers including CC Sabathia, Max Scherzer, and Hall of Famers Tom Glavine and Randy Johnson), he may have to endure the weight of the can’t-win-the-big-one tag until he wins one.

In a league where the last five Cy Young winners either didn’t pitch enough innings or were entirely out of the league, and where no one pitcher dominated enough categories to make a clear-cut case, Cole may yet bring home the award that eluded him in 2019, when teammate Justin Verlander narrowly beat him out. Such hardware could be cold comfort in the aftermath of the Yankees’ elimination. He’s hardly alone in terms of blame for the team’s precarious entry to the playoffs or its swift exit, but of all the players in pinstripes, Cole’s upcoming winter might feel the longest.





Brooklyn-based Jay Jaffe is a senior writer for FanGraphs, the author of The Cooperstown Casebook (Thomas Dunne Books, 2017) and the creator of the JAWS (Jaffe WAR Score) metric for Hall of Fame analysis. He founded the Futility Infielder website (2001), was a columnist for Baseball Prospectus (2005-2012) and a contributing writer for Sports Illustrated (2012-2018). He has been a recurring guest on MLB Network and a member of the BBWAA since 2011, and a Hall of Fame voter since 2021. Follow him on Twitter @jay_jaffe... and BlueSky @jayjaffe.bsky.social.

16 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Towel
2 years ago

“Kyle Schwarber sent a 103-mph rocket to center field on Cole’s fifth pitch, a 98.2-mph four-seam fastball that was more or less in the middle of the plate”

Uh… That pitch was several inches above the strike zone.

Okramember
2 years ago
Reply to  Towel

Schwarber hit his above the strike zone homer in the 3rd inning not the 1st.

hamiltonheights
2 years ago
Reply to  Towel

No it wasn’t. You’re thinking of the homer, Jay’s talking about the flyout Schwarber hit to Gardner in the bottom of the first.