Go! Multi Sport Start Analytics App

In competitive sports, victory is determined by milliseconds; at the 2025 World Championships, a mere 0.14-second gap separated first from fourth place. Rapid reactions enable the nervous system to process stimuli and trigger muscular commands, providing athletes with a vital edge. Reaction time measures the interval between a stimulus and the initial response; motion time refers to the physical movement during the start period. Although genetics and fatigue influence speed, systematic analysis and training build the muscle memory required for improvement. Available Apps for reaction time improvement include games, tapping movable lights on phone screens, and measuring reaction time with phone motion sensors, but this can be inaccurate.

Motivation

As a competitive speedskater, I often stood on the podium, wondering why first place remained out of reach. I eventually realized the differentiator wasn't just talent, but access to elite coaching. While Canada’s first-rated sport hubs are centered in major cities, athletes from rural areas often struggle to access high-level training. While others had expert guidance, I was forced to become my own coach—independently guessing how I was skating—and it inspired me to build an app to bridge this gap.

Goal

The purpose of this project was to create a new app that accurately measures motion and reaction time, helping athletes train and monitor their starts for improvement by visual analysis.

Application

This App will empower athletes worldwide across multiple sports to improve their reaction and motion times and give them an advantage to reach their potential.

Research

The first step was to research existing apps on Android and Apple platforms. I compared good features and shortcomings, such as a lack of reaction and motion time recording. I also researched the best available software programs for coding, as well as artificial intelligence (AI) tools to assist me. I conducted a literature review using published articles in well-known journals to improve my understanding of reaction and motion times.

Procedures

I coded an Android app using Android Studio Software on my computer. To make the coding process more efficient, I use Google Gemini, which helps generate precise programming prompts for Cursor. Each instruction was entered one at a time, building the app’s functionality. After compiling the app in Android Studio, I installed it on my phone and tested it, improving its functionality by refining the prompts and making small but important changes.

Experiment and data collection

Once the app was ready, testing began by using a phone with the GO! Multisport App, tripod and orange cones. After informed consent was given, four athletes, each specializing in long-track speed skating and long-distance running, participated in the study over five days. Each day consisted of a ten-minute warm-up, followed by five starts without using the App's analyzing function, only recording the times. This was followed by five starts using the App to track motion, reaction time, and video analysis.

Each start required a 10m sprint to the orange cones, followed by a 5-minute break. The measured reaction and motion times are automatically recorded in a data table and graph. The video is also automatically stored, and all data can be retrieved and analyzed immediately.

Variables

I controlled the independent variable by testing the same athletes across both speed skating and running to measure the dependent variables: reaction and motion time.

Statistics

To ensure correct statistical results, I performed an 80% power analysis with α=0.05. I determined that four participants would be sufficient. The results are displayed in box-and-whisker graphs. The graphs show minimum and maximum reaction and motion times in long-distance running and long-track speed skating. The line indicates the median, while the cross marks the average.

Results and analysis

In this project, I coded a new Android App named GO! Multisport Start Analytics App. I used AI-generated prompts, the Android Studio Platform, and Cursor software. The app visually tracks progress during motion-capture starts. It converts this to high-speed video analysis and data on reaction and motion times. Reaction time is measured from the “gun” sound of the voice command to movement at the ankle. Motion time is measured from the ankle movement until the ankle crosses the line. The app takes frame-by-frame shots. Artificial Intelligence analyzes shots to identify movements and thus determine reaction and motion times. The app turns these fast movements into measurable data. It summarizes the data in tables and graphs on the dashboard for historical analysis. This helps the athlete build a track record and monitor improvement trends. The video analysis lets athletes watch their movements in slow motion to refine start techniques and correct posture and body position.

After 5 days of training starts with and without using the GO! App, I gathered the following results: for long-track speed skating, the average reaction time with the app is 396.6 ms (milliseconds), compared to 427.44 ms without using the app. The average motion time using the app was 3.28 sec (seconds), compared to 3.47 sec without using the GO! app. This is a 30 ms or 8% improvement in reaction time and 0.19 sec or 6% improvement in motion time in long-track speed skating starts, achieved just 5 days after practicing with the app. In long-distance running, the recorded average reaction time was 409.6 ms with the app and 432.0 ms without the app, resulting in a 22.4 ms decrease, or a 5.5% improvement in reaction time. The average motion time for long-track running was 3.12 sec with the app, while 3.34 sec was measured without it. For long-distance running, the motion time improvement was 7% or 0.22 sec.

Developing and testing the GO! Multisport Start Analytics App showed that adding visual feedback to training leads to quick, measurable improvements. I used Android Studio, Cursor, and AI-generated prompts to build a platform that turns fast, complex movements into useful information. Through this project, I learned that when athletes can both feel and see their movements with frame-by-frame video analysis, they build the muscle memory needed for powerful starts more effectively.

