Team 4915 "Spartronics"'s Scouting Process

[phroa]

Hi all,

In the wake of our first two PNW regionals at Glacier Peak and Mount Vernon, we've created a quick article describing our process and wishes for the future.

We're intending to use it to guide us in the coming years when the seniors have left, and maybe provide some insight to other teams like you.

The report is available here: https://github.com/Spartronics4915/2...nt%20Vernon.md and references several Ruby scripts and an SVG file in the same repository.

Questions and comments are appreciated!

Thanks,
Jack Stratton

[BenjiSG]

The struggle to manage collecting both quantitative and qualitative data is definitely a tricky one. Our team's solution was to limit quantitative data-collecting to Day 1 of the competition, rank teams that night, and focus on using the Qualifiers on Day 2 to gather more qualitative data about teams we have interest in so that we can adjust their rankings accordingly as we go, similar to what you guys described.

An interesting difference however is that this year we begun to have the Head of Scouting make alliance selections rather than a member of drive team. This was born out of a scenario that happened when we were picking our alliance on Newton at Worlds last year as the 6th seed captain. Our pick for the second robot went fine, with the captain of our drive team following the list, but for the second pick he got mixed up and picked a third robot that used the feeder station instead of the highest recommended landfill bot on the list which ultimately made us go out in quarterfinals.

We actually do paper scouting based on team rather than field position, but we make sure to use one sheet per team and simply pull it out each time that team has a match, which winds up giving us a fairly comprehensive report on each robot.

Additionally, we make sure to do pit scouting and gather pictures of each robot there during setup so that we can keep them for reference during competition. Later, we can cross-reference our field and pit scouting to see if teams claim to do anything they haven't been trying.

[Liu346]

Hi guys it is very interesting how yall have approached this new issue of scouting at higher levels. Our team, 346, has taken a new approach this year to utilize our large team during this process. We have begun a 19 person scouting cycle. Yes this number may seem very high because most teams use only 7 or eight people but I will explain why it has become so useful. Since our team consist of new 70 students we decided to begin utilizing the resource of a large team so we began to break down this years game with the mindset of having a large scouting team. We have 2 people scouting each robot for each alliance. One is for qualitative data such as where it shoots from or why it has become stuck going over certain defenses. When the other scouter is looking for straight quantitative data. This would be like number of high goal shots and how many times they cross the ramparts. This is then sent to compilers who enter data into advanced spreadsheets that then tabulate the data and begin ranking the teams on the different data points. While the quantitative data is entered into the computer the qualitative data is fled in a filing box to be used later that night. We then have the data analyzed by a group of students and mentors to decide on strategies and defenses for the next match. This takes about 18 hard working students to pull off. This does not even include the students from other cycles so we can give our hard working team a break. Finally we use the qualitative data to cross reference data that night to create a pick list for the next day. Good luck in this years STRONGHOLD.