■ CRITICAL RISK ■ Sports & Entertainment
The recording and tabulating version, yes — optical tracking and computer vision log every pass and pitch automatically now. The analysts who turn that flood into decisions, and the official scorers who make judgment calls, are a much smaller and safer group.
“AI crunches stats in real-time. Your abacus is showing.”
Our AI replacement risk score — how we score jobs
The traditional sports statistician sits courtside or in the press box logging events: touches, rebounds, pitch types, completions, keeping the official book, feeding broadcast graphics, and compiling postgame packs. League stat crews, university sports information departments, and data companies employing armies of in-venue loggers built the numbers economy of sports. The skill was fast, accurate observation under a scoring rulebook — recognizing in real time that that was an assist, not a turnover.
Computer vision made the observation automatic. Optical tracking systems installed in major-league venues capture player and ball positions many times per second, generating datasets no human crew could — expected goals, pitch spin, defensive positioning — and AI event-detection turns raw video into tagged play-by-play even without installed hardware, which is pushing automated stats down into lower leagues and college sports that never had tracking budgets. Broadcast integration is automated; betting-data pipelines, the industry's money engine, demand machine-speed latency no scorer with a laptop can match. The manual logging layer of this occupation is being deleted venue by venue.
What survives splits in two. Down-market, humans still log games where camera coverage is absent or economics don't justify it — high school, small college, minor leagues — though phone-camera AI is coming for that too. Up-market, the growth is analysts: people who interrogate tracking data for tactical edges, build models for front offices, and translate numbers for coaches who don't want a dashboard, plus official scorers making rulebook judgment calls that leagues keep human on purpose. Our 79 reflects the honest asymmetry: the many logging jobs automate; the few analyst jobs thrive and multiply, but they're different jobs requiring different skills.
Automatability: our editorial assessment of current and near-term AI capability
Major leagues already run on automated optical tracking, and AI video analysis is now spreading the automation to venues that never had hardware budgets — the in-venue logging job is contracting right now and through this decade. Analyst roles interpreting the resulting data flood keep growing over the same period, but they demand data-science skills, not scorekeeping speed. Human loggers persist longest at amateur levels, on borrowed time.
Fewer every season for logging — optical tracking and AI video analysis record events automatically in major venues, and the technology is spreading downward. What teams increasingly hire are analysts who work with the tracking data, and leagues still employ official scorers for judgment calls. The clipboard jobs are converting to laptop jobs.
Yes — they're opposite ends of the same disruption. Automated tracking created far more data than anyone can interpret, so analyst demand keeps growing in front offices, betting firms, and media. The catch: those roles want data-science skills. Scorekeeping experience helps with domain intuition but won't substitute for Python and modeling.
For capture, yes, overwhelmingly — dozens of positional readings per player per second, spin rates, coverage metrics no human could log. For interpretation, it's murkier: ambiguous scoring judgments, context, and narrative still involve humans. That's why the official scorer survives while the stat-crew logger doesn't.
Add a programming language and get hands-on with tracking data — many providers offer public samples. Position yourself as a data-quality and interpretation specialist rather than a recorder. And use your access: relationships with teams, media, and leagues are how analyst openings get filled, and you already have them.