search

jeanlucj / FLRCpointFinder

Analyze data from the FLRC challenge to see what courses to focus on next
0 stars 0 forks source link

Add trail versus road and "degree of difficulty" as covariates #2

Open jeanlucj opened 3 years ago

jeanlucj commented 3 years ago

The really interesting point will be to have the interaction of these covariates with Athlete be a random effect. Do those interactions explain anything? That is, do some athletes do better on road than trail or vice versa. Do some athletes do better in short races than long races?

The degree of difficulty would be calculated from the mean time for a course, presumably after taking the worst time off of each Athlete for each course

jeanlucj commented 3 years ago

The current .Rmd has the relevant models. I would like to apply them to the "best N" results for each athlete on each course. Would like to figure out where N is kept. It would also be good to figure out how to present the results somewhat intuitively. Relative bar between Speed --------|------------- Endurance (slightly more of a speed person) Trail ------------------|--- Road (definitely more of a road person) TBH, those text bars look fine to me... Also, just rank the courses in terms of where the athlete is most likely to make the biggest improvements.