Adjusting Hoskins’ Batted Balls

Every year we have a number of players who make their debut towards the end of the season, wildly exceed expectations, and leave us wondering what the future may hold. Last year we had Gary Sanchez. This year, Rhys Hoskins.

Hoskins hit the ground running. I mean, how many guys reach double digit homers before they reach double digit singles? I could probably look it up, I’m not going to. I don’t want to know. Hoskins did it, and that’s good enough for me.

I mean, that’s totally crazy. Just think about it. There were 486 batters with at least 100 PA last year, and during a 14 day stretch Hoskins hit more homers than 259 of those guys hit, total. There were 32 players with 400 PA who had fewer homers than Hoskins hit during those 14 days.

Eighteen of his forty four major league hits are home runs. That probably isn’t going to continue, but how much will he be reined-in going forward? Clearly his batted ball distribution is skewed toward the high end, considering he hit a ridiculous number of home runs. However, we might be able to judge him based upon the balls he didn’t hit.

I’ve created two charts below that show the frequency of his exit velocities and launch angles compared to the major league averages. Let’s look at them one at a time and see if we can find something useful.

First, exit velocity. Notice how he has a large gap between around 88 and 98 miles per hour. The major league sample has a gap in a similar area, in large part due to a limitation of Trackman. However, Hoskins’ gap is unrelated to that issue. Hoskins is missing balls due to the funkiness of a small sample size. Similarly, you see a gap between 47 and 60 miles per hour.

Generally speaking, though, his high exit velocity balls are well populated, as you’d expect from him hitting 18 home runs in a span of 127 batted balls. We could assume that these high velocity balls define the upper limit of his exit velocity range, the area right before the distribution rapidly drops towards zero. His peak appears to be around 103 mph, while the MLB peaks around 100 mph. From this, we might assume he could have an exit velocity 3 mph above average.

The launch angle chart is more difficult to read. There are two large peaks on 14 and 28 degrees. The 25-33 degree area is well populated, as you’d expect from his home run rate. You’d expect that to regress over time. Furthermore, when I look for the highest density of high velocity batted balls, I find that 11 of his 58 balls hit above 95mph were hit between 14 and 17 degrees.

Back to the general shape of his batted balls, there are pretty large gaps in his balls hit between 0 and 10 degrees. And, really, all of his ground balls seem to be suppressed.

Combining this exit velocity and launch angle data, I see a player with a launch angle that is 3 mph above average and a launch angle that is 2.5 degrees above average. He has a small sample of batted balls, which is skewed toward the high end, with large gaps in the 88 to 98 mph range, 43-60 mph range, and to a lesser extent the 60 to 70 mph range. He also has gaps in the 1 to 10 degree launch angle range, sub 0 degree range, and the 20-26 degree range.

To a large degree most of this is common sense. He hit a ton of flyballs, so you’d expect him to hit more grounders going forward. However, it is valuable to note how many high exit velocity ground balls he appears to be missing from this sample. So even though you’d expect him to hit more softly hit balls in general, those soft hits will likely have decent batted average.

Balls in that 1 to 10 degree launch angle zone have a batting average of .476 and a .522 slugging. Those are valuable batted balls, and he is missing a large chunk of those. After accounting for Hoskins’ higher average launch angle, MLB hit 15% of their batted balls in what is equivalent to Hoskins’ 1-10 degree range. Hoskins has hit 9% of his balls in this range. So you’d expect him to hit roughly twice as many of these decently valuable batted balls over a sample of this size, all things being equal.

I have gone through and calculated the number of balls Hoskins hit into each launch angle bucket, and compared it to how many balls MLB have hit into that bucket. I have shifted MLB’s distribution of hits 3 degrees to account for Hoskins’ above average launch angle. Hoskins’ is 2.5 degrees above average, but my data is split into full degree segments so I had to round up to a 3 degree shift.

Hoskins vs MLB
Launch Angle Hoskins MLB AVG SLG
< 0 24.3% 29.0% .194 .209
0-10 9.3% 15.4% .476 .522
11-19 17.8% 14.7% .707 .964
20-26 8.4% 10.9% .540 1.252
27-39 22.4% 15.8% .313 .947
> 40 17.8% 14.2% .044 .089
MLB Numbers are shifted 3 degrees to account for Hoskins’ higher launch angle

Over 127 balls in play, you’d have expected Hoskins to hit 5 more balls beneath 0 degrees, 6.5 more between 1 and 10 degrees, 3.3 fewer between 11 and 19 degrees, 2.7 more between 20 and 26 degrees, 7.1 fewer between 27 and 39 degrees, and 3.8 fewer above 40 degrees.

Just using the league average success rates for these six categories, multiplying out Hoskins’ BIP in each category by the league stat in that category and dividing by the total BIP, the following chart shows the values before and after making this adjustment to his frequency distribution.

Hoskins
Stat Actual Values Redistributed BIP Frequency
BABIP .341 .348
SLGcon .604 .581

SlugCon, where you can celebrate your favorite mollusk with like minded individuals.

His BACON, and thus his batting average, went up by 7 points, but his slugging went down by 23 points. Rhys Hoskins’ xStats were: .260/.399/.600 with a .350 BACON. So, applying this sort of adjustment, his expected batting average might climb to .268 and his slugging drop to .577.

That is only after applying the launch angle adjustment. Following all of the above steps, except applied to exit velocity, the numbers adjust to .265 batting average and .579 slugging.

These adjustments only apply to his contact numbers, though. We are going to have a change to his walk and strikeout rate as well. The xStats projections 21.4 K% and 12.7 BB%, down from his 21.7 K% and 17.5 BB% in 2017. Applying these changes, plus the relevant regression toward the mean, I get a .262/.359/.507 projection for Rhys Hoskins. Which is a bit weaker than the Steamer projection. .261/.354/.518

Enough about that, though. Earlier I stated he appears to have a batting average 3mph above average and a launch angle 2.5 degrees above average. That would give him an average exit velocity of 89.5 mph and an average launch angle of 13.6 degrees. Similar batters are:

Similar Batters
Name EV V Angle
J. D. Martinez 90.6 15.3
Justin Smoak 89.5 17.1
Wil Myers 89.6 15.3
Josh Donaldson 90.6 13.6
Bryce Harper 90.1 13.7
Aaron Judge 94.5 15.5
Khris Davis 92.1 14.2
SOURCE: xStats.org

A solid list of names. Aaron Judge and Khris Davis represent exit velocity upside, and Josh Donaldson and Bryce Harper show launch angle downside. Wil Myers is probably the weakest guy here, and while Smoak had a great 2017 he’s a scary guy to bet on. Everyone else, though, has done very well. J.D Martinez would probably be an MVP candidate had he played a full season with the DBacks. Donaldson and Harper are both former MVPs, Judge is an MVP finalist, and Khris Davis is one of the most oddly consistent players in baseball.

This is good company for Hoskins to aspire to join.



Print This Post

Andrew Perpetua is the creator of CitiFieldHR.com and xStats.org, and plays around with Statcast data for fun. Follow him on Twitter @AndrewPerpetua.

newest oldest most voted
czelasko
Member
czelasko

With his exit velo and launch angles, I could see a 280/370/570 upside while flirting with 40HR