View Single Post
  #83   Spotlight this post!  
Unread 23-10-2014, 00:56
Citrus Dad's Avatar
Citrus Dad Citrus Dad is offline
Business and Scouting Mentor
AKA: Richard McCann
FRC #1678 (Citrus Circuits)
Team Role: Mentor
 
Join Date: May 2012
Rookie Year: 2012
Location: Davis
Posts: 982
Citrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond reputeCitrus Dad has a reputation beyond repute
Re: A comment about alliance selection in off season events

Quote:
Originally Posted by artK View Post
Emphasis mine. I was intruiged by your assumption that eliminations experience leads to learning, so I did some digging. I looked at a number of second bots that 254 has worked with over the past few years and looked at the eliminations histories of these teams. Though I didn't crunch numbers (in part due to a lack of a metric to compare elimination records), it seems that teams who played with us as a second bot did not seem to have an effect on their elimination records at other events: teams that won in the past kept on winning, teams that never went far into elims stayed that way (heck, one team hasn't seen eliminations since playing with us ).

But a few disclaimers:
  1. Learning could take multiple forms, but the easiest way to try and measure that is eliminations records.
  2. I looked at the second bots from regionals/offseasons for my dataset, because I knew that they generally have less eliminations experience than us or our partner. I also picked the teams 254 played with because I had experience with the teams (which may have opened the door to bias).
  3. I wasn't really sure how to measure elimination histories of teams, but the basic metric I used was to count how many times they made it to the semifinals or later each year. This metric is less than ideal, because it weighs wins and semis appearances equally. A fairer metric would be to do something like districts.
  4. I did this counting by hand, so it is not very precise. If I were to do this again with greater rigor (which I may well do if I have time), I would probably look at Michigan for cleaner metrics/data.

Tl;dr- Playing in elims once with really good teams doesn't seem to have an effect on future performance, though a more rigorous analysis is needed.
That's an interesting metric, but I think I"m looking for a different one: of those teams that played with you, how many are still operating compared to the overall average? I suspect the thrill of playing with 254 probably boosted the enthusiasm for those teams for some time and increased the number of team members in the future.
Reply With Quote