How many times do you scout a team???

[frc2590]

We do 3 separate scouts per team but mostly for the teams we play with and against hope that helps and be sure to make a pit scout in addition!
1 2 3
/ / /
scout scout scout=3 forms
+ a pit-scout(scout of the pits)
-there is no such thing as an over scouting :0

[Kims Robot]

Originally Posted by zant1234

Wow thanks everyone...you have given me some good ideas for scouting, now I just have to convince my team to let me do it

All of this said (and Crystal answered what 1511 currently does), if you are shorthanded there are two solutions.

1. What we did our rookie year
- Have 1 or 2 people at all times watching matches and taking data. (we used a single excel sheet that had all of the teams in each match listed, but you could also just have one sheet per team with their match numbers on it)
- They preselect 2 teams per match to watch for that match, and gather all the data they can.
- In the selection process, you make sure every team is covered 3-4 times.
- If one of the teams chosen is a DNS or just dies (or does very little), you gather data on another team in the match.
- Make sure you get one match per "time period"... we used to try for a practice match, a friday morning, a friday afternoon, and a saturday morning.

This method limits the number of people you need to scout but also gets rid of the "they were temporarily broken" or "temporarily awesome" issue. It is not as ideal as having data from every single match, but just like in Robot Design, you HAVE to do what your team is capable of or its just not going to be useful data anyway.

2. SHARE
Find another team that has awesome scouting and ask if you can help them in exchange for their data, or just have their data, or some other kind of trade. 1511 has worked with teams in this manner, providing our data to teams that don't have scouting systems or that come help us. Many teams are very open with their data.

Good luck!

[Dancin103]

Originally Posted by EricH

As many as possible. There is no such thing as too much data on a given team.

Definitely as many as possible. We do at least one time for every match a team plays between Friday and Saturday.

[EagleEngineer]

It matters on how important that team is to your schedule. you should scout at least every team once at a competition. But if a team is your opponent in a qualification match, then you might want to scout them more than once. Its all a matter of how many people you have on your scout team. The more the marrier.

[rajikurbaj357]

we have our 6 scouts and assign a spot for each, that way whatever team is in that spot is scouted, and we do that for every match up to the playoffs.

[NullEntity]

We have a person scout every robot every match. They fill in bubbles on a form then we place it into an automatic scanner. We can then generate reports on all of the robots!

It works surprisingly well.

[davidthefat]

We don't scout, but honestly, what do you scout for? Yes sure, how many points they scored, whether it was an offensive or defensive bot. But how do you get an honest answer regarding the robot itself? Most are not numbers on screen. You really can't judge a defensive robot's abilities by asking the teams. In other other words, how do you get an accurate judgement qualitatively and not on the quantity??

[ChrisH]

Of course, what you DO with the data collected is more important than the data itself. Team 330 generates spreadsheets that process the data collected. Weighting factors are used to assess the relative strengths of the robots.

Here's a simplified example of how that works. Assume that we figure all of the action is going to be tubes. Our initial cut at a spreadsheet will have a high weighting factor, say 70%, assigned to how many tubes a robot put up in a match. If at the event we find minibots to be more important, we simply change the weighting factors of the minibot and the tubes and the spreadsheet produces a different result, showing the high performing minibots.

These things can be set up and verified before the event so that while you are there you can focus on collecting the data.

Pure performace is not the only data to worry about either. While it is helpful in determining what a team is capable of, it does nothing to reveal unexploited weaknesses in a robot's design, like last year being able to be shoved into a goal and immobilized. This type of information takes a different type of scouting than just counting up scores.

How teams work together on an alliance is also important. I know of one team that might have lost out on a World Championshiop because they gave the person sent to talk to them about pairing up in eliminations the cold shoulder. That honor went to another team on our short list that was a little more friendly. I've always wondered what would have happened if things had gone differently....

[hat_ios]

Begin scouting early on Thursday and have a solid database of team info. This comes in handy during alliance selection. Being able to see how a robot/team has progressed could cause you to pick them or not pick them. I also suggest taking photos of each robot and having these photos available during your pick meeting. The drive team and pit crew often interact with many different teams so the photos are a good way to spark their memory.

[CalTran]

I agree with everything that has been said about scouting every match every robot and the pit scouting but I also have some stuff to add:

Media aspect! If you have someone with a video camera, get teams of interest ON VIDEO. If there's a team you see in the pit with a wicked awesome manipulator, VIDEO IT IN ACTION. If you hear about a team being a power house this year, VIDEO IT IN ACTION. Point is, if you get everything on video, you can review it in the pit, Friday night, and even during the day Saturday if you're having a hard time deciding on your dream alliance. A video that you can watch speaks volumes about how a team performs.

[Res0lu7ion]

as the person who's been head of scouting for 522 for the last 3 years you need to scout every robot in every match and talk to your team strategist every single time your team goes up. successful results every time

[Kims Robot]

Originally Posted by davidthefat

We don't scout, but honestly, what do you scout for? Yes sure, how many points they scored, whether it was an offensive or defensive bot. But how do you get an honest answer regarding the robot itself? Most are not numbers on screen. You really can't judge a defensive robot's abilities by asking the teams. In other other words, how do you get an accurate judgement qualitatively and not on the quantity??

I really hope you aren't serious on the "what do you scout for"... I never say things like this, but if you wanted to improve your record, you might consider scouting.

See this post for how we use "subjective" data.

We give relatively little weight to our pit scouting (what you ask the teams), but it is very relevant to deciding what you "might" see them do on the field against you. And watching matches is INVALUABLE... We use all of it to not only decide our match strategies but determine our alliance selection list, and have made it to elims all but 3 out of 19 events (alliance captain for many).

[bmlarson12]

It is best to try to get as many matches as possible. I also found this website that has some useful information, http://frcdb.net/

[roboX]

we scout every team every match if something happens to a robot, we know about it

[pyroslev]

Constantly. I tell my kids walk around with a notepad in their pocket. You scout every match they show up in. If they don't show for the match, you get someone to see why they are not on the field. If they are not there cause of major issues, that is worthy of note. If they are not there cause of adding a system, then it's worth having someone watching them on the practice field. There is no such thing as to much bad information.