We’ve done pit scouting for various reasons; we’ve match scouted; and we’ve used our version of the “Super Scout”.
I was wondering if any of you use the “Super Scout” concept. What do you task you super scouts to look for during the competition?
Can you define the “super scout”?
11 Likes
Our head scout was mostly on the lookout for strong defense because it was so difficult to scout objectively. He was also keeping an eye out for where particularly strong scorers liked to play so that we could play more effective defense in matches against them.
3 Likes
It can varies. The super scout will scout for specific abilities or details about specific teams. They have a very distinct set of characteristics they are looking for/at during matches.
I would recommend reading 1678’s 2018 whitepaper, they have a section that explains the role of the super scout, and how they oversee general scouting.
3 Likes
It is very difficult to scout defense reliably. We’ve tried to quantify it, but quickly realized that my 5/10 might by 7/10 by your standards.
We’ve tried by the effectiveness of team A stopping team B from scoring. Even that can be difficult due to team B’s inconsistency in scoring without someone playing defense against them.
However, usually a high scoring team is usually consistent.
-Comfort with certain game objectives (High Rocket and Climbs this year)
-Defense and how aggressive/comfortable they are playing it
-How do they drive? Does it look they they unintentionally swerve one way? Can they make small adjustments quickly? How quickly can they realign)
-Driver reaction speed and awareness (Are they getting in the way of their partners and can they respond to a partner getting in their way or to an unintended consequence like tipping over)
-Drive team “friendliness” (Do they seem to yell or be bossy to their partners?)
Those are most of the criteria I use while SuperScouting at least. Some of these are hard to see and dont end up being noted, but if they are noted, they can impact our pick list a fair bit.
1 Like
Maybe if we had more students on the team, we could have a super scout position.
Yeah we tried to explore some kind of DPR analogue but found there wasn’t really a big enough sample size that it mattered. It was pretty easy to catch by eye though so we went with that.
Our 2018 scouting whitepaper explains how 1678 uses super scouts in terms of actual scouting and the app interface, but the role of the super scout has actually changed quite substantially compared to last year – our 2019 white paper will be released soon (I’m not in charge of it).
In 2018, super scouts were responsible for turning qualitative data into quantitative data on things like driving speed and agility, defense, or good and bad decision making (which we changed this year). The super scout app didn’t change much between competitions in the 2018 season, because driver ability was really the only useful data point for that season, and our super scout data was reliable and accurate.
In 2019, we (along with pretty much every team) realized after week 1 that defense was going to be huge, but ended up using a very similar super scout app for our first two competitions at the Central Valley and Sacramento Regionals. It turned out that the simple rating of 1-3 for defense was not sufficient for us because the defense rating averages did not match up with strategists’ notes and observations, leading to second pick lists that were based off of anecdotes and very little defensive data.
We decided to revamp the super scout app to take inputs from both strategists and super scouts; strategists inputted much more complex defense ratings that included counter-defense and push ability ratings as well. Super scouts continued to input driver ability data. The key difference, though, was that the two strategists + three super scouts were all told to take written notes on defense bots at AVR and Champs, which gave us more eyes on the field and anecdotes that verified the defensive data. Subsequently, our second pick lists at AVR and Champs were much more reliable and data-accurate.
Please feel free to ask any questions about our strategy/scouting process! I’m happy to answer.
I don’t know what others do, but I have them watch teams closely that we’re playing with/against in our next match(es). I think of it as being a spy - we need to figure out what their strengths/weaknesses are and how they can be exploited/protected.
Additionally, they’re useful for getting additional info on teams you’re interested in for picking purposes.
1 Like
How do you guys decide who are strategists and super scouts?
Our core leadership team selected both the strategists and super scouts for this season. Next year, we are planning to have all super scouts and strategists be participants on the strategy subteam (this season super scouts were not on the strategy subteam, which was not ideal), as all super scouts and strategists participate in draft night at competition and should be knowledgable about our strategy process & any changes to it during competition season.
Offering the super scout position is a way to add prestige to scouting and also to encourage competition by pointing out the super scout is one who is most accurate and observant of the regular scouting team.
1 Like
I get 3-5 kids to work with per event, prestige doesn’t matter when we have enough issue getting data in general.
2 Likes
We have used the super scouting concept sometimes throughout competitions, but on a case by case basis. We usually send our top scouts to go to high variability teams and defense bots to watch how they play and see how and what they are working on in the pits. This allows us to better understand when we have weird data or only quantitative data for defense bots.
Ah that’s unfortunate. Generally we have no problem with sufficient numbers but this year was a bit lean for us also.
I’ve been in the easy position that there is typically only one student who wants to be the strategist - so that position gets self-assigned. As for super scouts, it was usually “I need someone to super scout this match” and whoever volunteered would do it. In general, teammates want to help the team and they can immediately see the effects of super scouting so I’ve always had volunteers for super scouting.
Scouting Lead for 2733 here.
For the past two seasons in which I have been scouting lead, I work on match strategy and give suggestions to the drive team as to what position we should play. Additionally, I tell them what abilities other teams have showcased in their matches. Because of this, scouts know that their data that they have been collecting for all the matches is used to help us win, and it is sort of used as an incentive to keep them scouting. I am also supposed to manage scouts but haven’t really been doing a good job at that so…
2 Likes
So we do pre-scouting, pit scouting, and regular match scouting, but we choose two people to be super scouters. The two people we choose are people that will be available for every event, so there’s not change in opinion/views.
So basically, our super scouts are each assigned an alliance. They enter the team numbers of all three robots on the alliance they’re watching and record the main focus of each team/driver individually, whether it be rocket, cargo ship, or defense. At the end of the match, they rank the robots 1-3 based on THEIR views after focusing on all three robots the entire match. They then score each robot’s ability from 1-5 and driver ability 1-5 (driver ability meaning drive team chemistry; how well they control the robot and how well they collaborate with the drive team from the other two teams).
If you have any more questions or want me to explain anything more, just let me know!