Quote:
Originally Posted by Brian Maher
Here are my thoughts: - Most of the pit scouting info can easily be observed from matches. I recommend using pit scouting to assess things that are harder to observe during the match, such as drivetrain details, type of shooter/climber/intake mechanism, or assessing robustness.
- Four balls per second is too low a maximum. Even if four balls per second were to be the hard limit for a shooter, consider that some robots will have multiple shooters.
- The sheets say nothing about auto. The gear is worth 50% more in auto (an extra half a rotor) and fuel is worth three times as much.
- The sheets say nothing about the number of shots or goals per cycle. There is no way to differentiate between a robot that puts 10 fuel in the high goal and one that puts in 50.
- The sheets say nothing about the number of fuel cycles or fuel-scoring actions. There is no way to differentiate between a robot that scores 50 fuel once a match and a robot that scores 50 fuel four times a match.
- The sheets say nothing about how game pieces are acquired. Loading station? Off the floor? Hopper?
- There's nowhere for miscellaneous notes.
Note that by collecting so much objective scoring info via pit scouting, your data will likely fall prey to what I like to call the "pit scouting effect". People tend to remember what their robot can do, not what it typically does. I'm not saying they lie, it's just what they remember. Keep this in mind when analyzing pit scouting data.
|
You should definitely observe what people do during their matches but there is some value in determining robot types on Thursday. If you're match 1 on Friday you won't know anything about what your partners or opponents do unless you ask and observe on Thursday.
And as Brian said, you really need to know where they load from. That will influence cycle times and defensive maneuvers.
Adding a place for generic notes is crucial. If a scout notices something crazy cool or crazy bad they need a place to record it.