Scouting sheet input requested

[Wildcat]

Attached below are the scouting sheets i have made for my team for this year, one is the sheet itself and the other is a direction sheet on how to fill it out. Some constructive input from the CD community on them would be greatly appreciated. Thank you.

[Ian Curtis]

You've got a great start here. Some things to consider:

1. Scouts get bored.
2. FIRST matches happen fast, so minimizing the amount of data you ask people to collect is paramount to getting data that isn't junk.

Do you need to know if they can score tubes every match? Do you need to know the autonomous path, or is just the starting point good enough? Does the minibot finishing need to be a separate question from where the minibot placed? Do you want to ask the scout to approximate a time to get tube and score (could tubes/match be an acceptable alternative that doesn't require guesstimation)?

Is there any other information you might want to know? Mecanum drive, 6wd, etc?

I'm a quantitative kinda guy, I really like to go for numbers with as little subjectivity as possible. The trick is collecting them without human error messing them up too badly.

2 cents.

[O'Sancheski]

[quote=Ian Curtis;1015259]You've got a great start here. Some things to consider:

1. Scouts get bored.
2. FIRST matches happen fast, so minimizing the amount of data you ask people to collect is paramount to getting data that isn't junk.

this is very important. My rookie year i was a scout and it gets very boring, and the matches happens very quickly. Two minutes is not going to be long enough to fill in all that data. Then having to shuffle through the other sheets to find the next teams. Good so far, just try to make it as basic as possible. Just remember that rookies are most likely to be scouts, so making it as simple as possible is great.

[Zoughtbaj]

Here's some things I noticed:

-The "did the minibots finish?" and "was deployment successful?" can be eliminated. I say this because since you have "did they attempt to deploy the minibot," and the placement, you can probably deduct whether or not it was a success by whether or not they mark a placement for the minibot.

-Something our team is considering is using a grid of the pegs to show both in an auto section and a tele-op section. If the ubertube was scored on a peg, the scout would mark in whichever box of the 3x6 grid where the ubertube was placed. Additionally, the same can be done for tele-op; you can even have the scout mark the shape so you can determine if they made a logo. This kind of system could replace a lot of the questions you have and save you some space. If you want a visual, I can make one for you.

- The "Lane entry" section could probably be made into a question. For example, "where did they enter the lane," under that, "entrance," under that, two check boxes, one saying "big" the other saying "small." could save you some space.

-You might consider having comments sections for all three phases of the game. While quantitative data is usually the most useful, sometimes, if the scout isn't vague, you can make some decisions based on what they remarked.

-For the Pizza orders, I recommend putting it on a different sheet instead of your scouting sheet. Like, create an excel sheet, have your team member's names run down it, and a column where they make their decision. The options can also be on the sheet somewhere. This would save you that little bit of space.

-For better quantitative data, I recommend not using the total score of the match, and instead using the total score of the robot itself.

-For reference, I recommend adding a match number. Quite often, robots will have difficulties early on but will refine the problems as they progress further into the game, making data without a match number sketchy sometimes.

In summary, experiment a little. Try to edit it and find what makes it better, and have fun with it! To me, making the scouting sheet is one of the funnest jobs on the team

[AlexD744]

Looks pretty good, but theres a few things I personally would take out because I simply don't see how it's useful. Remember that a lot of times scouts resent their job if they know that the information isn't going to be used. And a scouts job is made easier (and therefore the information is better) when the questions are succinct.

-I don't think someone ranking teams will care if the team loads over the top or through the slot, as long as they load quickly. However, knowing if they can floor load is important.

-I also feel that "Did they drop tubes?" is an unnecessary question. Every team is gonna drop a tube now and again, but I would say that if a team has a regular problem with holding onto their tubes it would be a good comment to make.

-I think that the place (1st-4th) or an apx. time would be sufficient in replacement of the two questions "Did they deploy a mini-bot?" and "did it reach the top."

In general, I've found that scouting sheets that short and sweet and to the point are the the best. Also, quantitative values are typically easier to use in a scouting meeting than qualitative ones.

