Skip to main content

2021 MLB Preview: The Art and Science of Pitching

2021 MLB Preview: The Art and Science of Pitching

2021 MLB Preview: The Art and Science of Pitching

The COVID-19 pandemic shortened and distorted the 2020 MLB season, so drawing any overpowering, sweeping conclusions from it is a fool's errand. On the other hand, as we head into the 14th season of the pitch-tracking era and the sixth year of the league's increasingly ubiquitous Statcast system, smaller and smaller sample sizes are able to tell us more and more. Further, many of the trends we saw develop in 2020 reflected or extended ones we'd seen over the previous handful of seasons.

The biggest theme of MLB, at this moment, is the extraordinary pace of change, and last year only magnified and accelerated that. In numerous ways, as the 2021 season dawns, pitching is an ultramodern art form built on the foundations of a recognizable but distinctly different one, like the modern metropolises that now stand on the footprints of great ancient cities.

Never Throw it Straight

Technology has driven much of this transformation, which is why it's so important to keep the advent dates of various tracking systems fixed in memory. The big leagues began logging pitch-by-pitch at-bat information (balls, strikes, swings, takes, foul balls) in 1988. Twenty years later, the next big breakthrough came, when PITCHf/x cameras captured and reported the speed and movement characteristics of each pitch in every big-league park for the first time. Then, in 2015, Statcast went live, bringing not only the same pitch-tracking power, but also the enhancement of reading spin rate, directly measuring (rather than estimating) the exact release point of each pitch in three dimensions, and capturing data on batted balls, as well as certain player movements.

Away from big-league fields, a further technological revolution also took place. A number of innovative firms invested in player instruction and development and began using more advanced motion-capture and radar technology to measure pitchers' movement throughout their deliveries, providing instant feedback a hurler could use to make tweaks within a bullpen session. From all of this data came some crucial information: Pitchers threw too many fastballs.

The more data that poured in, the clearer that became. The league began ditching the sinker around the time of Statcast's advent as teams gained the ability to reliably measure and compare pitchers' spin rates and could thus acquire pitchers with rising action on their four-seam fastballs, which leads to more swings and misses than does even a great sinker. From 2014-19, usage of the four-seam fastball rose from 35.3 percent to 37.4, while sinker usage plummeted from 21.4 percent to 15.0.

As those numbers suggest, however, there was another creeping realization going on alongside the growing distinction between fastball varieties: Often, the best pitch in a given situation wasn't any type of fastball, and in fact, fastballs could (in some cases) work best as off-speed pitches. If a pitcher could force opposing batters to sit on his slower stuff, he could throw the heat past those hitters more easily.

In 2020, with plenty of downtime before the season in which to refine their breaking balls and changeups, but without the opportunity to build up arm strength the way they normally would, pitchers put this evolving strategy to its most extreme use yet. League-wide, four-seam fastball usage dropped back to 2014 levels, while sinker usage was essentially flat. Meanwhile, sliders (18.8 percent of all pitches), changeups (11.7 percent), and curveballs (10.7 percent) all enjoyed their highest recorded usage ever. In each case, that is part of a multi-year trend, but the pandemic-altered season nudged the frontier forward.

The analytical underpinning for the shift lies in the league's obsession with strikeouts. As research has demonstrated, racking up whiffs is worth the extra walk or fly ball, from time to time, as long as a pitcher can sustain the former, and modern pitchers have realized that they can best do so with non-fastballs as their primary weapons. Yu Darvish, who finished second in NL Cy Young Award voting, had never thrown his four-seamer less than 25.9 percent of the time, until 2020. Last year, he threw it just 16.6 percent of the time. Because batters had to sit on his slider (46.7 percent), though, they whiffed at a jaw-dropping rate when Darvish did throw his pure heater. He'd never before topped 29.3 percent in whiff rate on the four-seamer, but in 2020, it was 42.3 percent.

Meanwhile, reliever Matt Wisler all but abandoned the fastball, with sterling results. The former top prospect, who had become a journeyman and was claimed on waivers by Minnesota just after the end of the 2019 campaign, has the ability to shape and locate his slider such that it behaves almost as two different pitches, so he threw it 83.4 percent of the time in 2020 — a record, for any pitcher with more than 20 innings pitched in a season — and (like Darvish) used the fastball as a change of pace. Wisler had a 1.07 ERA on the short season.

The only three pitchers to throw changeups half the time in a season, since pitch tracking began, all did so within the last two seasons, including 2020 NL Rookie of the Year Devin Williams. Minnesota's Tyler Duffey threw his biting curveball 56.7 percent of the time, a record.

Hitters want to hit fastballs. Great velocity can still beat them on that pitch, but for the most part, they want to hit fastballs because those pitches fly relatively straight. They can find the ball, track it better and make harder contact on it. Pitchers have used a bevy of tools to learn and hone non-fastballs, and now, they're refusing to give hitters what they want.

