In the composite rankings, Mile Split uses average time, 1-5 split, and points to rank teams. Which one is the most useful in predicting outcomes?
In the composite rankings, Mile Split uses average time, 1-5 split, and points to rank teams. Which one is the most useful in predicting outcomes?
Depends on the size of the race naturally. If your 5th guy is 2 minutes back, you might win smaller meets with less people because his impact is not as high compared to a larger meet. At something like NCAAs, you need a strong spread. The only exception I can think of is if you have an absolute monster at the front of the race. Your spread might be high because you have a superstar, but I would be willing to bet that good teams with a superstar probably have a phenomenal 2-5 spread instead.
Just my thoughts on the matter.
While I don't have the exact algorithm, basically you should rank every teams #1 runner and then factor the spread.