I’ve been working on some stats pulling from The Blue Alliance’s new BigQuery database, pulling all the data from every single year into my own postgres database that I will release eventually. That’s over 380MB of data, 113k matches, 28k awards, 5.9k teams, 1.5k events and 2.6M (yes, million) match score breakdown elements. The code that drives it can be seen here
To get into the meat of it, I’ve compiled the ‘standardized scores’ for each event since 2002 and put them on a map. These are computed by the formula (X - avg(year)) / stddev(year), where X is the match score per alliance. Those are averaged together for each event.
Bigger, more blue dots represent stronger scores, while smaller, more orange dots represent weaker scores.
I’ve also isolated the scores for 2017 and sorted them into districts (continental US / Canada only, gray = no district / offseason event).
To make it a bit easier to visualize, I’ve averaged out the standardized scores for each region and colored the regions accordingly (orange = weaker, blue = stronger)
… and the average (over lifetime) for the entire period of 2002 - 2017.
A while back, I compiled the scores/OPR/etc for Kansas teams. Not as systematically, as done here, but to see what teams could be considered the best at playing a particular year’s game. In that question, you do have to account for strength of the event attended, which I think isn’t necessary here because it is basically looking for the event strength and then combining that for the state-wide (or other geographical) average.
One thought is looking at the other parameters for scoring the event, which could give a greater insight into the difference in play styles between regions, if any, and also the progression of the meta-game. Of course, some of these are less accessible or more intensive to calculate than score. Also, there can be some inconsistency between years which makes interpreting them less robust than the score.
The maps look good, but I am sad that Missouri is boosted by the championship effect. I am wondering how money and time play into all this. Also, maybe team retention. I’d figure that for the most part that scores (game success) are better with more regional financial support, better retention, and older teams, but at the same time, there may be some instances where younger teams/regions are doing better, who knows.