|Seth Walder @SethWalder
Potentially quite interesting development for college football analytics. I asked Telemetry founder Jeremy Hochstedler a few questions about it.
Telemetry Sports @TelemetrySports
We're proud to announce our partnership with @Sportlogiq that will change how NCAA football programs analyze the game.
Read about it here: https://telemetrysports.com/sportlogiq-partners-with-telemetry-sports-to-bring-nfl-level-analysis-to-ncaa-football-programs/
So the way this works is Sportlogiq uses computer vision off of all-22 footage to create player tracking data, which Telemetry then analyzes. And provides that information and analysis back to teams.
Unlike with the NFL where there are chips in the shoulder pads and the league owns all the data, with CFB it will be decentralized. Sportlogiq/Telemetry aim to reach individual agreements with teams, who will provide the all-22 footage.
Teams can have their own tape analyzed, but critically: also their opponent’s tape.
They anticipate a 24-36 hour turnaround time from when all-22 footage is uploaded. Some human intervention is required to identify players. Per Hochstedler, some jersey numbers can be detected autonomously but not all, particularly with low quality video.
X and Y tracked – Hochstedler said Z-axis is possible in the future but not something they are doing now.