Strength of Schedule/Strength of Alliance

This is my first and last year as part of FRC. I was recruited by an existing member for my statistical analysis (I can program about as well as a ham sandwich, and I couldn’t even build that.) One thing I find very interesting is weighing the strength of teammates and opponents for the qualification rounds. A 6-8 team might be really good, just had terrible teammates and really tough opponents. By the same measure, a 10-4 team might be horrible but carried by strong alliances and dropped by a cupcake schedule. Do any of you all measure these statistics? If so, how do you measure them and how much do you weigh them?

Start by searching out Offensive Power Rankings (OPR) and Calculated Contribution to Winning Margin (CCWM).

Ed Law publishes some nice work in that vein.

What we, 68 Truck Town Thunder, and a lot of other teams do is scout all 6 teams each match. We count how many baskets they scored, how many shots they missed, how good they are at balancing, etc.

We put all this data into an excel spreadsheet. When it comes around time for eliminations; we have concrete data on how well each team did in the competition, rather then just going of how well it “seems” they did.

We also use our data to print out “match sheets”. We print a sheet for each of our matches, and it tells us important stats from each team like how well they score and how good they are at balancing. We then use this to decide with our alliance the best strategy for that match, like how to play defense against and who to coop with.

If you’d like, I should be able to get our excel spreadsheet and an example match sheet for you to look it. It’ll be write protected, so you’ll have to make your own, but it’s a good place to start.

I have a pretty good example of the importance of scouting. This year on Newton, my team was the 3rd Alliance Captain. Using our scouting data we were able to pick an amazing underdog alliance, partners with 330 and 639. 330 was around rank 24 and 639 was rank 89. Without our data, we never would of been able to pick our underdog teammates and become Division Finalists.

We used to get detailed data on every robot in each match, put in a spreadsheet, and calculate our own rankings. However, last year we ditched that and started using pencil and paper to write down qualities that stood out.