Sports Science: Week in Review, Oct 17-Oct 23

The first age of sports analysis is closing, the era that Bill James originated and which depended solely on numbers, statistical analysis and truth-telling. We’re entering the second era where the technology that enables data capture is as critical as the analysis.

Two categories of sensor technologies matter most. Wearable sensors attach to athletes or to objects, for what’s called the Internet of Things. Computer-vision technologies provide similar object tracking utility as wearables and can be easier for teams and athletes to use. Overall, the proximity of a wearable sensor to the thing it’s measuring make it a more precise, more accurate measurement.

Wearables like Lumo Run’s running form guage and GraphWear Technologies’ sweat patch are new products with features that are typical of what wearables do well.

These advances are just the beginning of what will be a long run for wearable technology innovation. Labs have recently reported breakthroughs like stick-on chemical sensors, safe brain implants for neural engineering and stretchable optical fibers that put light inside the body.

It’s encouraging to read how athlete performance professionals like Eamonn Flanagan, writing for PUSH, as well as the sports scientists at Sparta are able to grasp the benefits of athlete monitoring. But there’s lots of new technology coming on line and it’s not clear if Flanagan, Sparta and peers have the technical know how to scale up their current processes.

The camera-based computer vision systems are also coming on strong. Second Spectrum will take over the NBA’s player tracking with next-generation technology and improved data quality. The company’s founder, Rajiv Maheswaran, took questions on reddit.com/r/nba. ShotTracker, another computer vision basketball data capture tool, completed a $5 million venture round.

Computer vision research is also progressing rapidly. A steady stream of computing research awaits commercial technology transfer: SPiKeS superpixel-keypoints for tracking, action recognition methods and new outdoor multi-camera capture systems.

Sports teams vary in their adoption levels of new technology. The Dodgers are leaping forward, primarily with their own, best of its kind, startup company incubator. At the other end of the spectrum, Bill Belicheck has quit on the NFL’s tablet technology from Microsoft. In between you see Ohio State actively collaborating with military researchers on new sensor technology. The Boston Celtics are taking a wait and see approach with tablet technology. Adoption is spreading into high school sports, like football in Texas.

Sports media is an interesting position. As Ted Knutson notes in his critical assessment of commentators’ understanding of soccer analytics, there is unmet audience demand for in-depth, technical material. But then you see major outlets like ESPN give space to a technical understanding of sleep, travel and this season’s NBA schedule, but then say little that’s new or insightful.

More things that I read and liked last week:

Javier Baez and the Search for the Last Unicorn | VICE Sports (October 18, VICE Sports, Rian Watt)

[1610.07238] SPiKeS: Superpixel-Keypoints Structure for Robust Visual Tracking (October 23, arXiv, Computer Science > Computer Vision and Pattern Recognition; François-Xavier Derue, Guillaume-Alexandre Bilodeau, Robert Bergevin)

DARPA and Qualcomm brain implants for 6G and neural engineering (October 20, Quartz, Joon Ian Wong)

Predictive modelling of football injuries (October 23, Stelios Kampakis, The Data Scientist blog)

Gray Cook: “Movement Search: Connecting You to Your Movement Path” (October 21, YouTube, Talks at Google )

USI, Thunderbolts collaborate on sports science project (October 23, Evansville Courier & Press)

From Shared Dartmouth Floor, an Ascension for Alexi Pappas and Kyle Hendricks (October 22, The New York Times, Alexi Pappas)

Inside story of David Ortiz’s injury: ‘He was essentially playing on stumps’ (October 20, WEEI, Rob Bradford)

[1610.06906] Review of Action Recognition and Detection Methods (October 21, arXiv, Computer Science > Computer Vision and Pattern Recognition; Soo Min Kang, Richard P. Wildes)

[1610.06740] Model-based Outdoor Performance Capture (October 21, arXiv, Computer Science > Computer Vision and Pattern Recognition; Nadia Robertini et al.)

Despite new technology, Boston Celtics will lean on their iBrad (October 23, ESPN, Boston Celtics Blog)

Ohio State football players helping Air Force study performance | www.mydaytondailynews.com (October 23, Dayton Daily News)

I’m Rajiv Maheswaran, CEO of Second Spectrum. We build machines that understand sports and just partnered with the NBA to bring new tracking and analytics tech league-wide. AMA! : nba (October 22, reddit.com/r/nba)

Karl-Heinz Rummenigge: English clubs effectively ‘kidnapping’ young players (October 21, The Guardian, Football )

Mary Cain Leaves Nike Oregon Project – John Henwood, Not Alberto Salazar, In Charge of Her Training (October 22, LetsRun.com)

Football injuries and their prevention with Swedish football injury warriors (with images, tweets) · Nim_Perera · Storify (October 17, Storify, Nirmala Perera)

Lumo Run sensor: Tracks runner hip, pelvic motion (October 18, SI.com, Tom Taylor)

Reliability, validity and usefulness of 30-15 Intermittent Fitness Test in Female Soccer Players (October 17, Frontiers in Physiology)

Dodgers Lead Arms Race In Sports Tech (October 20, The Post Game, Jonathan Crowl)

How DeSoto, Coppell joined Cowboys, major college programs on cutting edge of football technology — and why (October 20, dallasnews.com, SportsDay)

“Our sport is your sport’s punishment”- A Brief Look at Motivation and Punishment (October 20, Steve Magness, Science of Running blog)

University of Maryland, College Park and University of Maryland, Baltimore team up to turn Cole Field House into national hub for sports medicine – Baltimore Sun (October 20, Baltimore Sun)