Data Golf: Great site. Have you examined the correlation between players’ scoring averages over the season with their scores for a specific tournament? A positive correlation would show that the tournament sorted players according to ability. A zero correlation would show that the tournament scores are random with respect to ability. Some of my colleagues and I tried this for an entire season and found that some tournaments were better predictors of ability than others. But we only have the data for one year. If someone tracked the correlations for golf courses over the years, they would find which courses are the best tests of player ability. We called the tournaments with the highest correlation ( season scoring average/tournament score) the most valid test of player ability. Betters: Valid courses would have more predictable outcomes. Just a thought.
1 Like
Yes, we have looked into this in a few different ways over the years. We actually made a plot looking at this question in our post-PGA newsletter:
And then we also have an older blog post on this topic.
Our course fit tool also essentially gets at this question. Courses that favour most or all of the sub-skills will separate more on overall skill as well.
4 Likes