Quote:
Originally Posted by BigJ
I am guessing a lot of teams are duplicating a majority of the data they take (how many penalties a robot commits, how many objectives a particular robot completes, etc). This is a wonderful opportunity to crowdsource the statistics on a web app or something, but the crowdsourcing idea runs into problems OTHER than not having internet at the venue.
If every team thinks that other teams will be recording the statistics, you end up with a dearth of contributors. Some teams will not want to put their scouting trust in a system of other teams. If the system/app goes down everyone participating loses their statistics, etc.
|
A crowd source app would be great, there have been many attempts at it over the years, however I think someone should take some initiative and get a Android/iOs App for this together. Figuring out who is inputing data, and who's data is correct is a major challenge.
Quote:
Originally Posted by Chris is me
What could possibly be more important than an objective measure of how many points a robot scores? With 100 teams in a division and a dozen or so students watching matches, trying to get consistent comparisons between teams without some objective metrics is impossible. I don't think a purely number-of-balls pick list is a good idea, and qualitative statements are important, but the more you can make objective, the smoother the process will be.
|
Its not about who scores (that is an easy question to answer), its where they shoot from and the strategies they use. You can make this process a lot more efficient if you apply
technical analysis to the game statistics. How an alliance works as a whole changes the situation a large amount every match. See below.
Quote:
Originally Posted by XaulZan11
I'm very curious to know how you scout. Could you please share?
|
I love questions! This is exciting.
The scouting/strategy method that I use was actually designed mostly by my Dad, but I do a lot of work with it. Both of us have backgrounds/careers in Finance but somehow found our way into FIRST. We apply the same ideas of stock/equity analysis (finding a stock to buy) to scouting/strategy. Most of it is Technical Analysis (ie - looking at match score averages), this allows us to sort through all of the teams at an event very quickly. Much more efficient than using scouts. After that we get into
Fundamental Analysis/Qualitative aspects (ie - where do teams shoot from) to better determine what we want to do in matches/alliance selection.
The key is to understand the short comings of your statistics and then do some minimal match/pit scouting to better answer any questions that could be raised. The less time you spend gathering the data, the more you can use your time to use more effective scouting strategies.
This thread really isn't the correct place to discuss all of it so if you guys are interested I'll start a new one. But look into 1114's presentation they gave on scouting, it is on the right track.