Thread: Scouting
View Single Post
  #7   Spotlight this post!  
Unread 09-07-2012, 08:11
IKE's Avatar
IKE IKE is offline
Not so Custom User Title
AKA: Isaac Rife
no team (N/A)
Team Role: Mechanical
 
Join Date: Jan 2008
Rookie Year: 2003
Location: Michigan
Posts: 2,149
IKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond reputeIKE has a reputation beyond repute
Re: Scouting

On Team 33, we have a lot of active members engaged in scouting. We typically have 6 scouts (1 for each robot), and 1 master scout organizing assignments. We scout each robot to a sheet that has multiple cells. An example of the 2012 sheet is here. We often have specialty roles as well like human player scouting in 2009, Minibot scouting in 2011, and pit scouting.
We also typically have a "master scout" that does match scouting. Match scouting is where you scout th 5 robots who will be in your match before the match starts. We like to use the previous match data, though first match of the day relies on pit scouting, and after 3-4 matches, you can use "data trends" more reliably though this data is still pretty spotty. This master scout reviews what they think the opponents will likely do, and what they are capable of. They also cover what your partners are likely capable of. Lastly they develop a general match strategy to review with the coach. Ultimately, strategy is up to the coach, but it is the scouts responsibility to let them know what to expect, and to give a recommendation.

I have also put together a "generic" scouting booklet located here. There are lots of issues with subjective data, but it is usually better than no data. This booklet gives the user some nice options to try to organize observations about a particular team, and to ensure you get some data on every team.

Remeber, the msot important rules to scouting:
#1 Know thy customer: A bunch of useless data is useless. Collect meaningful data for its intended use (match strategy, pick list, future robot improvements, a white paper on the correlation of # of wheels with success...).
#2 Data in a computer with a dead battery is essentially useless. The coolest custom program still needs electrons to run. Have a good power plan that does not rely on the venue providing power in the stands.
#3 Web based systems require the web. There are some awesome web based systems that teams ahve showcased, that they could not use in competition becuase lack of signal (poor signal, or venue not allowing wi-fi, or spontaneous field issues resulting in wi-fi shut downs or...) Have a back-up plan that uses paper and pencil.
#4 The top teams are easy to pick out, but sorting 20th to 28th best is often very difficult. Pay attention to everyone.