scouting 2011

[h1n1is4pigs]

hey, i'm trying to figure out what important information i should include in a scouting form for this years game. what would you guys recommend putting into this form and how would you recommend weighting the information?

thanks in advance!

[Chris is me]

What are you looking for in an alliance partner?

What would you want to know about your opponents before a match against them?

[MagiChau]

You have to think of which robot is most compatible with the possible strategies for you. I would look out for defense ability, ability to control the field, tube scoring ability, speed in traveling the field, and minibot deployment speed.

[Colin P]

It'll be hard, but try to calculate average return time. I.E. how long it takes the team to grab a tube, score it, and come back for an offensive robot.

It's hard to formulate hard data for defensive bots, because it's only in reaction to offensive plays. Try to teach your scouts how to discern between good play and poor play, and have them keep note of it. Hard data doesn't always work.

Number of tubes scored, minibot wins/ minibot speed, breaks, penalties are all going to be important this year.

When designing a scouting sheet, just imagine a game being played in your head. You'll start thinking of things you wouldn't have thought about if you were just looking at the manual.

[xSAWxBLADEx]

this is cause we have an amazing head scouter (me) but we are only going to have

-robot start (outer,middle)
-auto score (yes or no)
-pegs ("X" for tube "O" for uber)
-minibot (1st,2nd,3rd,4th)

[MagiChau]

Originally Posted by xSAWxBLADEx

this is cause we have an amazing head scouter (me) but we are only going to have

-robot start (outer,middle)
-auto score (yes or no)
-pegs ("X" for tube "O" for uber)
-minibot (1st,2nd,3rd,4th)

I believe there needs to be written or typed, in words, a description of the autonomous mode. The autonomous program(s) could work well with your team or against it. For example if they can score an uber-tube on the column right by the lane (has to go diagonal) I would be impressed more than by scoring by going straight. So I would weight it more at the least.

[KC1AJT]

What my team has found in the past is that open response of any form is not good. Especially when you have 6 or more people scouting. My team uses schedules for scouting so we can end up with 20 different scouts. Here's how to do it:
-give options (check boxes are good)
-yes and no is always a plus

Also, on a side note:
I am currently developing a very large scouting system. It will be setup like a kiosk (hopefully) at our pit where other teams can come in, type in a number and print a report with scores, penalties, pictures, and mechanical details. If anyone would like to help out, I am currently looking for multi-user use of an OOBase database, if its possible. I will be releasing this when it is done.

[Waffles]

Usually our team has 6 scouters, one per robot.

This year, we're looking to have an addition of someone watching analyts/feeders. <--Believe it or not, some of them will have crazy tossing abilities. XD

Obviously, there's a couple things you can determine ahead of time with pre-match sheets (aka run around the pits and ask teams what their robot does).

(There's a few cons to PRE-MATCH scouting and then competition scouting<-- it let's you know if teams made any major improvements and you get a better indication of how they work if they say they do something and don't... because when it comes down to it, it's what you see on the field.) We also take photos of every single robot at competition.

Here's what we decided for our scouting database as far as criteria goes (we'll probably share our database in the near future):

Prematch:
-Scouter name
-Match #
-Alliance color / Team #
-Starting position
-Human player position

Mini-bot:
-Attempted deployment
-->Location: (lane, mid)
-Successfully scored
-->Blocked?
-Engagement time w/ tower

Autonomous:
-Starting position
-->Stayed on tracking line?

-Ubertube
-->Attempted?
-->Scored?

Teleoperated:
-Placed a second tube on same peg?

-Create a logo?

-Pick up off floor? feeder? both? another robot?

-Alliance role (for majority of match)
-->Supplier, score, defense? (CHECK ALL THAT APPLY)

-Placement speed
-->Slow, medium, fast

-Pickup speed
-->Slow, medium, fast

-Drive skill
-->Maneuverability
-->Precision

-Feeder skill
-->Throwing distance

Something like that... our students are still working on creating the database!

Hope this helps!

[xSAWxBLADEx]

Originally Posted by Waffles

Prematch:
-Scouter name
-Match #
-Alliance color / Team #
-Starting position
-Human player position

Mini-bot:
-Attempted deployment
-->Location: (lane, mid)
-Successfully scored
-->Blocked?
-Engagement time w/ tower

Autonomous:
-Starting position
-->Stayed on tracking line?

-Ubertube
-->Attempted?
-->Scored?

Teleoperated:
-Placed a second tube on same peg?

-Create a logo?

-Pick up off floor? feeder? both? another robot?

