It shows average time of possession (in auto and tele), average cubes in the vault, average robots crossing the line in auto, and average climbed robots (including levitate) for each team at each event.
Some fun facts:
If you average above 110 seconds of scale possession, you’re in the top 1% of teams.
225 and 1538 had the most average seconds of auto scale possession in the world (8.0) at CASD and MRCMP respectively.
Only 19 teams averaged over 11 seconds of auto switch ownership. No one averaged 12+.
The distribution of scale possession looks pretty amazing (that’s 254 at Arizona North way up there at the top): https://i.imgur.com/15ebwAn.png
I think we should expect that scale ownership averages will go down from this point in the season onward, because keeping it tilted your way becomes more difficult as you face better opponents. Actually, total time of ownership will decrease because the scale will spend more time balanced.
Hey I hate to ask, but do you by any chance have the equation of that graph? I’m trying to make a .js application that will allow my team and others to determine some quick facts about how a good a robot will be with them fast, ad that would allow us to add percentile to that list. Cheers!
It looks like a sigmoid graph oriented vertically. I don’t know the exact numbers but the rough estimate would be something like: Percentile = 100 / (1 + e^(-possession / 10 + 5.5)). The graph shown has possession on the Y axis and rather than percentile uses “total rank” (I suppose) where the maximum is approximately 6400.
