Is 2021 the year of the no-hitter? We’ve seen four no-hitters already, and that’s not even including the seven-inning masterpiece that Madison Bumgarner tossed (I would count it if the MLB did but I want the data to remain consistent). That might not seem like a lot, but it means we’re on pace for nearly twenty: more than we’ve seen in the last seven years, combined. Is this just a fluke in the random noise of such rare events, or is there something more to it? Let’s analyze.

One interesting trend to observe related to the spike in one-pitcher no-hitters is its relation with complete games. The two show a moderate correlation of 0.498, which improves to 0.573 when excluding this year’s anomalous performance. It makes sense, as only complete games can result in a one-pitcher no-hitter, the type that I am addressing. Since 2010, complete games have been on a steady decline. However, after plateauing a bit in 2019, 2020 and 2021 have returned to complete game numbers from the middle of the decade. Why? The past two years, combating modern philosophies that focus on the bullpen and have resulted in innovations such as the opener, there has been a universal DH.

No-Hitters and Complete Games per 4320, 2010-2021 (data via Baseball Reference and Wikipedia) — No-Hitters and Complete Games per 4320, 2010-2021 (data via Baseball-Reference and Wikipedia)

Having a designated hitter affects pitcher longevity in more ways than you might think. While hitting as a pitcher can result in injury or fatigue, one of the primary motivators behind pulling pitchers out is pinch-hitting (and pinch-running). Late in a game, there are often critical at-bats where keeping a pitcher in makes little sense. If they’ve already pitched a lot and getting a hit is super important, most managers would opt to trust their bullpen on the mound as opposed to their pitcher at the plate.

This assumed relationship between the complete game and the designated hitter is backed up very well with historical data. In 1972, the year before the American League added the DH, the League had five fewer complete games than the National League. The next year, with pitchers hitting only in the NL, the AL had 167 more complete games, a bump of 22% from the year prior. With the DH going majors-wide, we’re seeing more opportunities for one-pitcher no-nos than would be expected given the steady decline in complete games in the last few decades.

Of course, this isn’t the only factor. Say that we only had two one-pitcher no-hitters so far instead of four. It would still be the largest conversion rate of complete games to no-hitters since at least 2010. Based on this chunk of data from the 2010s and 2020s, we should have maybe one no-hitter. Instead, we have four. Should this pace continue to the end of the season, a normal bell curve would suggest that the random chance of it happening is all but negligible. Using the same data, we would find that this conversion rate of complete games to no-hitters is more than three standard deviations away from the mean, giving it less than a 0.1% chance assuming no external factors. The chance of one such season happening in a full century would still be less than ten percent.

Conversion Rate of Complete Games to No-Hitters, 2010-2021 (data via Baseball-Reference and Wikipedia)

There does persist one issue, of course: sample size. To adjust, let’s not project the current number of no-hitters over a full season, but instead fill it in with a more reasonable rate. Using the highest non-shortened-season complete game to no-hitter rate to fill in the season, we end up with nine no-nos on 72 complete games. This is still a very high rate but is given a much more reasonable 7.6% chance with no external factors. Over a century, there is a 99.9% chance that a season like this would occur at least once. Still, though, other than the number of complete games, why this season?

Well, a no-hitter is all about limiting hits, and this season has had the fewest hits per game in over a century. Although some of that can be attributed to the seven-inning double-header rule, it is still one of just three seasons under eight hits per nine innings. Other than pitching, one reason is the shift and other infield tactics. The league average batting average on balls in play has gone from .300 just four years ago to .285 this season, a trend well illustrated with this graph:

Hits and Batting Average on Balls in Play, 2010-2021 (data via Baseball-Reference)

Almost no matter how dominant a pitcher’s performance, at least 15 balls will be put into play. While pitchers themselves can obviously influence the batting average on these balls, let’s assume it’s up to random chance. With the .300 batting average on balls in play, 0.475% of complete games with 15 balls put into play would end with a no-hitter compared to 0.653%. The chance of a no-hitter occurring in one of these 15-ball-in-play complete games goes from 9.08% to 13.28%. For the reasons mentioned above, the probability is significantly higher in reality, but this goes to show what a difference infield tactics can make.

Another driver of the no-hitter spike is strikeouts. Instead of relying on balls in play, which will always have a luck component, pitchers are getting batters out at the plate more than ever. In fact, there have been two more strikeouts per nine innings this season than there were just ten years ago. The no-hitters themselves featured 37 strikeouts in 36 innings (according to FanGraphs), or pretty much exactly the league average.

In the end, there’s nothing that I can point out to fully explain the unprecedented amount of no-nos this year. Instead, it’s a combination of small sample size, strikeouts trending up and hits down, and the new designated hitter rule leading to more complete games. At the end of the day, much of it can simply be attributed to a statistical anomaly. Even the rarest things are bound to happen eventually. So far this season, no-hitters have hit that “eventually.” Enjoy it while it lasts.