On the afternoon of March 12, in the hours after MLB announced that spring training would be suspended because of COVID-19, several teams in ongoing games played out the string in Florida. The last game going was Phillies vs. Rays in Port Charlotte, which ended with a faceoff between Phillies reliever Jonathan Hennigan and Rays pinch hitter Ruben Cardenas. Hennigan got ahead 1-2 and delivered again. “The pitch is taken outside corner, called strike three, and that is your ballgame,” Rays radio broadcaster Neil Solondz said. “And we don’t know how long it’ll be our final ballgame.” That strikeout, which concluded the last competitive major league game for four-plus months (at minimum), raised the combined Cactus and Grapefruit League strikeout rate in 2019 to 24.1 percent, up from 22.3 percent last spring and 23.0 percent last regular season.
If the spring training schedule had been completed, the winnowing of rosters and the demotion of minor leaguers like Hennigan and Cardenas might have suppressed that strikeout rate slightly before Opening Day. But the final figure almost certainly would have ended up higher than the previous season’s spring strikeout rate, auguring yet another increase in the regular-season rate. A trend toward decreased contact is the closest baseball has come to a constant during an era of offensive upheaval: MLB’s regular-season strikeout rate has climbed in every season since 2005. That streak of 14 consecutive increases is the longest in the game’s history, easily exceeding the previous high of 11 straight seasons from 1955 to 1965, which was abetted by a strike-zone expansion in 1963.
In March, all signs still pointed toward the streak extending to a 15th season. But the pandemic-driven adoption of the designated hitter in the National League, along with other rule changes and conditions unique to this small-sample, 60-game season, may conspire to stop the streak. It will probably be close, though, and like everything else about sports (and, well, life) in 2020, the outcome is dependent on a confluence of factors that are difficult to predict.
Several causes have contributed to this prolonged strikeout rise, including technology-driven growth in the size of the strike zone, a focus on catcher framing, rising pitch velocities, shorter outings by starting pitchers and fewer plate appearances against the same pitcher within the same game, an emphasis on spin rates and whiff-happy breaking balls, a sabermetrics-inspired recognition of the value of strikeouts for pitchers and, perhaps, an embrace of power-oriented, uppercut swings in the age of launch-angle measurements, optimized defensive positioning, and a historically lively ball. Most—if not necessarily all—of those forces are still in effect this year, much to the dismay of fans who lament the lack of balls in play and the resulting scarcity of fielding and baserunning action in the modern game. The question, then, is whether the universal DH and any other strikeout suppressors can counteract the K-rate increase that we would expect to see under normal circumstances, at least temporarily arresting the inexorable rise of the most frequent of the three true outcomes.
As pitchers have become increasingly overmatched at the plate, their strikeout rates have skyrocketed even faster than those of non-pitcher hitters. But as starters get yanked earlier in games, they make fewer trips to the plate. Last year, pitchers struck out in 43.5 percent of their plate appearances, but they accounted for only 2.7 percent of all MLB PA, which limited their ability to elevate the league’s overall K rate.
Designated hitters have whiffed more often than the league-average batter in every season since 2011, but they still strike out far less often than pitchers did (24.9 percent of plate appearances last season). Replacing those pitcher plate appearances on NL teams (and on AL teams in interleague road games) with DH plate appearances will bring the K rate down, but by how much?
Because NL teams didn’t plan around the DH when they were assembling their rosters over the winter, their designated hitters are likely to be worse (and more strikeout prone) than those of AL teams. (Since the institution of season-long interleague play after the 2012 season, AL DHs have struck out in 22.4 percent of their plate appearances, compared to 24.7 percent for NL DHs, in far fewer PA.) However, there’s another way in which eradicating pitcher hitting can depress strikeout rates. Many pinch-hit plate appearances come in place of the pitcher, and pinch hitters also tend to strike out often (29.0 percent of PA over the past two seasons), both because they aren’t great to begin with and because they’re suffering from the pinch-hit penalty. With pitchers removed from the starting lineup, many of the plate appearances that have gone to pinch hitters in the past will also be reallocated to more capable bats.
We can get some sense of the magnitude of the potential strikeout savings by comparing NL teams’ performance in road games against other NL teams—with pitcher hitting in effect—to their performance in road games against AL teams, played under DH rules. Over the past two seasons, NL teams have struck out in 23.2 percent of plate appearances in the former situation (or 23.4 percent if we remove strikeout-suppressing Coors Field) and 22.7 percent in the latter situation. That differential hasn’t held true for all previous seasons, but it seems like a reasonable place to start.
For another look at the question, I asked FanGraphs author Dan Szymborski, who operates the ZiPS projection system, to project MLB’s strikeout rate for this season with and without pitcher hitting. ZiPS spit out a strikeout rate 0.5 percentage points lower in the DH scenario, after accounting for the pinch-hitting effect and apportioning playing time to the hitters FanGraphs expects to get time at DH. That’s virtually the same difference we get if we simply substitute last year’s DH strikeout rate for the pitcher-hitter strikeout rate in the 5,098 plate appearances pitchers made last year. For what it’s worth, when the AL added the DH in 1973, the AL strikeout rate decreased by 0.7 percentage points more than the NL strikeout rate, although starters stayed in games much longer then and pitchers weren’t quite as incompetent at the plate.
If adding the NL DH slashes roughly half a percentage point off the MLB K rate, then the fate of the streak could come down to the wire (assuming the pandemic permits the season to come down to the wire). Over the past five seasons, the average annual increase in strikeout rate has been—yes, you guessed it—0.5 percentage points (0.52, to be precise). And the increases in three of the past four seasons have surpassed 0.6 or 0.7 percentage points.
