Really interesting data. A couple other stats of interest came to mind that might put things in perspective:
-Rank percentile of each team at that event
-Week number of the event. Everyone knows play ramps up as the season goes on, so a high RS/match early in the season was probably harder to obtain
I think it would also be cool if you could re rank the teams by overall RP for the year instead of by each event. So like teams are ranked by their average RP from all the events they have competed in so far instead of by each event individualy.
So this is the first ‘ranking’ system where I see everybody ranked. But having watched this game, I feel like RP’s aren’t really a very good measure of a team’s ability.
We have numerous threads out there ranking OPR, ranking the Missouri teams with some ‘mystery’ number’, ranking Michigan teams by OPS.
I want to hear some discussion by some real statistician-type people that tells me what the ‘best’ ranking system is,
Then it would be great if, like on this thread, RP’s/match, you made a spreadsheet to rank everybody.
I didn’t realize that our ranking system was still a mystery. Turns out the following quote does leave all of the readers in limbo. Our private e-mails with Missouri teams have been more specific.
Allow me to reveal the mystery.
We used FIRST’s district ranking system as described in section 7 of the admin manual for a team’s first 2 events. In the cases where a team only attends 1 event their points (except for the culture awards - EI, RCA, or RAS) were doubled.
So Alpha Beta, is this the ‘best’ ranking system and why?
I’ve just looked at the formula for District ranking (for the first time–don’t have districts in Kansas) and “wow”, it certainly looks like the most complicated!
Thanks for telling where you got the numbers for MO.