[Pamelot2067]

This is what I've come up with so far. Its quicker to fill out and less confusing

Team: __________ Match Number: _________
AUTONOMOUS
• Placed an Uber Tube? YES NO
• What row did they score on? TOP MID BOTTOM
• Line tracking? YES NO
• Vision tracking? YES NO
TELE-OP
• Can They Score Tubes? YES NO
• Did they create the logo? YES NO
• Highest row scored on? TOP MID BOTTOM
• How many tubes did they score? ______
• Did they drop any tubes? YES NO How many? ____
• How did they collect tubes? GROUND SLOT OVER TOP
• Aprox. Time to get tube and score? _____
• Offence or Defensive play?
END GAME
• Did they deploy a minibot? YES NO
• Did their minibot finish? YES NO
• What place did their minibot come in? 1st 2nd 3rd 4th
List Any Penalties This Team Received
__________________________________________________

[HannahF]

Things you could add:

• Was the team playing Offense or Defense

• Did the robot fall over/ Is it easily tipped/Could it right itself

Could you explain to me the part at the bottom about the pizza orders....Is this some kind of code Why are you using your scouting sheets to order pizza?

Besides that I'd say this is a good scouting sheet.

~Hannah

[Wildcat]

Originally Posted by HannahF

Things you could add:


Could you explain to me the part at the bottom about the pizza orders....Is this some kind of code Why are you using your scouting sheets to order pizza?


~Hannah

last year at states, my teams scouting group ordered pizza one day, the next day more people wanted to order pizza, so we ordered more. by the third day we were getting pizza for my whole team which overall was a huge hit, so we decided to integrate it into the scouting sheet to make taking orders easier on us

[BJC]

That is a pretty good scouting sheet. I do, however, have a couple of suggestions.

- we like to use scale of 1 2 3 4 5 (circle one) to judge things like speed, driver precision, tube pick up speed, and other qualitative yet important bits of data. We have found that although people judge the scale differently when you put all of a perticular robot's scouting reports togeather it gives a fairly clear idea of what that robot can do.. anyway, it's better than "fast- (yes) (no)."

-Our team is old fashioned-we also use paper. We like to fit every team onto one page and file them away. Every page has probably 8ish matches per sheet so nearly every match of every team is recorded and filed. We have found that this goes a long way in staying organized when using a paper system. Theres no, "wait wheres team #### I though you had their 3rd match? Where is it?" etc.

-We also like to have lots of people cycle through match scouting. We have several main scouters which are filing and keeping control over the match scouting, one or two who go and pit scout, and 2ish mentors who make specific notes about robots (which will need to be further reviewed on Saturday), take pictures, and keep general tabs on the overall scouting process. It's a good system which has worked very well for us in the past.

Again, good start.

[XaulZan11]

Originally Posted by BJC

we like to use scale of 1 2 3 4 5 (circle one) to judge things like speed, driver precision, tube pick up speed, and other qualitative yet important bits of data. We have found that although people judge the scale differently when you put all of a perticular robot's scouting reports togeather it gives a fairly clear idea of what that robot can do.. anyway, it's better than "fast- (yes) (no)."

Be sure your not mistaking Precision for Accuracy. Being able to say "This team on average has a speed rating of 3.54" is still based on the biased perception of the scouts.

In terms of the sheet, what about teams that fetch tubes (ie pick up tubes and drop them into the home zone for other teams to score)? For our scouting sheets, I'm thinking about keeping track of the number of tubes brought into the home zone (even if they are scored or just 'fetched'), but that could get redundent.

I doubt our scouting sheets will have the mini-bot place (1st through 4th) because that is heavily dependent on partners and opponents mini-bot speed. I think we will keep track of time it takes for the mini bot to reach the top from the time the robot touches the tower.

For autonomous, you can make it a little easier and say "Where did they score (circle one): top, middle, bottom, none".

I think the "Approximate time takes to score tube" will be shown in the number of tubes scored. I think you could get rid of that and make it eaiser on the scouts.

A minor thing, in the tele-op box, I would re-order the items so the mini-bot stuff is all at the end, so the scout can just go right down the list.

Another tip that we have done the last few years, for things that teams will do several of (like score tubes, score balls last year, laps in 2008...), we have put little numbers on the sheet that the scout checks off during the match. Its a lot easier than trying to count on your hands, especially when there are several things your keeping track of.

Overall though, I think this is a very good scouting sheet!

[BJC]

Originally Posted by XaulZan11

Be sure your not mistaking Precision for Accuracy. Being able to say "This team on average has a speed rating of 3.54" is still based on the biased perception of the scouts.

Just to clarify my above post:

I would consider Precision how well the drivers drive. Although it is not easily defined it is very apparent which robots are well driven and which are not.

