Cavan Biggio- Credit: DaveMe Images

Blue Jays hitters hold up pretty well to Predictive wOBAcon analysis

The Toronto Blue Jays could run out a lineup that produces quite well according to analysis of predictive metrics





Anyone who has read a few of my posts is likely aware of my love of Statcast’s x-stats (xwOBA, xwOBAcon, xBA, xBABIP, etc.). They are an interesting set of tools that can help one compare how a batter’s results at the plate compare to the results their batted balls earn on average (based on exit velocity and launch angle, but not launch direction). It’s worthwhile to know whether a given fly out was the kind of batted ball that is almost always a fly out or the kind of batted ball that often generates extra-base hits.


However, when using these x-stats, it’s vital to be careful how one interprets and discusses them. While they tell us something useful about a player’s past performance, they are not designed to be predictive and are, thus, imperfect predictors. Tom Tango, MLB Advanced Media’s Senior Data Architect (and a great Twitter follow), examined precisely this idea in a blog post on Sunday, titled “xwOBAcon v Predictive wOBAcon”.


As an example, Tango compared two hypothetical batted balls: one with an exit velocity below 95 mph and a 28 degree launch angle; and another with a 95 mph-plus exit velocity and a -4 degree launch angle. Essentially, a weakly-hit air ball vs. a hard-hit grounder. While these two balls would have roughly the same xwOBAcon (.269 for the weakly-hit air ball and .287 for the hard-hit grounder), what they say about a hitter’s potential going forward is noticeably different — Tango calculates a .369 Predictive wOBAcon (pwOBAcon) for the weak air ball vs. a .579 pwOBAcon for the hard grounder — as an ability to hit the ball hard is more predictive of future success than hitting balls in the air.


We can take this idea to the maximum extent with another example. On average, a ball hit with a 52 degree launch angle generates poor outcomes regardless how hard it’s hit. When these very vertical batted balls have a 95 mph-plus exit velocity, they have a xwOBAcon of .019, whereas their weakly-hit cousins have a nearly-identical xwOBAcon of .024 (slower is actually better in this case, probably because these balls stay in the air for less time, making a fly out less likely).


However, these two batted balls actually have the highest (1.738) and lowest (0.075) pwOBAcon among the 16 that Tango analyzed (same average outcomes, completely different predictions). A weakly-hit pop up has two suboptimal traits, bad launch angle and a low exit velocity, so batters who generate a lot of them are less likely to generate batted balls that have good launch angles and/or high exit velocities. On the other hand, as Tango notes, “it takes an immense amount of power to mishit a ball at such a horrible angle that you can STILL launch the ball off the bat at 100 mph”. It’s funny to think that, if you had one batted ball with which to assess a batter’s talent, the one who hits a 100 mph pop-up would be the most likely to experience future success.


Tango concludes his post by offering readers four useful categories for thinking about the predictiveness of batted balls by exit velocity and launch angle:

● balls hit 95+, at 24+ degrees: these are the best indicator of a batter’s talent

● balls hit 95+, under 24 degrees: these are a very solid indicator of a batter’s talent

● balls hit under 95, at 8-32 degrees: this tells you how average a batter is

● balls hit under 95, outside of 8-32 degrees: this tells you everything bad about a batter, mishitting on both speed and angle


Put simply, it’s a really good sign when a hitter hits the ball hard often. It’s an especially good sign when a lot of those hard-hit balls have a meaningful launch angle. However, it’s a very bad sign when a hitter is generating a lot of weakly-hit balls with either high or low launch angles.


With an eye on predicting how various Blue Jays will hit in the upcoming season, I though it’d be interesting to examine how their 2019 batted balls fit into these categories and how they compared to other major leaguers. [In order to ensure that there is a meaningful sample size for each batter, I’m focusing only on those who made at least 150 plate appearances in the majors in 2019.]


Cavan Biggio

Biggio is the Blue Jays’ stand-out within this batted ball framework, producing an absurdly-high rate of hard-hit balls with a 24 degree-plus launch angle in 2019, which lead all Jays hitters and ranked among the league leaders. In fact, he was one of only three qualified batters across the majors (alongside Joey Gallo and Matt Carpenter) who hit more hard-hit balls with a launch angle above 24 degrees than hard-hit balls with a sub-24 degree launch angle. He also hit relatively few weakly-hit balls with a particularly low or high launch angle, ranking in the 3rd percentile league-wide.


