Quote:
Originally Posted by Chris Hibner
You also don't have a way to determine which robots scored the points in each match, which is why we use OPR. You can use the OPR algorithm on coopertition points to get a "coopertition OPR" that should in theory determine which robots are doing the coopertition bridge balancing.
|
The problem is that for every match, I at least get a (foul + baskets + bridge) score for each alliance, which is then used to determine OPR. I don't, however, get a coopertition score for each alliance every match, so you can't apply the same regression math to try and tease out which teams actually earned the coopertition bridge points. With the data actually available in the FRC results, the best you can do is divide each team's CP total by their # of matches played, and hope that correlates with the teams that are actually cooperteting.
If I could parse the twitter feed or if it was available at a more convenient place than twitter (anyone know a link?), then I'd be able to break out fun things like Bridge Power Ratings, Hybrid Ratings, Basket Power Ratings, Foul Ratings, and Coopertition ratings, which would probably help a prospective picking team pick out complementary teams. Come to think of it, that would be useful enough to be worth trying to do.