This is a review of the Bayesian hierarchical latent variable models conducted by Kyle Burris and Greg Appelbaum.
Many experts consider hitting a pitched baseball to be the hardest thing to do in sport. The athlete must determine the type of pitch and its trajectory, decide to swing the bat in coordination with the path of the ball – all done within milliseconds.
In this study, Sensory Station assessments from 252 professional baseball players were compared to game statistics to study the relationship between sensorimotor skills and baseball production. The assessments consisted of nine computerized sensorimotor tasks, each designed to evaluate a specific aspect of the player’s visual-motor abilities. The nine tasks measured the player’s ability to: make out small details on distant objects, detect the contrast between an object and its background, quickly and accurately measure the relative distance of an object, shift their attention and recognize peripheral targets, determine the number of targets both near and far way, remember and recreate visual patterns, respond to rapidly changing targets, and respond to “go” or “stop” stimuli, and finally the ability to quickly respond to simple visual stimulus.
Baseball production is measured using five different statistics. On-base percentage (OBP) measures how frequently a batter reaches base per plate appearance. Times on base includes hits, walks and hit-by-pitches but do not include the times a player reaches base due to defensive errors. Players need to be able to consistently hit the ball or draw a walk in order to have a high OBP. Walk percentage (BB%) is the percentage of a batter’s plate appearances that result in the player being walked. Players with a high walk percentage are able to accurately determine whether a pitch is a ball or a strike. Strikeout percentage (K%) measures the number of times a batter strikes out. Slugging percentage (SLG) measures the power of the hitter…recognizing the fact that not all hits are equally valuable.
Analyzing the data showed that sensorimotor abilities are related to on-base, walk and strikeout percentages, but not to slugging percentage. These relationships are self evident. Players with high on-base and walk percentages combined with low strikeout percentages must be able to determine the location of the pitch in the strike zone and decide whether to swing or not. Slugging percentage would deal more with strength, rather than sensorimotor abilities. Players with a high walk percentage typically scored high on tests relating to hand and eye coordination as well as response times. Players who have a low strikeout percentage tend to score high on the tests relating to measuring the relative distance of an object, recognizing peripheral targets, and responding to rapidly changing targets.
These relationships would indicate that sensorimotor screenings would be highly beneficial for player scouting. Screenings would help differentiate those players who are more likely to excel in the professional arena.
How mature is your team’s analytics program? Take the Sports Analytics Maturity Assessment.
Learn about the Groundbreaking Sports Analytics Model coaches and sports analysts are talking about!
Learn all about Sports Analytics here.