Can Statcast's Expected Weighted On-Base Metric Help Predict Future Fantasy Stars and Busts?
Editor's Note: For this piece, we are integrating Genius anotations to define advanced statistics that may be unfamiliar to the everyday reader. The advancements made by Statcast, Fangraphs and Baseball Prospectus (among others) are fundamentally altering the way that organizations evaluate players and construct rosters. Definitions for these advanced stats are derived from the Statcast Glossary unless otherwise noted (the Fangraphs Glossary is another essential resource).
Statcast is amazing. The leaderboards are invaluable tools for anyone fascinated by baseball analytics. From a fantasy perspective, Statcast data is relatively untested compared to old mainstays, but the trove of data offers plenty of insight.
This piece focuses on Expected Weighted On-base Average (xwOBA), a stat that calculates what a player’s Weighted On-base Average (wOBA) should have been given his contact tendencies (or, inversely, a pitcher). Importantly, this is not designed to be a predictive tool. In order to make it predictive, one must find the difference between xwOBA and actual wOBA to find players who are under/over performing contact profile. Even then, in many cases, hitters are likely just underperforming their batted balls in play (BABIP). However, xwOBA is an excellent shorthand for arcane batted-ball numbers, and with a little interpretation, analysis and old-fashioned scouting, significant xwOBA/wOBA differences give us a solid list of players who made very good contact in 2018 without commensurate results.
This piece will focus on hitters. Baseball Prospectus’ Jonathan Judge showed xwOBA data is only slightly better than Fielding Independent Pitching (FIP) as a predictive tool (he also ran tests on batters and found little significance, but I think there’s more value to the stat on the hitting side). However, the expected stats function as another potential analytical tool, and at the very least is a different way to perceive batted-ball profiles.
xwOBA Regression Candidates
Eddie Rosario, Twins:.041 difference wOBA-xwOBA, (.288/.323/.479)
Rosario is going in the top 100 of most drafts after a solid 2018 with 24 HRs, nine steals and 87 runs scored. There’s plenty to like about Rosario, who joined in on baseball’s fly ball revolution last year, but the 27-year-old’s Statcast numbers are not on the list. His career xwOBAs have either tracked below or even with his wOBA, which makes 2018 a major outlier. Rosario’s BABIP was a solid .312, and his average was still high, but the contact he made was not indicative of that. The jury is out here, but the numbers state he’s a little overvalued.
Miguel Andujar, Yankees: .038 difference wOBA-xwOBA, (.297/.328/.527)
Andujar is already a polarizing real-life player, but he was an excellent fantasy asset in 2018. Andujar was extremely aggressive: His 4.1% walk rate and 16% strikeout rate was not ideal, but he made up for it with 27 homers and a .297. His .316 BABIP was a tad high, but it doesn’t scream regression. His wOBA-wxOBA split is not as promising, and his expected slugging percentage (xSLG) difference is even worse relative to the league. His infield fly ball rate is 14.5%—over four percentage points above league average—indicating that he’s prone to harmless infield pop-ups. We have no data suggesting that Andujar can sustain his 2018 performance. That’s not to say he can’t, but it’s worth monitoring.
Adam Eaton, Nationals:.031 difference wOBA-xwOBA (.301/.394/.411)
Adam Eaton is not going very highly in drafts this year (198.3 ADP on FantasyPros), and his batted ball statistics are not going his way either. Eaton has declined in hard-hit percentage and exit velocity, and the poor xwOBA numbers reflect that. He has always been able to maintain a high BABIP and get on base, but his increased power in 2015-16 is gone, and he’s on the wrong side of 30.
xwOBA Breakout Candidates
Gary Sanchez, Yankees: -.036 difference wOBA-xwOBA (.186/.291/.406)
After his disastrous 2018, the narrative of Gary Sanchez improving on his .197 BABIP and recovering from numerous injuries is established. It is worth noting, however, that his batted ball profile supports the signs. He was better than league average on all of his expected numbers. Improved health and better luck should follow.
Ryan Braun, Brewers: -.038 difference wOBA-xwOBA (.254/.313/.469)
Braun may be 35, but his batted ball profile suggests he’s as good as ever. Unfortunately, his overall numbers were mediocre last year. Braun posted his best hard-hit rate and expected batting average in the Statcast era in 2018, but failed to gain any appreciable results for it. His .274 BABIP was not good, but even with that, Braun deserved a little better from 2018 and could be primed for some improvement in 2019.
Kendrys Morales, Blue Jays: -.044 difference wOBA-xwOBA (.249/.339/.438)
Inexplicably, Kendrys Morales hit the ball as hard as ever in 2018, but had worse results. His age and reduced playing time and age don’t portend increased success, but I suppose you can’t argue with a Statcast-best hard-hit rate and an exit velocity in the top 6% of the league. Morales is likely to go undrafted for good reason, but I wouldn’t be shocked to see a random hot streak at some point.
A Pitcher to Watch
Brad Peacock, Astros: .035 difference wOBA-xwOBA
Peacock should already be on everyone’s radar as he has emerged as the frontrunner for the Astros’ fifth starter spot this spring. Peacock had a 2.82 xFIP in 65 innings of middle relief last season, and his Statcast numbers argue that he was even better than that. The contact against him suggested he was significantly better: in addition to the xwOBA split, his expected slugging against of .346 was 85 points lower than his actual slugging against of .431. His strikeout rate improved from 10.98 to 13.29, and his walk rate fell by a walk per inning. Even if Peacock’s strikeout rate and other peripherals decline slightly as he gets stretched out, he is very undervalued in drafts.
