Hard Contact Rate and Identifying Breakout Candidates

Sophomore year of high school, I was the statistician for the Junior Varsity baseball team. By that, I mean that I was not good enough to play and spent my bench time coming up with new ways to evaluate our players. But, JV baseball is a brave new world in terms of statistical analysis. Sample sizes are too small to properly determine much of anything, and fielding is so shoddy that offensive value is shockingly overestimated. So, I had to create an entire new suite of measurements.

I had a fair amount of data on contact quality, although it was subjectively assessed. But, I was able to cobble together some rate statistics to roughly determine hitting ability.

In doing research on MLB players, I thought that perhaps I could rely on my JV toolbox to identify top prospects. By simply multiplying “hard-hit rate” and “contact rate,” I am able to estimate the probability of a given swing resulting in hard contact. It neglects many factors, of course; for instance, contact in the zone may be more likely to result in a hard hit than contact elsewhere. But, this “hard contact rate” gives a reasonable approximation of the desired probability.

So, how does this statistic perform in evaluating players? Quite well, in fact. Looking at all qualified players in the 2015 season, there is a strong correlation between hard contact rate and wRC+.

So, hard contact rate is a fairly good predictor of overall offensive success. But, is it a repeatable skill? How consistent is it? To answer that question, let us look at the same qualified hitters in the two halves of the season.

View post on imgur.com

Once again, we see a relatively strong correlation. Although the sample size is not massive, it seems that hard contact rate stays more or less consistent. It is not subject to the constant fluctuations of something like batting average or BABIP. Thus, prospects with strong hard contact rates are likely to maintain that ability. As an indicator of offensive success, this statistic has proven quite strong.

In order to use hard contact rate to identify top prospects, we have to examine how it changes over time. Then, we can use the aging curve to spot those players who are performing better than their age mates. Here is that aging curve, drawn from all qualified hitters between 2011 and 2015.

Looking at players between the ages of 25 and 32, we see a clean curve predicting average hard contact rate over time. We must omit the players on either end of this 25-32 range, since that sample size is too small and characterized by exceptional players. There are not many league-average 21-year-olds, nor are there many under-performing 36-year-olds who still have a job.

But, we can still use the averages for those young players to identify truly exceptional talent. By filtering 2015 data to find players under the age of 23 whose hard contact rate is above average for a 23-year-old, we find the following list:

Harper, Machado, Sano, Correa, Schwarber, Bird, Conforto, Betts, and Odor.

Clearly, the system works to some degree.

I am particularly fond of the Odor pick. While he was a highly regarded prospect prior to his major-league debut, his freshman and sophomore seasons largely disappointed. However, I see a bit of Bryce Harper in him. Like his predecessor, true achievement is likely in his future; as the aging curve shows, hard contact rate peaks later in a player’s career than many other stats. Therefore, he is my pick for breakout candidate over the next few years.

By expanding this research, hard contact rates could be used to identify prospects and breakout candidates. I have yet to examine how the stat predicts success among minor leaguers, for instance.

In a future article, I will examine just that. Also in the pipe is an exploration of contact stats in predicting home runs. Whether or not hard contact rate holds up under further scrutiny remains to be seen.





5 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
evo34
8 years ago

Interesting. If hard contact rates peak at age 30, why do you suppose overall hitting value peaks much earlier? Is it all loss of footspeed?

Also, wouldn’t you want to check the correl. of first half Hard% to second half wRC+, and compare that to the correl. of first-half wRC+ to second-half wRC+?

Matt
8 years ago

Will have to find the stats and articles but I am fairly sure the 1st half vs 2nd half of Carlos Gonzalez last season was a great example. He was near the top of MLB in hard contact rate in his terrible start then we all know what he did after May.

Eric
8 years ago

I propose two new metrics to be added to this, UC and UH.
Meaning ‘ugly contact’ and ‘ugly hit’

Meaning ugly hit is like Alcides Escobar’s 4 base error of a home run in the bottom of the 1st in the 1st game of the world series, where he gets credit for a “legitimate” HR on the play, yet between Matt Harvery and Yoenis Cespedes there were three errors, two mental and one physical, accounting for all 4 bases, I don’t care how you divide it up.

Meaning ugly contact is a duck snort, dying quail single or a full swing ‘swinging bunt’ that wasn’t really a bunt etc.

Add those two to your stats above and I bet you will find, as UC and UH go up in volume, since its weak contact, inverse holds in terms of real batting average and slugging and OPS plus production, runs and RBI.

Alex
8 years ago

Deja vu! I have used the same process to track my youth team hitting. With fielding at this age so poor, hits are hard to determine, so like you I combine contact rate and hardhit balls. It correlates well with my homemade version of RC. Do you have any pitching ones from your JV days? I’d be curious to hear.

A