This site is about the scientific approach to improving health and performance in sport. The increasing use of wearable sensors provides a growing source of data that is ripe for the application of machine learning algorithms and model-based statistical analysis. The aim is to provide new insights into the performance of individuals and teams.
Gavin Francis studied mathematics at St. John’s College, Cambridge, before pursuing a quantitative finance career in the investment management industry. Having published a paper on the application of neural networks for stock selection in the early 1990s, he has consistently worked on the use of algorithmic approaches to managing risk in the financial markets, particularly in the field of foreign exchange. Gavin has maintained a strong interest in Bayesian techniques, with has practical experience in managing portfolios for sophisticated global investors, using machine learning techniques.
Since 2016, Gavin has specialised in applying his expertise in data science and algorithmic techniques to analysing sports performance, particularly in the fields of cycling and deep learning. He has conducted research with his wife, Dr. Nicky Keay, on the health and performance of cyclists, leading to publications in academic journals. He contributes articles to the cycling press.
As a keen sportsman, Gavin has won national medals in Masters Cycling. He has represented Great Britain at age group level in World and European duathlon championships. He is an accomplished skier and snowboarder, while in the summer, you may find him kitesurfing, windsurfing or pursuing the family tradition of rock climbing.
Female football specific energy availability questionnaire and menstrual cycle hormone monitoring, Nicola Keay, Eddie Craghill, Gavin Francis, medRχiv, 2021
Clinical application of monitoring indicators of female dancer health, including application of artificial intelligence in female hormone networks , Nicola Keay, Martin Lanfear, Gavin Francis, medRχiv, 2021
Clinical application of interactive monitoring of indicators of health in professional dancers, Nicola Keay, Martin Lanfear, Gavin Francis, medRχiv, 2021
Indicators and correlates of low energy availability in male and female dancers Keay N, AusDancers Overseas, Francis G, BMJ Open Sport & Exercise Medicine, 2020
Strava inflation: Why getting a fast time always gets harder, Francis G, Cyclist, 2019
Clinical evaluation of education relating to nutrition and skeletal loading in competitive male road cyclists at risk of relative energy deficiency in sports (RED-S): 6-month randomised controlled trial Keay N, Francis G, Entwistle I, Hind K, BMJ Open Sport & Exercise Medicine, 2019
Infographic. Energy availability: Concept, control and consequences in relative energy deficiency in sport (RED-S) Keay N, Francis G, British Journal of Sports Medicine, 2019
Low energy availability assessed by a sport-specific questionnaire and clinical interview indicative of bone health, endocrine profile and cycling performance in competitive male cyclists Keay N, Francis G, Hind K, BMJ Open Sports and Exercise Medicine, 2018
Decomposition of Emerging Market Currency Risk: A Hedging Application, Francis G, Musli E, Cella T, Journal of Performance Measurement, Vol. 18 #1 – Fall 2013
Revealing the information content of investment decisions, with Noriyuki Okuyama, chapter in Handbook of Behavioral Finance, published by Edward Elgar, 2010
Dicing with the devil, Francis G, FX Trader Magazine, Oct 2010
Risk Assessment in Currency Programmes in “Foreign Exchange: A Practitioner’s Approach to the Market” RiskBooks, with Michael Shilling, 2008
Quantifying the Information Content of Investment Decisions in a Multiple Partial Moment Framework: Formal Definition and Applications of Generalized Conditional Risk Attribution, Okuyama N, Francis G, The Journal of Behavioral Finance, Vol. 8, No. 3, 121-137, 2007
Disentangling cognitive bias in the assessment of investment decisions: derivation of generalised conditional risk attribution, Okuyama N, Francis G, The Journal of Behavioral Finance vol. 7, No. 2, 75-87, 2006
Stock performance modeling using neural networks: A comparative study with regression models, Refenes A, Zapranis A, Francis G, Neural Networks, 1994
Trained to Forecast, RISK, Jan 1993