Leaderboard.bike pulls data from road-results.com so that I can implement a better ranking system than what USAC provides.

How are these rating points different from USAC ranking points?

• Leaderboard.bike uses an Elo rating system, which is statistically meaningful. USAC points are based on witchcraft as far as I can tell.
• Here, more is better. The more points, the better the rider.
• Points here never expire. If a rider takes some time off from racing, then comes back, their points are right where they left them. However, a rider will not appear in the leaderboard if they haven't raced within the last 12 months. USAC ranking points only considers the past 12 months. This might not sound like a big deal, but doing that actually wreaks havoc on the entire system.
• Leaderboard.bike differentiates between road, crit, masters', and women's categories separately.
• Rating points here are localized only to races in Washington State. Any rating system is only useful for making comparisons. Since most bike racers only race within their local association (e.g. the WSBA), it makes sense for a rating system to stay within that racing ecosystem. If a WA racer races in another region, those races have no effect on the racer's rating score here.
• Every race counts. USAC ranking points only considers your best five races in the last 12 months. But Leaderboard.bike considers every race. This means you can watch your score change after every race.
• Every rider counts. USAC ranking points only considers the top finishers in its algorithm for updating each riders' points. Leaderboard.bike updates each rider's rating based on every other rider that was in the race.

Why should anyone care about ranking points anyway?

Rating points are about more than just seeing who is the best. As we all know, bike racing is hard, and we can't all win every race. Having a reliable system for rating points provides each rider with something to measure their own performance with. If you already have goals that keep you going, then you probably won't care about these points. But rating points provides an objective way to measure how you're improving, even if the riders you are competing against change each week.

How does it work?

Your rating points can change with each race. Your rating points will change depending on how well you perform relative to the other athletes you were competing against. New athletes start with a number of points equal to the median number of points amongst the other athletes in their first race. As an athlete's points rise, their score will become less sensitive to change. This allows new riders to more quickly settle into their true score, and prevent experienced athletes from oscillating too much. The actual algorithm is a variation of the Elo rating system -- the same system that is used in chess, tennis, and many other sports. I've implemented it as suggested by this article.