Clearly, a pitcher’s skills determine whether he’s an ace starter or a swingman/sixth starter. However, there are a few elements of performance which are consistent, irrespective of talent level. That said, when a round bat meets a round ball, variance ensues, exposing the pitcher to good and back luck.

A few weeks back, several ERA estimators were discussed. Now, a few factors contributing to fortune or misfortune will be reviewed. In simple terms, we have already explained that a pitcher can be lucky or unlucky. Next up, we’re going to show the reasons, beginning with BABIP (Batting average on Balls in Play).

BABIP with respect to hitters was previously discussed. By means of review, it’s standard batting average, but leaving out homers and strikeouts while including sacrifice files:

BABIP = (Hits – homers)/(At bats – homers – strikeouts + sacrifice files)

Several years ago, Voros McCracken noted that pitchers tend to settle around the same BABIP, regardless of their skill level. Pitchers posting a BABIP lower than league average were considered lucky, and their BABIP the following season was expected to rise. Those with a BABIP higher than league average were unlucky, with the anticipation it would fall the ensuing campaign.

Later, it was fine-tuned to different batted ball types sporting different a different average BABIP. Line drives being the highest, then ground balls with fly balls checking in as the lowest. As such, a fly ball pitcher could consistently register a lower BABIP than a ground ball pitcher.

Intuitively, a pitcher’s ability to avoid yielding line drives should help suppress BABIP. However, research indicates pitchers have very little control over their line drive rate, whereas they are largely responsible for their ground ball or fly ball tendencies.

Below is a chart illustrating the range of expected BABIP based solely on batted ball distribution. The BABIP is essentially a weighted average of the ball in ball in play components. The component BABIP are from the 2023 season and are strictly from starting pitchers. For simplicity, bunts are counted with grounders and popups are included with fly balls.

The top line is the league average distribution from last season. Line drive rate is kept constant since it is very much out of the pitcher’s control. A rate below 25.2% can be considered lucky, helping to lower BABIP. A level above 25.2% indicates the hurler was unlucky, pushing BABIP up.

As an example of how to use this data, Logan Webb’s ground ball rate using this methodology (which excludes homers since we’re just focused on balls in play) was 64.9%, so he should be expected to sport a high BABIP. However, his line drive rate was just 20.8%, lower than average. His xBABIP is .305, as compared to the actual .302. Webb’s BABIP reflected his batted ball distribution almost perfectly.

That’s backwards looking. Our objective is projecting Webb’s BABIP for the 2024 season. The ground ball rate was his career-high, so it’s best to anticipate a slightly lower mark, let’s call it 59%, which is in line with what he’s done over the past three seasons. A pitcher doesn’t have complete control over his line drive rate, so let’s split the difference between what he yielded and the league average and call it 23%. This renders a .308 projected BABIP. There was a time projecting a BABIP 11 points over league average would be considered foolish but breaking it down by components suggests otherwise.

Marcus Stroman also sported an extreme ground ball rate of 58.7% using this process. His line drive rate was 24.2%, just under the league average. His xBABIP is .314, much higher than the actual .283 mark. On the surface, this screams good luck. However, Stroman has generated a BABIP well below average for the past three seasons. He may have been fortunate for three seasons, but not to this extent. Stroman’s ground ball BABIP last season was .174, 74 points below the league average.

Some could be good infield defense, but some can also be Stroman inducing soft contact. Unfortunately, Stroman’s ground ball HardHit% is above league average. He appears to have been blessed by Lady Luck, perhaps for the past three seasons.

There may be something in Stroman’s underlying numbers not yet captured by next level analysis. Statcast uses next level methods. His xBA was .265, compared to the actual .235. This indicates Stroman yielded seven fewer hits than expected. Adding back those seven hits elevates his BABIP to .301, which is still lower than his .314 xBABIP, but now we’re within shouting distances.

For projection purposes, despite posting a BABIP of .286, .272 and .283 over the past three seasons, the prudent expectation is in the .300 neighborhood. His ERA will still be low, since his ground ball lean keeps the ball in the yard, but Stroman’s WHIP should suffer.

What about a hurler on the other end of the spectrum. Using this methodology, Cristian Javier posted a 27.8% ground ball rate and 47.4% fly ball rate. His actual BABIP was .272, compared to a .270 xBABIP. It wasn’t long ago that most everyone would run away from Javier expecting steep BABIP regression towards league average. However, his low BABIP was completely supported by his batted ball distribution. Of course, homers are usually an issue with so many fly balls, but that’s a story for another day (this time next week, for those keeping score at home).

Some of you are probably wondering about the elephant in the room. It’s been referred to a couple of times, but let’s address it before calling it a day (and year). Very little of the discussion involves the pitcher’s ability to influence BABIP. The two most obvious possibilities are minimizing line drives and inducing weak contact. Current studies indicate that a pitcher’s line drive rate is random, with perhaps a small level of control. Discerning how much a pitcher induces soft contact is still a work in progress. The extent to which a pitcher generates grounders and fly balls is greater than the scope he can bring about soft contact. The holy grail is, how much greater?