When a sports performance is at its peak, it is akin to a musical performance in the sense that each player seems to perform their part effortlessly, creating a rhythmic flow of counterparts all moving as one. Rhythm and timing are vital elements in sports like basketball in which syncopated passing and shooting appear to facilitate accuracy. This study tests if shooting baskets “in rhythm,” as measured by the catch-to-release time, reliably enhances shooting accuracy. It then tests if an “in rhythm” timing is commonly detected and agreed upon by observers, and if observer timing ratings are related to shooting accuracy. Experiment 1 tests the shooting accuracy of two amateur basketball players after different delays between catching a pass and shooting the ball. Shots were taken from the three-point line (180 shots). All shots were recorded and analyzed for accuracy as a function of delay time, and the recordings were used to select stimuli varying in timing intervals for observers to view in Experiment 2. In Experiment 2, 24 observers each reviewed 17 video clips of the shots to test visual judgment of shooting-in-rhythm. The delay times ranged from 0.3 to 3.2 seconds, with a goal of having some of the shots taken too fast, some close to in rhythm, and some too slow. Observers rated if each shot occurs too fast, in rhythm slightly fast, in rhythm slightly slow, or too slow. In Experiment 1, shooters exhibited a significant cubic fit with better shooting performance in the middle of the timing distribution (1.2 sec optimal delay) between catching a pass and shooting. In Experiment, 2 observers reliably judged shots to be in rhythm centered at 1.1 ± 0.2 seconds, which matched the delay that leads to optimal performance for the shooters found in Experiment 1. The pattern of findings confirms and validates that there is a common “in rhythm” catch-to-shoot delay time of a little over 1 second that both optimizes shooter accuracy and is reliably recognized by observers.

Machine learning is a rapidly growing field, with no doubt in part due to its countless applications to other fields, including pedagogy and the creation of computer-aided tutoring systems. To extend the functionality of FACT, an automated teaching assistant, we want to predict, using metadata produced by student activity, whether a student is capable of fixing their own mistakes. Logs were collected from previous FACT trials with middle school math teachers and students. The data was converted to time series sequences for deep learning, and ordinary features were extracted for statistical machine learning. Ultimately, deep learning models attained an accuracy of 60%, while tree-based methods attained an accuracy of 65%, showing that some correlation, although small, exists between how a student fixes their mistakes and whether their correction is correct.