No Hit Parade

In 2020, the league hit .245, the lowest overall mark since another shortened season, 1972. After that campaign, the American League instituted the designated-hitter rule. Last season was the first ever played under DH rules in both leagues, yet, by one measurement, offense was back to pre-DH levels.

It's not that simple, of course. Batting average is no longer the offensive currency of baseball. In 1972, the league had a .311 on-base percentage and a .354 slugging average to go with their ugly batting average. Last year, those figures were .322 and .418. The modern game is very much about payoffs, rather than probabilities: Games are won not by singles and sacrifice bunts, but by walks, doubles, and home runs. The league is selecting hitters better than ever, and teams are even able to put great hitters at positions (especially the middle infield spots) formerly reserved for lithe, light-hitting glove men.

Still, the falling global batting average tells us something of which we must not lose sight: The progress of baseball, at this stage, continues to be a march toward the death of offense as the game has known it for most of its history. The average strikeout rate continued its uninterrupted two-decade climb, hitting 23.4 percent in 2020. Walks were up, and the ball remains lively, helping to boost power and (thereby) propping up offensive numbers, but as long as there remain 60 feet and six inches between the plate and the mound; 90 feet between each base and the next; and 30 teams in the league, pitchers will continue to steadily widen their edge over batters.

Defensive shifts play a role in that. Even without them, though, the odds would be stacked against hitters. As the league gets more athletic, on average, hitters will be able to hit more home runs (all else being equal), but they won't be able to collect more hits on balls in play. The increased benefits of hitting the ball a bit harder on the ground and on a line are washed out by the better athleticism of fielders, data-informed positioning, and the relentless rise of the strikeout.

Most of MLB's history has been defined by a perpetual scarcity of good pitching. Now, coming up on a quarter-century after the last expansion in 1998, that scarcity is fading. Population growth, expanding international markets, better player development, improved sports medicine (especially when it comes to treating arm injuries), and a wave of new information and technology have made even 13- or 14-man pitching staffs chock-full of hard-throwing guys with acceptable control and above-average breaking balls. Hurlers are forcing hitters ever further into a corner, and without a change to one of the vital dimensions of the game, that will continue, even if the ball keeps flying and runs continue to cross the plate at reasonably normal rates.

Lower and Lower

To get maximum value out of all of the good arms available, though, teams have had to change the way they use their staffs. Thanks to the pandemic, we saw that trend accelerate in 2020 as well. Starting pitchers are becoming, if not an endangered species, at least a marginalized population. For several years, the modern bullpen has been cannibalizing the traditional starting rotation, and it took its biggest bite yet last year.

Starting Pitchers Percentage of Batters Faced by All Pitchers, MLB


Contact %













In the 1990s and 2000s, starters still faced two-thirds of all batters, even as the last generation of old-fashioned workhorses grew old and departed the scene. Now, that share is barely over half, and it is likely to nose even lower in 2021.

The odd structure of the pandemic season, with its bifurcated preseason preparation period and condensed schedule, led to increased injury rates. This year, experts anticipate teams to be cautious about starter usage, especially for young or oft-injured hurlers, because (in addition to the other issues) no one was able to rack up the innings totals in 2020 that would typically keep them strong and ready for the following campaign.

One reason why the league relies so much less on fastballs now is that more of the pitches are being thrown by people who know they won't see an opposing hitter more than once or twice in a game. There's less to be gained by holding anything in reserve than there used to be, even for starters, and relievers have never had to worry much about that, anyway. As teams, players, agents, and outside analysts have come to the conclusion that pitching while tired and needing to get opponents out multiple times within a single game make a pitcher less effective, they've made the logical decision to focus on giving each arm on a staff a few good pitches, having them blend them evenly, and getting them out of the game as soon as logistics and rules allow it.

To be sure, something is lost, aesthetically. Narratives about toughness and endurance were, perhaps, always overblown, and the individualistic heroism of a no-hitter or a complete-game playoff gem has always been a bit of an illusion. Run prevention is a team effort, after all. Still, there are dynamics created by having a starter pitch deep into a game that give the contest extra depth.

A pitch in one at-bat sets up a pitch in another. A hurler learns pinpoint command of his fastball so that he can induce weak contact with it when needed, to save bullets, and have something left when the game is on the line in its late stages. A batter and a pitcher develop a longer and more meaningful log against one another, probing for weaknesses, trying to surprise one another, despite facing off in 10 plate appearances a season. The fifth pitch in a pitcher's repertoire can be thrown a handful of times in a game, but to enormous effect, if the spots are chosen correctly.

These nuances all make pitching more interesting, but in today's bullpen-mad version of the game, many of them are rendered moot.

— Written by Matt Trueblood (@MATrueblood) for the Athlon Sports 2021 MLB Annual. At 224 pages, it's the largest on the newsstand and the most complete preview available today. Click here to get your copy.

(Yu Darvish photo courtesy of @Padres)