Toronto Blue Jays xwOBA over and underperformers

xwOBA is a new stat that uses Statcast data of launch angle and exit velocity to determine a player's expected wOBA. It attempts to strip luck from the equation and focus solely on the quality of contact made by a hitter. A more complete description can be found here. If you're unfamiliar with regular wOBA you can find a primer here. xwOBA data can be found at Baseball Savant. It should be noted that Baseball Savant's wOBA numbers differ slightly from those found at Fangraphs, but not considerably.

A player's speed can factor into whether they over or underachieve with wOBA. Slower players are more likely to underachieve because they are less likely to turn weak hits into infield singles or hits in the outfield into extra bases. It doesn't always explain the difference, but it's a factor worth noting. Ballpark factors may also play a role.

With that said, Justin Smoak underachieved in the 2 years before breaking out in 2017 so there may be some predictive value in the stat. If a player is hitting the ball hard and at the right angle it's only a matter of time before their results catch up.

Among players with at least 400 at bats Kendrys Morales was the 2nd biggest underperformer in baseball last season according to xwOBA. His xwOBA came in at .359 vs. his actual wOBA of .326. The .037 gap was topped by only Miguel Cabrera. If Morales had performed closer to his xwOBA he would have produced a stat line closer to Jay Bruce's .254/.324/.508 rather than the .250/.308/.445 that he actually produced. Morales is probably a good candidate to exceed his 2017 production. Morales had the 11th highest average exit velocity among batters with at least 250 recorded results.

Devon Travis also had a large gap between his xwOBA and wOBA. His .311 mark lagged behind the expected .338. If his results fell closer in line with his batted ball quality he may have produced a batting line similar to Adam Jones' .285/.322/.466. If Travis continues to hit the ball hard and stay healthy in 2018 we could see 30+ doubles and something around 20 home runs.

Russell Martin produced a batting line that was quite a bit better than the average catcher and may have even had slightly bad luck. His xwOBA trailed his wOBA by .010. Martin isn't likely to see a huge uptick in hitting performance in 2018, but if he keeps hitting the ball with the same authority as he did in 2017 we should see similar results.

Justin Smoak's numbers had just a .005 gap between them which suggests that Smoak could be in for another quality season in 2018. Hopefully he doesn't return to his xwOBA underperforming ways. Still, his .383 xwOBA led the team last year, so even if he regresses some, he should be a solid hitter in 2018.

On the overachiever side of things we have Teoscar Hernandez. His actual wOBA of .377 was way higher than his expected .317. If Hernandez had performed closer to his expected numbers his .261/.305/.602 line would have looked closer to Chris Davis' .215/.309/.423. This seems reasonable because it is unrealistic to expect Hernandez to maintain a .260 AVG with a 37.9% strikeout rate and his 30.8% HR/FB rate is not going to be repeated. Hernandez does barrel the ball often and has very good exit velocity, but he doesn't make contact nearly often enough if he is to expect to have long term success as a hitter in the majors. Hernandez has a lot of tools that make a great hitter, but clearly still needs refinement and that's likely why the Blue Jays went and got some outfielders rather than counting on him to repeat his scorching September performance.

Ezequiel Carrera overproduced nearly as much as Hernandez did. He had a -.052 gap between his wOBA and xwOBA. If he had hit more like his xwOBA expected him to then his numbers would have looked similar to Billy Hamilton's .247/.299/.335 except without the elite defence and stolen bases. 2017 was the first time Carrera had ever produced a batting line above the league average and anybody expecting a repeat performance is likely to be disappointed.

Based on the players above it would seem that xwOBA does have some value in determining who should have produced better or worse than they really did. It's a new stat that I am still getting familiar with and I look forward to seeing what research is done on the subject and how we can better use it to predict future performance.