ConductVision Sports

Check the research behind every measure

This page lists the studies behind every measure on these basketball pages and the published accuracy of video measurement in sport.

Plan a pilotReading list
Gaps between better and weaker shooters against the error of markerless video, in degreesGap between better and weaker shootersMarkerless video error, root-mean-square with 95% CI0°5°10°15°20°KneeKnee: 18.9° gap between groups18.9°Knee: error 4.4° (95% CI 2.9° to 5.9°)4.4°HipHip: 12.8° gap between groups12.8°Hip: error 5.3° (95% CI 2.9° to 7.6°)5.3°AnkleAnkle: 12.2° gap between groups12.2°Ankle: error 4.9° (95% CI 3.9° to 5.9°)4.9°

Each gap between better and weaker shooters is two to four times the error of markerless video at that joint.

Accuracy

Published accuracy of video analysis

Markerless systems track joints from video without body markers. Lab motion capture, which tracks markers stuck to the body, is the standard that video measurements of joints are judged against.1,4 Each row reports a published result for video analysis, against lab motion capture, a force plate or timing gates.

MeasurePublished result
Knee and ankle angles, side viewRoot-mean-square error of 4.4 degrees at the knee and 4.9 degrees at the ankle, combining 20 studies of jumping.1
Hip angle, side viewRoot-mean-square error of 5.3 degrees, with wider variation between studies.1 Repeat measurements of hip bend varied by 4.0 to 11.1 degrees.2
Jump height, markerless2.9 cm lower than lab measurements on average, across the combined studies.1
Jump height, phone videoIntraclass correlation coefficient of 0.997 with a force plate and a mean difference of 1.1 cm, in 20 recreationally active men.8
Sprint split times, phone videoCorrelations of 0.989 to 0.999 with timing gates for split times over 40 meters, in 12 highly trained sprinters.9
Change-of-direction times, phone appCorrelation of 0.964 with timing gates, in 20 adolescent athletes.10
RepeatabilityMost repeat measurements varied by less than 5 degrees across 53 studies.2
Basketball jumpsJoint-angle differences under 8.2 degrees from lab motion capture.3 Knee and ankle angles were accurate.3 Hip angles showed a statistically significant difference from lab motion capture.3
Hip and knee joint centersConsistent offsets of about 30 to 50 mm from lab motion capture positions for three joint-tracking methods.4
Two athletes filmed togetherBody-segment angles agreed within 12.5 degrees, except the hand.5 Wrist angles were the only joint angles without acceptable agreement.5
Video at 25 frames per secondHow far each joint moved agreed closely (r = 0.916 to 0.994) in three athletes.6 Accelerations agreed poorly (r = 0.232 to 0.677).6
Event timingDetecting foot contact required video at 100 frames per second or more.7 An error of 20 milliseconds in timing foot contact changed the knee angle measured at that moment by up to 20 degrees.7
Ball flight from practice videoNot covered by the studies on this page. Each pilot checks ball measurements against sample frames measured by hand.

Using the numbers

How a pilot uses these findings

Profiles compare players measured with one system and one camera setup, because joint-tracking methods differ in where they place the hip and knee.4 Jump and sprint tests are filmed at 100 frames per second or more, the rate one study found necessary to detect when a foot touches down.7

Most of the cited studies tested college, professional or adult players. Two tested adolescent athletes, so a pilot compares each academy player with their own earlier sessions as well as with published results.10,19

Before any profile is shared, a ConductVision engineer measures sample frames from the academy's own video by hand, and the pilot reports how closely ConductVision agrees.

Reading list

Every study cited on the basketball pages

Accuracy of markerless video analysis

  1. Meta-analysis of markerless accuracy and reliability in jumping tasks.Ogura A, Florio E, Wileman TM, et al. (2026). Are we there yet? A systematic review and meta-analysis of the validity and reliability of automated markerless motion capture systems during jumping tasks. Journal of Sports Sciences.
  2. Review of markerless reliability and validity in sports and functional tasks.Yoma M, Llurda-Almuzara L, Herrington L, et al. (2025). Reliability and validity of lower extremity and trunk kinematics measured with markerless motion capture during sports-related and functional tasks: A systematic review. Journal of Sports Sciences.
  3. Markerless systems and lab motion capture compared in basketball jumps.Wei L, Li M, Yang C, et al. (2025). Comparison of lower limb kinematics and kinetics estimation of basketball players during jumping with markerless and marker-based motion capture systems. Acta of Bioengineering and Biomechanics.
  4. Joint-center accuracy of three pose-estimation methods.Needham L, Evans M, Cosker DP, et al. (2021). The accuracy of several pose estimation methods for 3D joint centre localisation. Scientific Reports.
  5. Markerless accuracy with two athletes filmed together.Oonk GA, Kempe M, Lemmink KAPM, et al. (2026). Examining the concurrent validity of markerless motion capture in dual-athlete team sports movements. Journal of Sports Sciences.
  6. Basketball movements captured at 25 frames per second.Li Z, Tan Z, Zheng W, et al. (2026). Evaluating a Multi-Camera Markerless System for Capturing Basketball-Specific Movements: An Exploration Using 25 Hz Video Streams. Sensors.
  7. Frame rates and event detection in two-dimensional sports video.Mundt M, Colyer S, Wade L, et al. (2024). Automating Video-Based Two-Dimensional Motion Analysis in Sport? Implications for Gait Event Detection, Pose Estimation, and Performance Parameter Analysis. Scandinavian Journal of Medicine & Science in Sports.