The results show the app makes a real difference. In just five days, long-track speed skaters improved their reaction and motion times by 8% and 6%. Long-distance runners improved by 5.5% and 7%. These quick gains suggest even more progress is possible with longer use. The app stands out because it lets athletes track their performance over time, use image annotation to fix body position, and set clear targets for reaction and motion times.

These results show that the benefits of elite coaching can be turned into digital tools. By tracking ankle movement from the starting signal to the finish line, the app provides athletes with the accuracy they need for top-level competition. I found that tracking progress this way not only records improvement but also speeds up development. The GO! App helps athletes improve their technique on their own, showing that even in sports where milliseconds matter, accessible technology can make a big difference.

Moving forward, I want to expand the App to include more sports, such as swimming, cycling, and rowing, making it more versatile. I also plan to add multilingual starter commands, which will be essential if the app is used globally. Another goal is to evolve the app into a camera and computer system to improve accuracy and open new possibilities. Ultimately, I aim to get the app onto the Play Store and see it used worldwide by athletes and training centers.

I would like to sincerely thank William Thompson for supporting me in pursuing an innovative idea. Thanks, Mom, for attending all of my fairs, staying up with me while I practiced my presentations, and for her patience throughout the process.

My thanks as well to Ms. Jewell of Marymount Academy for her guidance and encouragement. Finally, I would like to thank Paul Woodall, the official starter for Speedskating Ontario, for providing the voice recordings of long- and short-track speedskating used in the app.

References

Google. (2026). Android Studio (2026.1.2) [Computer software]. https://developer.android.com/studio

Backyard Brains. (n.d.). The science of your reaction time. Retrieved from backyardbrains.com. https://backyardbrains.com/pages/the-science-of-your-reaction-time?

CAS (2024). Sports technology is enhancing athletic performance. Retrieved from cas.org. https://www.cas.org/resources/cas-insights/latest-sports-tech-boosting-performance

Centre de glaces Intact Assurance - Skating | Activities and Attractions | Visit Québec City. https://www.quebec-cite.com/en/businesses/centre-de-glaces-intact-assurance

Cursor. (2026). Cursor (Version 0.x) [Large language model]. https://cursor.com

Davis, T. L., & Fang, J. Y. (2010). Movement Time. In K. Kompoliti & L. V. Metman (Eds.), Encyclopedia of Movement Disorders (pp. 219-220). Academic Press. doi.org. https://catalog.nlm.nih.gov/discovery/fulldisplay?

Derderian, C., Shumway, K. R., & Tadi, P. (2023). Physiology, Withdrawal Response. In StatPearls [Internet]. StatPearls Publishing. Available from: nih.gov. https://www.ncbi.nlm.nih.gov/books/NBK544292/

Frontiers for Young Minds. (2022). Watch and Learn: Athletes Can Improve by Observing the Actions of Others. Retrieved from frontiersin.org. https://kids.frontiersin.org/articles/10.3389/frym.2022.70278

Google. (2026). Gemini (January.2026) [Large language model]. https://gemini.google.com/

Grammarly. (2026). Grammarly [Computer software]. https://www.grammarly.com

Jain, A., Bansal, R., Kumar, A., & Singh, K. D. (2015). A comparative study of visual and auditory reaction times on the basis of gender and physical activity levels of medical first year students. International journal of applied & basic medical research, 5(2), 124–127. https://pmc.ncbi.nlm.nih.gov/articles/PMC4456887.

Liebermann, D. G., Katz, L., Hughes, M. D., Bartlett, R. M., McClements, J., & Franks, I. M. (2002). Advances in the application of information technology to sport performance. Journal of sports sciences, 20(10), 755–769. https://pubmed.ncbi.nlm.nih.gov/12363293.

Mirshams Shahshahani, P., Lipps, D. B., Galecki, A. T., & Ashton-Miller, J. A. (2018). On the apparent decrease in Olympic sprinter reaction times. PloS one, 13(6), e0198633. https://pmc.ncbi.nlm.nih.gov/articles/PMC6021049.

Noah Lyles wins a historically close Olympic 100-meter sprint by five-thousandths of a second - Bloomberg.https://www.bloomberg.com/en/news/thp/2024-08-04/noah-lyles-finishes-2nd-in-olympic-100-meter-semifinal-heat-but-advances-to-race-for-gold

BioLogically. (2021, May 7). Reflexes v. reactions [Image]. BioLogically. biologicallyblog.com

Swivel Vision, LLC. (n.d.). How to Improve Reaction Time: The Ultimate Guide. Retrieved from swivelvision.com

World Athletics. (2025). FINAL | 100 Metres | Results | Tokyo 25 | World Athletics Championship. Retrieved from worldathletics.org. https://worldathletics.org/competitions/world-athletics-championships/world-athletics-championships-tokyo-2025-7190593/results/men/100-metres/final/result