What should this year's scouting app look like? What questions should be asked? Any tips?

The 2024 season has officially started and it’s time to make a new app! I haven’t seen any topics about scouting yet so I wanted to start one. I am the scout lead for team 8046 and I use MIT App Inventor to program and design an app. I then use google sheets with a bunch of formulas, which is where I transport the data to. I have started brainstorming questions to ask but I wanted to see what other scout leads were thinking! This years game is very complicated with scoring and different tasks throughout, what are the ways your team is thinking to find the data throughout a match?

8 Likes

This has been going through my mind quite a bit since we would previously assign a member to each alliance for scouting. However, I believe this year, it may help to assign multiple people to an alliance for a more in-depth scouting review. I think the organization of scouting teams is critical this year for scouting. If you have the resources, maybe have a team scouring the live stream of matches who can report more important details than just the Blue Alliance page.

That is very interesting! Our team assigns 1 person per robot per match. We have a team roles sheet where we assign each team member to a roll for a certain amount of matches so people get to do many roles, takes turns, make sure everyone gets a pit spot, and so there isn’t too many people in the pit! I think this has worked pretty well for us! But it is hard with a smaller team. Our team is about 25 members and it works well for us.

1 Like

We want all the following data if isn’t too overwhelming for scouters to collect it:

  • Every action that can result in points
  • Cycle times
  • Shooting positions
  • Pickup locations
  • Pickup method (ground, source)
  • Strategies that need to be defended like stashing pieces next to the speaker to wait for amplification
  • Defense effectiveness

Scouters miss things if you ask for too much from them though, so usually not possible to get it all. Here is the scouting form mockup that we are proposing for our scouting app.

1 Like

Your list looks similar to ours. We divide our scouts into 3 groups.

  • Pit scouts: walk around and gather basic robot features & capabilities
    • swerve vs tank,
    • can score in goals X, Y & Z,
    • what autons can they do, if any,
    • can endgame, etc
    • (so we have something in early matches before we’ve seen them)
  • Quantitative scouts: 6 per match, each recording the “numerical” things about that robot’s performance (app based):
    • points scored (&where),
    • penalties earned,
    • cycle times<1>, etc
  • Qualitative scouts: generally 1 per alliance, watches for non-numerical things:
    • played defense,
    • was well defended,
    • locations used,
    • bad driving, etc

<1> if you record both pickups and shots/drops, you should be able to get cycle times “for free”

2 Likes

How do your scouters know this? We don’t collect penalty data at all because it is hard to watch both the bot and the refs. And even if you see a ref give a penalty, it isn’t always obvious which bot it is for.

Blue Alliance reports the number of penalties for the alliance. So if you are team x and take the average number of penalty points that are added to team x’s opponents, then that will probably give you the best idea of how many penalties team x is causing.

Best effort. Sometimes they’re pretty obvious, some times not.

A strategy to better utilize match scouters could be to just have them focus on actual match scoring for a single robot per game. A lot of data can just be gained from pit scouting.