Jump and sprint timing from phone video

  1. Jump height from high-speed iPhone video compared with a force plate.Balsalobre-Fernández C, Glaister M, Lockey RA (2015). The validity and reliability of an iPhone app for measuring vertical jump performance. Journal of Sports Sciences.
  2. Sprint split times from high-speed iPhone video compared with timing gates and radar.Romero-Franco N, Jiménez-Reyes P, Castaño-Zambudio A, et al. (2017). Sprint performance and mechanical outputs computed with an iPhone app: Comparison with existing reference methods. European Journal of Sport Science.
  3. Change-of-direction times from an iPhone app compared with timing gates, in adolescent athletes.Balsalobre-Fernández C, Bishop C, Beltrán-Garrido JV, et al. (2019). The validity and reliability of a novel app for the measurement of change of direction performance. Journal of Sports Sciences.

Shooting mechanics

  1. Free-throw mechanics of proficient and weaker shooters, measured without markers.Cabarkapa D, Cabarkapa DV, Miller JD, et al. (2023). Biomechanical characteristics of proficient free-throw shooters—markerless motion capture analysis. Frontiers in Sports and Active Living.
  2. Three-point mechanics of proficient and weaker shooters, while loading the shot and at release.Cabarkapa D, Cabarkapa DV, Fry AC (2026). Biomechanical determinants of proficient 3-point shooters: markerless motion capture analysis. Frontiers in Sports and Active Living.
  3. Release-speed consistency and shooting accuracy.Slegers N, Lee D, Wong G (2021). The Relationship of Intra-Individual Release Variability with Distance and Shooting Performance in Basketball. Journal of Sports Science & Medicine.
  4. Release speed and coordination in missed and swished free throws.Mullineaux DR, Uhl TL (2010). Coordination-variability and kinematics of misses versus swishes of basketball free throws. Journal of Sports Sciences.
  5. Simulated release conditions for the free throw.Tran CM, Silverberg LM (2008). Optimal release conditions for the free throw in men's basketball. Journal of Sports Sciences.
  6. Body-pose attributes from broadcast video in made and missed NBA three-pointers.Panna F, Lucey P (2017). “Body Shots”: Analyzing Shooting Styles in the NBA using Body-Pose Attributes. MIT Sloan Sports Analytics Conference.

Fatigue and shooting

  1. Shooting-arm position under increasing fatigue in one NBA player.Erčulj F, Supej M (2009). Impact of Fatigue on the Position of the Release Arm and Shoulder Girdle over a Longer Shooting Distance for an Elite Basketball Player. Journal of Strength and Conditioning Research.
  2. Game-simulation fatigue and three-point shooting across positions.Bourdas DI, Travlos AK, Souglis A, et al. (2024). Basketball Fatigue Impact on Kinematic Parameters and 3-Point Shooting Accuracy: Insights across Players’ Positions and Cardiorespiratory Fitness Associations of High-Level Players. Sports.
  3. Repeated sprints and three-point kinematics in elite under-18 players.Slawinski J, Louis J, Poli J, et al. (2018). The Effects of Repeated Sprints on the Kinematics of 3-Point Shooting in Basketball. Journal of Human Kinetics.

Measuring and projecting shooting skill

  1. How many shots it takes to rank shooters, and evaluation from shot trajectories.Marty R (2018). High-resolution shot capture reveals systematic biases and an improved method for shooter evaluation. MIT Sloan Sports Analytics Conference.
  2. Stability of basketball metrics, including free-throw and three-point percentage.Franks AM, D’Amour A, Cervone D, et al. (2016). Meta-analytics: tools for understanding the statistical properties of sports metrics. Journal of Quantitative Analysis in Sports.
  3. Field goal percentage estimated from shot trajectories.Daly-Grafstein D, Bornn L (2019). Rao-Blackwellizing field goal percentage. Journal of Quantitative Analysis in Sports.
  4. NBA three-point percentage predicted from pre-draft data.Berger T, Daumann F (2021). Increasing the shot at a quality draft-decision – A Bayesian approach to improve predicting three-point accuracy translation in the NBA Draft. MIT Sloan Sports Analytics Conference.

Athletic testing

  1. Combine measures of drafted and undrafted players, 2000 to 2018.Cui Y, Liu F, Bao D, et al. (2019). Key Anthropometric and Physical Determinants for Different Playing Positions During National Basketball Association Draft Combine Test. Frontiers in Psychology.

Tracking and scouting from video

  1. Tracking data from college broadcast video and prediction of NBA talent.Patton A, Scott M, Walker N, et al. (2021). Predicting NBA Talent from Enormous Amounts of College Basketball Tracking Data. MIT Sloan Sports Analytics Conference.
  2. Player tracking from one moving broadcast-style camera.Johnson N (2020). Extracting Player Tracking Data from Video Using Non-Stationary Cameras and a Combination of Computer Vision Techniques. MIT Sloan Sports Analytics Conference.
  3. Defender distance and shot success in college tracking data.Tenan MS, Rezai AR (2023). Player Tracking Facilitates Valid Causal Inference: The Average Treatment Effect of Defender Proximity on Scoring. MIT Sloan Sports Analytics Conference.

Plan an academy pilot

Tell us about your players and the first question you want answered. A ConductVision engineer replies with a proposal.

Plan a pilotBasketball academies