I consider Accuracy how precise the drivers need to drive to be successful. For example this year if you have a roller claw to pick up tubes off the ground you generally need to be less accurate then if you are picking tubes using the inside hole.

and Yes, the 12345 metric can be bias, however, it is not used to find a average. It is instead a tool for scouters to consider when making a pick list. It is not perfect, however despite all flaws it has worked amazingly well for our team. Combined with all the other scouting info it gives a great general idea of how good a team is at said function.

[XaulZan11]

Originally Posted by BJC

I would consider Precision how well the drivers drive. Although it is not easily defined it is very apparent which robots are well driven and which are not.

I consider Accuracy how precise the drivers need to drive to be successful. For example this year if you have a roller claw to pick up tubes off the ground you generally need to be less accurate then if you are picking tubes using the inside hole.

I was talking about Precision and Accuracy in the statistical/numbers sense, not in the robotics sense. Basically, precision does not mean accuracy. I can say there are "1,002,353,234,354,787" grains of sand on a beach and be very Precise, but still be innacurate. On the other hand, I can say there are over 1 billion grains of sand. I wouldn't be Precise but I'd be Accurate. Obviously the key is to balance precision and accuracy.

Back to the robotics, by putting a number on the speed/driving ability/pickup speed, you are precise, but not necessarily accurate. I think people have a tendency to see precision (a number) and assume it is accurate.

[Wildcat]

attached below are the most recent updates of my sheets

[Sue-San]

That is a pretty nice scouting sheet

Just a few suggestions/comments:

• Like stated before, try to take out the subjective stuff. Is there any way you can remove the whole ranking box and make it into yes/no questions to avoid too much judgement made by the scouts?
• Have you considered adding anything about defense?
• I like the portion about the lane entry

Good Luck!

[IKE]

Originally Posted by BJC

-Our team is old fashioned-we also use paper. We like to fit every team onto one page and file them away. Every page has probably 8ish matches per sheet so nearly every match of every team is recorded and filed. We have found that this goes a long way in staying organized when using a paper system. Theres no, "wait wheres team #### I though you had their 3rd match? Where is it?" etc.
.

This is in my opinion the best tip of all (hey it our system, so I am biased). The nice thing about this is you have a clear overview on a single sheet of how a team does. Doing 1 sheet per team per match will lead to 80*6 or 480 sheets to sort through from 40 teams at a District event. If you have a single sheet you can then review 8 or so matches at a glance. Each team has a sheet, thus we have 40 sheets at a district, 64 at MSC, and 80-ish at the championship. Much more manageable, and more importantly Actionable! Here is our Championship sheet from last year. It only has 5 matches, but this covered through Friday, and gave really good trending data.
http://www.chiefdelphi.com/media/papers/2370
Teams frequnetly remember their best matches and believe that is their "Average" performance. A team that we are good friends with asked us to pick them because they could score 11 balls in the home zone. We had a frank discussion with them about their consistency. I showed them their sheet. Yes, 1 match they scored 11 balls. It was the only match they scored more than 4. Most matches they were under 2. If you just looked at their average, around 4/match, it wasn't too bad. If you ignored the one good match, their average was closer to 2.

By the Championship, we had a "Summary box where we summed up attributes. Passes and scores and hangs were summed into a metric we called positive actions. This was an wonderful indicator of total offensive capability, and a good initial ranking metric. Talking to a scouting team I admire a lot, they did something similar, but weighted the passing as not every passed ball scored.

*****************
As far as qualitative goes, we use it for judgement metrics. A qualitative how fast 1 2 3 4 5 tells me a lot more than they can go 12 FPS in high gear, but spend the entire match in low and look slow.

*****************
Remember, the most important part of scouting is having actionable data. Data your comp team can use in their matches, and data you can use to make a good pick list.
Ask yourself, what is important for your team to know strategically about partners and opponents?
What does your team need to know to make a good pick list?

Pit Scouting:
What ideas is your team looking for to make your program better?
Instead of how many wheels, how about:
How do you tension your chains?
How do you hold position with mechanism X?
How do you control the motion of your arm?
How do you clutch your gripper?
How did you decide to use that particular material?
I would love for 1 weekend if my pit-crew didn't have to answer how many wheels the robot has. See, they are right there. They are the round things touching the ground...
***********************
This system picked 23/24 teams in elims at the MSC last year. Your mileage may vary.