Using Sabermetrics for Fantasy Baseball: Pitcher BABIP

While FIP is a useful tool to predict a pitcher's future ERA performance, fantasy owners should remember that ERA, not FIP, is what really matters in most formats. This means that we are interested in the "luck" that separates the two statistics.

While some of this luck is unpredictable, we can and should predict some of what goes into a pitcher's bottom line. BABIP plays a big role in the variation of a pitcher's perceived luck, but it may not be as clear-cut as it seems.

Let's get to it!

How to Interpret BABIP for Pitchers

When calculating BABIP for hitters, we assume a neutral defense because they figure to see a balance of poor and skilled defenders as they travel around the league. This is not true for pitchers, as they always pitch in front of their own club's defenders. A team with Victor Robles and his 23 Outs Above Average (or OAA) figures to provide better defense to its pitchers than a team that lacks a premium glover. A better defense helps pitchers sustainably outperform their FIP.

Outs Above Average is a Statcast metric that makes it easier to look at the quality of a pitcher's defense. Outs Above Average, or OAA, measures each player's defensive contributions using Catch Probability. If a batted ball is caught by a player, the player receives OAA credit equal to 1 - the ball's Catch Probability. For example, a successful catch on a ball with a 40% Catch Probability is worth 0.6 OAA (1 - 0.4 = 0.6).

Players also lose points equal to the batted ball's Catch Probability if they flub the catch. Missing the ball in the example above would subtract 0.4 from the player's OAA. Another great feature of OAA is that the stat is sortable based on a shift. If you want to know how a third baseman fares when shifted to a traditional shortstop position, OAA makes it easy to look at that data. Robles led all outfielders in OAA last season, while Javy Baez (19 OAA) took the top spot among infielders.

There are other defensive metrics, but they are much more abstract than OAA while also leaving out important pieces of the puzzle. Ultimate Zone Rating (or UZR) makes no effort to account for the shifts in today's game, rendering it completely obsolete. Defensive Runs Saved (DRS) has one fantasy purpose: measuring the value of a pitcher's defensive contributions to his own cause.

For example, Dallas Keuchel finished second among pitchers in DRS last season despite only throwing 112 2/3 IP. This is nothing new for Keuchel, who has a whopping 50 DRS over his 1,302 career IP. Fantasy owners have known for years that Keuchel posts lower than average BABIPs when he's on despite being a ground ball pitcher, but the reason isn't some magical contact suppression ability. It's the fact that Keuchel rates as roughly double the defender Javy Baez is if you prorate his DRS over a position player's number of innings.

What else impacts a pitcher's ERA?

BABIP is also partially determined by a pitcher's style. An extreme ground ball pitcher may have a higher BABIP against because grounders have higher BABIPs than fly balls (.236 to .118 in 2019.) This stylistic difference also changes how much a given pitcher will benefit from (or be hindered by) a particular defender on his team. For instance, a fly ball pitcher would love to pitch in front of Robles, while a ground ball specialist would benefit more from an elite infielder like Baez instead.

While defense is largely out of a pitcher's control, some pitchers can control their BABIP to a degree. For example, you would probably be tempted to say that the .218 BABIP Justin Verlander allowed last season was a fluke, and you would be partially right. However, Verlander combined a strong fly ball tendency (45.2 FB%) with an above-average IFFB% (12.4%). Flies have the lowest BABIP of all batted balls. Furthermore, pop-ups are rarely hits, so inducing them consistently enables a pitcher to post better than average BABIPs. The combination would be expected to produce a low BABIP allowed.

The same principle holds for pitchers who can limit line drives, but this skill is not quite as sticky as pop-ups. Liners post very high BABIPs but randomly fluctuate, as we have seen in a previous article.

Every pitcher allows a few hits, and the sequencing of these events may also cause a difference between a pitcher's FIP and ERA. Allowing three base hits over three innings is probably harmless, while allowing three hits in one inning and then nothing in the next two frames likely puts a run on the board.

Sequencing luck is measured by strand rate, or LOB%, and research shows that it is largely an unstable, luck-driven stat. In 2019, the league average LOB% was 72.3%, with higher numbers generally forecasting a higher ERA moving forward. Elite strikeout guys tend to be the best at getting the K "when they need it," and as such may sustain slightly elevated strand rates.

Conclusion

To conclude, a pitcher's BABIP includes some unknown variables but also some predictable inputs. The quality of his defense can help or hurt him. Sequencing does not affect BABIP, but can impact a pitcher's ERA substantially. A given pitcher's style, as a ground ball or fly ball specialist, may also impact his performance. If you would like to learn more about other advanced stats, check out this link for a full glossary.