Put simply, in his rookie season, Biggio produced a great deal of the best kind of batted balls and a relatively low number of the worst kind of batted balls. This strong batted ball profile, combined with his above-average walk-to-strikeout ratio and above-average second base defence, makes him the kind of young major leaguer worth betting on to have a strong sophomore season. Projection systems have him as a roughly-average hitter in 2020, but this data makes me even more confident that he can be an above-average regular this year and beyond.


Vlad Guerrero Jr.

Vlad’s batted ball numbers serve to corroborate what countless people have said over the last number of months: he often hits the ball hard, but too often at low (and less productive) launch angles.


Vlad produced a ton of hard-hit balls last season (27.3% of all plate appearances), even more than Biggio did (22.7%). However, whereas Biggio’s hard-hit balls had the loft to do some serious damage, Vlad’s often did not. That dichotomy reflects the two young players’ status in baseball: Biggio has the lower ceiling, but has already produced near that level; Vlad has a sky-high ceiling, but needs to make some changes to get there.


A point of optimism lies in Vlad’s physical development over the off-season. Last month, Shi Davidi wrote about how Vlad’s improved conditioning could help him offensively in 2020. In particular, it seems Vlad was tiring as the season progressed (something Ben Nicholson-Smith wrote about in January).


Hitting coach Guillermo Martinez noted that “the fatigue led [Vlad’s] body to work inefficiently, with negative body movements keeping him from consistently putting his best swing on the ball”. In contrast, Vlad is now “mechanically sound, he’s way more aware of his body right now, and he’s going to be able to make adjustments. He has a better idea of what his body is doing.”


Vlad’s batted ball profile in September 2019 certainly looks like that of a young man who was physically tiring after a long rookie season. Through the end of August, Vlad produced 0.71 hard-hit batted balls for every weakly-hit ball, better than the league average of 0.63. In September, for each of his weakly-hit batted balls, he produced only 0.36 hard-hit balls. Hopefully, with better fitness comes both improved body control at the plate from Opening Day and improved stamina in the later months.


Bo Bichette


Bo had the smallest 2019 sample size of the Jays’ Big Three, but there’s a lot to like. While his rate of hard-hit balls with loft was only league-average, he produced a ton of hard-hit balls with shallower launch angles. In fact, his overall rate of hard-hit balls (29.2%) surpassed even Vlad’s, a sign that Bo should continue to produce a lot of extra-base hits going forward.


Moreover, when Bo produced a weakly-hit ball, a relatively high proportion of them had intermediate launch angles, while a relatively low proportion had particularly low or high launch angles. The implication is that Bo does a good job of generating batted balls in the “sweet spot”. That, coupled with his ability to regularly generate high-exit velocity batted balls, is a predictor of future success.


Danny Jansen


Jansen’s numbers were probably the most surprising among Jays hitters. He produced a below-average ISO (.153) and BABIP (.230) in 2019, so I wasn’t expecting much from his predictive batted ball profile. However, it turns out that he produced a great deal of 95 mph-plus batted balls (27.6%), edging out Vlad.


Like Vlad, Jansen just didn’t get meaningful production from his hard-hit and elevated batted balls. Among 245 major leaguers with at least 20 batted balls in the first category, Jansen’s wOBAcon (.723) ranked 220th, while Vlad’s (.628) ranked 233rd. It’s hard to say exactly why this was, but it does suggest that both vital members of the 2020 Blue Jays should see improved batting performances in the coming season.


Teoscar Hernandez


Over the 2019 season, Teoscar had a pretty solid batted ball profile, ranking highly in terms of the batted balls predictive of strong future production, while also producing a relatively low rate of weakly-hit batted balls. His outfield defence remains a big question mark, but his ability to hit the ball hard does not.


Teoscar was a much more effective hitter after a stint in Buffalo, posting a 47 wRC+ beforehand and a 126 wRC+ afterwards. This difference is reasonably apparent in his batted ball profile as well. Pre-demotion, 7.8% of Teoscar’s batted balls were hard-hit and elevated, 12.8% were hard-hit but not-so-elevated and 37.6% were weakly-hit. After his recall, each of those numbers trended in the right direction: 9.3% were hard-hit and elevated and 15.2% were hard-hit but not-so-elevated, while only 28.5% were weakly-hit. In essence, not only was Teoscar a more productive hitter after his time in Buffalo, he also displayed the kind of batted ball profile that predicts success going forward.


