About This Site
Built to track autocross history, results, classes, and rankings in one place.
AutocrossRank.com brings together event results, driver history, class and PAX data, uploads, claims, and public rankings so racers and fans can follow performance across seasons.
What AutocrossRank Does
The site is designed to make autocross data easier to explore and easier to trust. Visitors can browse events, drill into class results, look up drivers, review class and PAX information, and see public driver rankings built from imported event data.
For signed-in users, the site also supports uploads, result claiming, and account-level history. For admins, it adds review tools, parsing workflows, class management, and ranking controls.
Quick Links
About Chris Hammond
AutocrossRank.com is the work of Chris Hammond, an autocrosser from St. Louis who has been part of the sport for more than 25 years.
Chris first launched Christoc.com's Racing Forums, which later evolved into Pylon.cc and ultimately SOLO2.ORG. This site continues that long-running effort to give the autocross community useful tools, shared history, and a place to follow the sport online.
Pick The Winners
AutocrossRank.com is also the home of the annual Pick The Winners contest. That tradition is coming back again for the 2026 season.
Learn More About Pick The WinnersWhat's Public Today
Public visitors can already browse events, drivers, classes, and rankings without needing to log in. As more data is reviewed and imported, the historical picture should keep getting better.
How The Current Ranking Algorithm Works
AutocrossRank uses the Glicko-2rating system — the same family of algorithms used in competitive chess and many online rating systems. Every driver starts at a rating of 1500, and that rating moves up or down based on head-to-head class results at each event. The system also tracks how confident it is in each driver's rating and how consistent their results have been over time.
1. Each driver carries three numbers: Rating, Rating Deviation, and Volatility.
Rating is the skill estimate (starts at 1500). Rating Deviation (RD) measures how confident the system is — a new driver has high uncertainty (~350) that shrinks as results confirm the estimate. Volatility tracks how much a driver’s performance is expected to fluctuate.
2. Rankings are computed inside each class, not across all cars at once.
For each event, the algorithm groups drivers by class and only compares drivers who actually ran against one another in that class.
3. Beating a higher-rated driver is worth more than beating a lower-rated one.
Every head-to-head pairing within a class counts as a separate outcome. If your time beats a driver whose rating is above yours, the system rewards you more than if you beat someone rated lower.
4. Losing to a weaker-rated driver hurts more than losing to a stronger one.
The same logic applies in reverse. The expected outcome of each matchup is calculated from the ratings involved, and your rating shift reflects how surprising the actual result was.
5. The system tracks certainty, not just score.
Drivers who haven’t competed recently see their RD widen — the system becomes less certain of their current level until they race again. High-RD drivers can move quickly up or down when new results arrive.
6. Bad or incomplete results are excluded.
A result is excluded from ranking comparisons if it has no usable event time, or if it appears to be an incomplete multi-course result where only part of the required day was actually completed.
7. Final order is by rating first, then wins and event count.
After all events in the chosen scope are processed, drivers are sorted by rating. Wins and total ranked events are then used as tie-breakers.
In plain English: the algorithm tries to reward drivers for consistently beating strong competitors in their class, adjusts for the expected outcome of each matchup, and ignores result rows that look incomplete or unreliable.