-Alliance role (for majority of match)
-->Supplier, score, defense? (CHECK ALL THAT APPLY)

-Placement speed
-->Slow, medium, fast

-Pickup speed
-->Slow, medium, fast

-Drive skill
-->Maneuverability
-->Precision

-Feeder skill
-->Throwing distance

50% of that is opinion...

[MagiChau]

Originally Posted by xSAWxBLADEx

50% of that is opinion...

I am unsure what you are implying with that statement but I view that list of criteria that Waffles is going with has too much baggage. I see that something like "staying on tracking line" is way too limited to provide enough use for wide scale scouting. Light sensors aren't required to make a good autonomous.

There is way too much things to fill out as that list looks intimidating. I would rather have like a box to draw robot path in autonomous and two scoring grids to show position of tubes scored. This would eliminate stuff like "create a logo" and "scored over a uber-tube"

[xSAWxBLADEx]

Originally Posted by Waffles

-Engagement time w/ tower

-Alliance role (for majority of match)
-->Supplier, score, defense? (CHECK ALL THAT APPLY)

-Placement speed
-->Slow, medium, fast

-Pickup speed
-->Slow, medium, fast

-Drive skill
-->Maneuverability
-->Precision

-Feeder skill
-->Throwing distance

stuff like that is mostly opinion

[MagiChau]

Originally Posted by xSAWxBLADEx

stuff like that is mostly opinion

throwing distance and engagement time looks more like estimations of numbers than opinion. I see the point of how judging pickup and placement speed makes scouting hard since for example, the team I am in attend a Week 1 Event, and no one has any idea what is the difference between slow, medium and fast until plenty of matches have already gone by to judge others on. Perhaps an approximation of the time is needed.

I think driver-skill is plausible somewhat since it should be obvious if a robot has clumsy movements compared to fantastic ability to play defense or move across the field.

I am disputing mainly with your 2 examples of throwing distance and engagement time with tower.

[nitneylion452]

Originally Posted by xSAWxBLADEx

stuff like that is mostly opinion

How are things like distance and speed opinion? I'm sure they have criteria (such as approximate time) for what constitutes slow, medium, and fast.

[MarkoRamius1086]

Originally Posted by xSAWxBLADEx

"Originally Posted by Waffles
-Engagement time w/ tower

-Alliance role (for majority of match)
-->Supplier, score, defense? (CHECK ALL THAT APPLY)

-Placement speed
-->Slow, medium, fast

-Pickup speed
-->Slow, medium, fast

-Drive skill
-->Maneuverability
-->Precision

-Feeder skill
-->Throwing distance "

stuff like that is mostly opinion

Perhaps "opinion" is a harsher word for "Qualitative Observation?" Many of these do have the ability to peer into a situation and see who did the real work, and who did something other than productive.

-Feeder Skill

Unless a Human Player is strong and skilled enough to throw tubes across the feild onto the lower racks (really improbable), middle racks (highly impobable), or high racks (astronomically improbable) there is no way to put a Quantitave Observation on Human Player performance. Perhaps Human Player performance is not quite as important as it was in say... Lunacy... but a good human player who can watch the game and throw the game pieces where and when they are needed is something every strong alliance should have.

Originally Posted by Waffles

-Alliance role (for majority of match)
-->Supplier, score, defense? (CHECK ALL THAT APPLY)

Matching and cooperative Alliance Roles are undoubtedly one of the key factors when trying to create an ideal winning alliance.

Originally Posted by Waffles

-Drive Skill
-Engagement time w/ tower
-Placement speed
-Pickup speed

These four give a solid indication to the fluidity of a teams driving or robot mechanism. Pure numbers and ratios can give a great indicator to who the top teams are, but in most cases it is not as clear cut as that and some serious in-depth research on how a team performs under pressure can make the difference between victory and defeat.

Every team has good days and bad days, matched by good luck and bad luck in their matching, and due to this numbers alone cannot make a valid decision in my eyes. For the record I believe that ratios and other match numbers are just as invaluable (to showing the top teams for alliance pickings) as qualitave info. It takes a combination of qualitative and quantitave observations to truly make sure you are picking the team you want to.

My $0.02
Happy scouting and good luck this year!

[Chris is me]

The important thing to note is to only track stats you're actually going to use. Speed is good to have - but it will be evident in scoring ability.

Here's my list as of now:

• # of Tubes on rack
• # of Tubes Misplaced (put in wrong position for a logo)
• # of Tubes Dropped
• Time on clock when Minibot scores
• # of Tubes Passed

What else would you need that's not subjective or obvious?