Because the addition of the DH doesn’t definitively settle the streak’s fate, it makes sense to consider other elements of this strange season that could nudge the K rate one way or another. Pitch velocity, a major driver of whiffs, has also ticked up dramatically, not only during the pitch-tracking-technology era but likely long before. However, the surge in speed seems to have plateaued lately, and it’s unclear which way velo will swing this season, given the possibility of quick hooks and relatively ill-prepared pitchers.
“If we have shorter stints and guys are throwing fewer bullets this year, then it stands to reason that they might throw harder,” says Doug Thorburn, a former pitching analyst for Baseball Prospectus and motion analyst for the National Pitching Association. “But competing against that will be the abrupt/unorthodox training regimen leading up to the season, which would very well leave a lot of pitchers falling short of top form.” Thorburn speculates that average velocity will be slightly lower compared to last year, although he expects wide variation from pitcher to pitcher or team to team.
Former major league outfielder Sam Fuld hazards the same guess, for some of the same reasons (and some different ones). “I think more than anything you’ll see slightly lower velocities because of empty stadiums/less adrenaline,” says Fuld, who’s now the director of integrative baseball performance for the Phillies. “I think guys trained pretty well during the break, but the abbreviated Summer Camp might lead to less velo because guys haven’t built up to that higher-intensity workload that they normally do. But we’ll have warmer weather to start, so there will be a lot of confounding variables!”
Those words, “confounding variables,” came up in multiple responses from the public analysts and team personnel I surveyed. As Fuld indicated, the average gametime temperature typically rises throughout the season, as does the average speed of four-seam fastballs.
According to temperature-corrected data from Pitch Info, fastball speed is basically constant once the effect of temperature is removed, but that statistical adjustment doesn’t make those pitches look any slower to hitters in higher-temperature months. This year, though, MLB bypassed the cold-weather months, and although that also may mean skipping the lowest-velo period for pitchers, that could still be a boon to hitters, who’ve historically struck out more often in the first month of the season than they do during most of the summer months.
Research by former Baseball Prospectus author (and current Cleveland Indians analyst) Max Marchi suggests that after accounting for temperature, hitters are actually more productive than pitchers at the start of the season, despite the traditional March/April strikeout spike. It’s possible, then, that hitters could have the upper hand at the start of this season, too—and because the temperatures won’t be working against them, they won’t have to dig themselves out of an early-season strikeout-rate hole. It’s notable, too, that the biggest bump in strikeout rate has historically come in September/October, when roster restrictions have loosened. This year, late-season rosters will be limited to 26 players, which may ameliorate that late-season strikeout-rate rise. Then again, rosters will be bigger than usual throughout the season, starting at 30 players and shrinking after two-week intervals to 28 and then 26.
MLB’s transition from Trackman to Hawk-Eye tracking tech this season may make it tough to tell at first whether any apparent pitch-speed difference is real or simply a product of different standards. The effects of other potential strikeout-rate influencers are even more murky. We know from more Marchi research that pitchers do dial up their velocities in certain counts and game situations and against good hitters, but we don’t have the data to determine whether their pitch speeds would sink in empty ballparks. Fan-free stands could also lead to less home cooking from umpires on pitch calls, a possible component of home-field advantage.
In a short season with reduced travel, fatigue may matter less than it usually does, which might (or might not) lead to better swing decisions. The three-batter-minimum rule might trim a few strikeouts by depriving pitchers of the platoon advantage, but ultra-short outings were too rare already for there to be a big effect. The new automatic-runner rule will cut down on extra innings, when strikeouts tend to be a little less common (by almost a percentage point over the past five seasons). MLB could follow through on its threats to crack down on foreign substances, which would hamper pitch movement and whiffs. And then there’s the specter of COVID-19, which could sideline some players for sizable percentages of the season and has already convinced other players (and umpires) to opt out. It’s impossible to predict stats with any certainty when the makeup of team rosters, the quality of umpiring, and the viability of the season itself are so subject to change.
The strikeout-rate mystery is a manifestation of how hard it will be to assess anything this season. Considering all the uncertainties, it’s no surprise that the sources I surveyed were split on the direction of the sport’s strikeout rate in 2020. Fuld and Thorburn guessed slightly lower; Harry Pavlidis, director of R&D at Baseball Prospectus and founder of Pitch Info, guessed slightly higher, citing higher temperatures, while Dan Aucoin, co-manager of baseball analytics at Driveline Baseball, guessed slightly higher because of the presumed prevalence of bullpenning. Former major league pitcher and current Twins scout Wesley Wright considered “so many variables,” including fan-ban-induced adrenaline deprivation, but he tentatively concurred with Aucoin, saying, “I do think you could see a slight uptick in the average velocity because with the importance that will be put on every game to start the season I think we will see more relievers being used, which should boost average velocities a tad.” Driveline founder and Reds director of pitching initiatives Kyle Boddy, meanwhile, hypothesized that the pitcher-batter balance of power would be about the same, expressing some surprise that the hitters and pitchers he’s seen so far have looked like their usual selves.
I’d lean slightly toward MLB’s swelling strikeout stalling or reversing itself for this season, thanks to the universal DH, the possible avoidance of the first-month strikeout-rate uptick, the elimination of 40-man September rosters, and many players’ atypical training routines. Even if the streak is snapped, though, the NL DH alone—which could become permanent—wouldn’t be enough to restrain the rising strikeout rate for more than one year. And if it takes a pandemic and the associated strangeness of a 60-game season to stop the strikeout-rate increase, then the powers that be at MLB should see it as a sign that the long-term trend toward less contact can’t be contained without a more lasting alteration to the rules.
Thanks to Lucas Apostoleris of Baseball Prospectus for research assistance.
