Strava announced new features and updates to the app and platform today, including changes to make the leaderboard more fair and accurate. The company said earlier this year, 4.45 million activities were removed from the leaderboard, and the service has improved detection of e-bike rides that were accidentally or intentionally uploaded.
According to a press release, Strava has introduced new machine learning models to identify “irregular, unlikely, or impossible efforts,” and is beginning to use the models to remove problematic activities from the app's segment leaderboards. Google states, “Machine learning is a subset of artificial intelligence that automatically enables a machine or system to learn and improve from experience. Instead of explicit programming, machine learning uses algorithms to analyze large amounts of data, learn from insights, and make informed decisions.”
Leaderboards are used in the context of STRAVA to track the fastest time across a specified cycling and running segment, and many athletes will see how they can be used to compare with others. If the time posted appears to be suspicious, athletes can manually flag the activity for investigation.
With the latest changes, Strava reports that flagged activities related to the use of e-bike are down nearly 85% year-on-year, along with overall flagged activities. The company says U-Bike Rides will be automatically detected and deleted upon upload and moved to E-Bike Leaderboards separate from the leaderboard for Assisted efforts.
Strava did not share how the model works, as sharing this information would facilitate new ways for riders to fool the system. Athletes can manually flag any suspicious efforts that the model may have missed.
In many cases, vehicles are unintentionally misunderstood by Strava users, and this update promises to address these vehicles as well. Before riding on the platform, the athlete declares whether the activity has been completed on e-bike. If you're not a regular EMTB rider, you might forget to update this setting. Leaderboards can be affected when riders forget to stop GPS after riding, and drive homes are recorded as (very fast) rides.
Strava's Chief Product Officer Matt Salazar said the improved models should be a “fair platform where competition takes place.”
Additional product updates
In addition to improving leaderboard integrity, Strava has also announced a major update to its Mobile Route Builder tool. The update is said to include additional route details along with integration with the Strava Global Heatmap so that Riders can quickly identify popular routes.
New Power Skills and Training Zone features will also be released on the platform and should be available from September.
