I'm a data nerd and wish there was a better way to rank individual runners throughout the country, even though most will never end up facing each other head-to-head.
The Elo rating system is a method for calculating the relative skill levels of players in competitor-versus-competitor games such as chess. Everyone starts with a score of 1,000 and your score is adjusted each game based on the probability of you winning. If you are the favorite and end up winning, you gain fewer points for beating a weaker opponent than if you upset a better opponent. You're also rewarded for quality over quantity, so 3 tough wins are graded more favorably than 10 easy wins. Long-term success is rewarded more than a team/player having a short peak.
A lot of sports use Elo ratings to rank teams and individual players, so I was wondering if it could be applied for track-and-field or cross-country. The problem is that games like chess or basketball are 1-on-1 competitions. It gets tricky to accurately model rankings once 3+ teams/players are involved. It's also easier to model a larger sample size of 82 games for a pro sport or hundreds of chess matches because they happen so quickly. Cross-country teams have maybe 10 meets per year, and the best runners might only race half of them.
Cross-country races sometimes have hundreds of runners at a time. A hilly course at altitude on a cold day in CO will have slower times than a flat course in perfect weather in CA, so comparing times at different venues is impossible if the athletes never face each other head-to-head. There are so many variables that make quantifying relative performance almost impossible.
Track is a little easier because you'll rarely have more than 8 people in a sprint heat or more than 24 people in a distance heat. However, you have to consider the fact that there are heats. You see it all the time: 10 guys miss the finals despite running a faster time than the 2 qualifiers from the slowest heat because that's the way track works. But is the slow heat winner actually a better runner for winning a slow race than someone who ran faster than him? After all, the goal is to get from start to finish fastest, and they were only the fastest in their slower heat. At the end of the day, you can pretty easily compare times on a track since courses
Are there any Elo rating systems for cross-country? I know Speed Ratings exist for HS cross-country, but I honestly don't know the math behind them. I'd love to dive more into the topic and have a decent statistics background, but I'm looking for where to start if I was to develop a new ranking system. If you guys have any input on this topic that'd be awesome.
I've looked up a few articles regarding this topic in other sports/games and here's what I found.
http://www.gautamnarula.com/rating/
http://runningmagazine.ca/chess-and-running-what-counts-as-elite/