Lourdes Gurriel Jr.


Overall, Gurriel had a productive 2019 season, evidenced by both his full-season 124 wRC+ and strong batted ball profile, underlined by his high rate of hard-hit batted balls (particularly those of the elevated variety). One knock on Gurriel was he produced a league-average rate of weakly-hit batted balls with extreme launch angles.


Gurriel’s 2019 had a similar arc to Teoscar’s, with a slow start (40 wRC+ through mid-April), followed by time in Triple-A, followed by a strong rest of season (137 wRC+ from May 24th onward).


Moreover, like Teoscar, Gurriel saw a meaningful change in his batted ball profile before and after his time in Buffalo, suggesting that his stronger post-recall results are sustainable. Gurriel hit hard-hit batted balls with elevation more than twice as often after his recall (10.7%) than before (4.5%). His rate of hard-hit balls with a sub-24 degree launch angle similarly more than doubled after his Triple-A stint, rising from 9.1% to 20%. In contrast, he produced a much lower rate of weakly-hit batted balls with extreme launch angles post-recall (27.4% vs. 43.2%).


Randal Grichuk


Like the other two regular 2020 outfielders, Grichuk’s 2019 batted ball profile was fairly good, with an above-average rate of hard-hit batted balls the main positive. His league-average rate of weakly-hit batted balls with extreme launch angles is the main negative.


Looking ahead, I have a (hopefully-not-misplaced) belief that Grichuk could have a very strong 2020 season, based largely on how he finished the 2019 season.


The most consistent knock on Grichuk is his high strikeout rate, evidenced by his 28.3% career mark. In a Birds All Day interview posted on June 21st, Ross Atkins said that Grichuk was actively working to cut down that strikeout rate. In early September, I followed up on that thought and found that Grichuk had maintained a 22.6% strikeout rate from June 21st to September 8th.


Over 323 plate appearances from June 21st through to the end of the season, Grichuk ran a 22.9% strikeout rate. He also hit a relatively high rate of hard-hit batted balls, whether with a 24 degree-plus launch angle (8.7%) or not (21.1%), producing the 43rd-most hard-hit balls in the majors over that stretch. A 2020 season with a close-to-league-average strikeout rate, his usual power and average CF defence would make Grichuk’s contract a steal.


Travis Shaw

Shaw is the lone batter in this post whose entire 2019 sample was with another team. Shaw was a relative steal for the Jays this off-season — the Jays are paying him $4 million this season and he’s arbitration eligible in 2021 — because of his very rough 2019 season with the Brewers (47 wRC+ over 270 PA).


The Jays are believers in Shaw as a bounce-back candidate and his batted ball profile backs that confidence up. On the one hand, Shaw generated an above-average rate of hard-hit balls with loft, while on the other, he effectively avoided the worst kinds of contact (low exit velocity, especially with extreme launch angles). This suggests that his massive ISO drop off from 2017-18 to 2019 (.239 to .113) should be reversible.


Rowdy Tellez

Over the off-season, no player helped me learn more about effectively using Statcast data than Rowdy (though I should also give credit to Eno Sarris and Tom Tango for providing the lessons).


Rowdy (91 wRC+) was a guy I was initially down on. His 2019 walk rate (7.1%) and strikeout rate (28.4%) were each worse-than-average. I looked to Statcast for signs that he was short-changed on batted balls, but came up empty — his xISO (.225) was not meaningfully higher than his ISO (.221). While his xBABIP (.330) was meaningfully higher than his BABIP (.267), such gaps are common among slower runners who regularly face the shift.


Eno Sarris provided my first bit of Rowdy-hope by pointing out to me that a player’s barrel rate is a better predictor of their future ISO than is xISO. Players with a similar barrel rate to Rowdy’s 8.3% mark averaged an ISO in the .255-.260 range, notably better than Rowdy’s actual .221 ISO. Coincidentally, Rowdy posted a .255 ISO over 123 plate appearances following a late July/early August stint in Buffalo.


The table above, based on Tom Tango’s framework, provides even more bits of hope for a 2020 breakout. Rowdy generated above-average rates of hard-hit batted balls, as well as an average rate of weakly-hit/intermediate launch angle batted balls, but managed to produce a very low rate of weakly-hit batted balls with extreme launch angles.


