As baseball has eased its way back into our lives, it’s around the time that writers and fans release bold predictions about the Major League Baseball season. While that isn’t the point of this article, I will share one player I believe is due for positive regression, one I believe is due for negative regression, and how we can back up those claims with expected outcome statistics from MLBAM’s Statcast.
Statcast expected statistics are based on quality of contact (exit velocity and launch angle) and amount of contact, not outcomes. Each batted ball has an expected batting average (xBA), expected slugging percentage (xSLG), and expected weight on base average (xwOBA). The “x” before a metric means it is an “expected statistic.” An expected statistic removes luck to better understand the quality of a player’s skill. Strikeouts, walks, and hit by pitches are then factored in to find a player’s expected outcome over the course of a season. In 2019, there was one player that stood out to me that had a significant factor of bad luck in his batted balls.
Justin Smoak of the Milwaukee Brewers is likely to collect a significant amount of plate appearances at first base. Smoak is a switch-hitter who has fared much better from the left side as of late, posting a .260 xwOBA as a right-handed hitter, but a .385 xwOBA as a left-handed hitter (explanation on xwOBA coming below). I would imagine a platoon role is in play. On the bright side, he is a candidate for a bit of positive regression.
With the Toronto Blue Jays in 2019, Smoak posted a .323 wOBA (.320 league average), but an xwOBA of .366. wOBA (weighted on base average) takes a similar approach to on base percentage but weights each aspect of hitting in proportion to their actual run value. To learn more about wOBA, read more on FanGraphs here. Additionally, xwOBA is wOBA’s expected statistic counterpart and is a popular metric to value a batter’s skill.
On a tangent from xwOBA, Smoak played most of his games in Toronto last season, which has a home run park factor of 101 for left-handed hitters, whereas that number in Milwaukee is 112 (both numbers from FanGraphs’ 2018 Park Factors). As I mentioned previously, Smoak is a switch-hitter but with right-handed hitter Ryan Braun in the fold, Smoak figures to be utilized almost exclusively against right-handed pitchers.
As of 2019, Statcast includes sprint speed as a variable for certain types of batted balls (topped and weakly hit), which removes bias for players with the ability to beat out infield batted balls. Smoak himself is not a speedster, so this won’t matter too much in this case. Despite that, his wOBA has underperformed his xwOBA four times in the last five years, and by no small margin (in 2019 it was the largest at a 0.043 difference). One question that would be interesting to research is how different batted ball profiles affect expected outcome statistics. Smoak’s Statcast similarity score aligns closely with Alex Bregman, Juan Soto, and Bryce Harper, among many others. For Brewers fans, there is a lot to like here.
Ketel Marte of the Arizona Diamondbacks posted a similar xwOBA (.369) to Smoak (.366) but experienced a much different 2019 season (.405 wOBA). Like Smoak, Marte is a switch-hitter, but somehow posted identical .369 xwOBA values from each side of the plate. In the scatter plot below from Baseball Savant, you’ll notice that the two players share a similar position on the y-axis but have a drastic difference on the x-axis. Did Smoak have an unlucky season or did Marte experience a lucky one? It’s a loaded question and not one that can be fully explained in this article. The short answer is that both are due for regression to the mean.
Marte enjoyed a bit more success than his xwOBA suggests he should have, but there are positives to look at, such as career highs in hard-hit percentage (39.4%), barrel percentage (9.3%), exit velocity (90 mph), and launch angle (11.3 degrees). On top of that, he accumulated 7.1 fWAR (FanGraphs WAR), which ranked sixth amongst all position players in baseball. Marte had a career year and deserves the recognition.
While the metrics listed above are career highs for Marte, elite players are putting up slightly better numbers each year. Marte’s profile could potentially be exploited once teams figure him out. Now, this isn’t to say that Marte will have a poor 2020 season because a .369 xwOBA is still well above league average. Nonetheless, expectations for another 7.1 fWAR season might need to be tempered.
It’s interesting to compare and contrast situations where players like Smoak and Marte finish a season with similar expected metrics, but what matters at the end is results. Can their xwOBA values alone back up my claims that both are due for regression to the mean? Well, no. It’s important to note that xwOBA does not have much predictive power from year to year. Expected outcome statistics are more indicative of a player’s skill than their result-oriented counterpart. While defense is removed, variability is controlled for, and only measures such as exit velocity and launch angle are considered, these are still just results of two players performances during 2019, and not highly correlated to a successive season’s performance.
Anyways, let’s enjoy the 60-game season as we are fortunate to turn on the television and watch live baseball. Wear a face covering, wash your hands, and distance from others when possible!
|Crafting a Gameplan… on The Jordan Hicks Dilemma: A Qu…|
|SB on Zach Davies: What Changed?|
|Can Detroit Tigers B… on Improving Pythagorean Winning…|
|Avoid the Dead Zone:… on An Analysis of Jakob Junis, Ar…|
|Predictive Fitness a… on Press Release: BaseballCloud A…|