It’s also important to reflect back on Rowdy’s production from August 14th on, after a stint in Triple-A (where he posted a 186 wRC+). Over 123 plate appearances, he slugged his way to a 111 wRC+ thanks to his ability to barrel the ball (8.9% barrel rate, 90th percentile over that stretch) and generate extra-base hits (.255 ISO, 75th percentile). One knock on this good run is that it happened in August and September, like 2018 (151 wRC+ over 73 PA), when the quality of competition sags due to 40-man call-ups. That said, of the eleven barrels, seven came against pitchers who had logged plenty of innings before September, while an eighth came against a Top 100 prospect (Bryse Wilson) who had made a few appearances throughout the season. A 6.5% barrel rate would rank in the 71st percentile over that month-and-a-half, suggesting that his results (in terms of ISO) were well-supported by the underlying contact.


Derek Fisher

Throughout his pro career, Fisher has consistently walked and struck out more than the average batter. This was certainly evident in 2019, when his cumulative walk and strikeout rate (46.7%) ranked ninth among 411 batters who made at least 150 MLB plate appearances.


One side effect of this is the relatively low percentage of his plate appearances that resulted in each of the four types of batted balls highlighted in the table above. As such, it’s useful in Fisher’s case to also examine his mix of batted ball types: for every Category 4-batted ball Fisher produced in 2019, he produced 0.24 Category 1-batted balls (53rd percentile), while for every weakly-hit ball he produced last season, he produced 0.81 hard-hit balls (82nd percentile).


The implication of these two paragraphs is that while Fisher’s batted ball quality is fine-to-good, he really needs to up the quantity of batted balls he produces, namely by reducing his strikeout rate. Admittedly, this is probably not a controversial or revolutionary thought, but it is nevertheless the task that is in front of him in 2020. For what it’s worth, he’s only struck out three times so far in Spring Training, good for a 17.6% K rate (I’ll have more to say about the usefulness of ST stats in my next post).


Brandon Drury

Among all of the Blue Jay hitters I examined for this post, none had more average numbers than Drury. In terms of the three most predictive categories (one and two are positive predictors, while four is a negative predictor), Drury finds himself between the 45th and 55th percentiles. In terms of Category 3, which Tom Tango says “tells you how average a batter is”, Drury stands out.


Combined with a poorer-than-average walk rate (5.6%) and strikeout rate (25.3%) in 2019, this analysis seems to corroborate how many Jays fans feel about Drury: a guy without a standout skill (and a few shortcomings), who likely won’t be a member of the next Jays playoff team.


Billy McKinney

The Blue Jays’ current front office is, in my opinion, a Top 10 group across the majors. I think that they have made more good moves than bad and even when a choice didn’t work out so well, I generally feel that the logic behind the move was sound.


The Happ for Drury and McKinney trade is one of the bad ones. Happ was never going to result in a true haul, but he seemed like a really good addition in a relatively weak 2018 trade market. Rather than get one or two higher-ceiling (and higher-risk) prospects for him, the Jays opted for higher floors. I can’t speak to the packages other teams were offering at the time, which is an important piece of missing information. However, there were clear worries at the time that they were both Quad-A types and their performances since have only reaffirmed those concerns.


Like Drury, McKinney posted a subpar walk rate (6.9%) and strikeout rate (26.4%). Unlike Drury, McKinney produced hard-hit balls at a below-average rate and weakly-hit balls with extreme angles at a higher-than-average rate. Unless he can show something different in Buffalo this season, where he’ll be competing for playing time with Jonathan Davis, Forrest Wall and Josh Palacios, I don’t see a future for him in the organization.


Concluding Thoughts

In 2020, the Blue Jays could put out a batting lineup made up exclusively of guys with solid batted ball profiles, in terms of pwOBAcon. Biggio, Teoscar, Rowdy and Shaw each produced above-average rates of hard-hit balls with loft and low rates of weakly-hit balls with extreme launch angles. Gurriel, Jansen and Grichuk each produced average rates of the worst kind of batted balls, but also produced very strong rates of hard-hit balls. Bo and Vlad could stand to generate more loft, but nevertheless produce a lot of hard contact and are still only in their early-twenties.


While each Blue Jay has things to work on in order to squeeze out every drop of potential, the foundation for a positive offensive season is there.




*Featured Image Courtesy Of DaveMe Images. Prints Available For Purchase.



Support Jays From The Couch On Patreon






Jeff Quattrociocchi

I'm an economics professor in the GTA whose lifelong love for the Jays was reignited by that magical August of 2015 and the amazing moments since.