Go to Post Watch out world, FIRST is going to become mainstream! - Church [more]
Home
Go Back   Chief Delphi > Competition > Rules/Strategy > Scouting
CD-Media   CD-Spy  
portal register members calendar search Today's Posts Mark Forums Read FAQ rules

 
Reply
Thread Tools Rate Thread Display Modes
  #1   Spotlight this post!  
Unread 05-02-2017, 20:47
RyanSchwartz's Avatar
RyanSchwartz RyanSchwartz is offline
Registered User
FRC #4611 (Ozone Robotics)
Team Role: Electrical
 
Join Date: Jan 2017
Rookie Year: 2014
Location: Lewis Center, Ohio
Posts: 14
RyanSchwartz is an unknown quantity at this point
Key Scouting Points?

This year I've been the one to spearhead the dive into the scouting programs, with a very different objective than in the past. I noticed our past leaders have seemed to ask the question of "What data can we collect?" and figure out what to make of the data after it's collected. This year, we're asking "What information can we give to our drivers to perform better?".

We've separated this into two sections: Game-to-game strategy and Top-8 strategy. Our data in the past has been oriented towards which teams we should pick if we reach the top 8 (Top-8 strategy), but now we're trying to include the strategies that change our game play from round to round (game-to-game strategy).

My questions for the game-to-game strategy are as follows:
  • Should we be focusing more on our opponents or teammates when developing our strategy?
  • What other things could we focus on other than scoring as many points as possible?
  • What data points (other than the amount of points a team scores) help your drivers perform as a team player and formidable defensive opponent?
  • Do you think game-to-game strategies are even worth it? Will other teams listen to our strategies say "this is what we think is best"?

Thanks for your time and I look forward to your input.
Reply With Quote
  #2   Spotlight this post!  
Unread 05-02-2017, 21:10
CalTran's Avatar
CalTran CalTran is online now
MST Rolla Senior
FRC #2410 (BV CAPS Metal Mustang Robotics)
Team Role: College Student
 
Join Date: Apr 2010
Rookie Year: 2010
Location: Overland Park, Kansas
Posts: 2,438
CalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond repute
Re: Key Scouting Points?

Quote:
Originally Posted by RyanSchwartz View Post
My questions for the game-to-game strategy are as follows:
Should we be focusing more on our opponents or teammates when developing our strategy?
Both honestly - it depends on what job you plan on playing in the match. If you're going with an offensive strategy, you should probably focus on how to best work with your teammates to maximize your score. If you're going with a defensive strategy, I'd largely ignore what your teammates are doing (but still be ready to help run a gear or two if necessary, or know when to drop playing defense all together to help close the gap in scoring.

Quote:
What other things could we focus on other than scoring as many points as possible?
When running a cycle, can you afford to also play hit and run defense? Does it look like an opponents robot is harassing a high volume shooter on your alliance, and can you assist them (block the defender) so they can score?

Quote:
What data points (other than the amount of points a team scores) help your drivers perform as a team player and formidable defensive opponent?
Scoring position - Do they appear to only be able to shoot from a single spot? Can I delay, or even prevent, them from getting to their spot?
Cycle Time - How long is their cycle? Does adding a few seconds to it make a difference? This year it might be the difference between starting the 4th rotor or not.
Coach - Your coach needs to have match awareness. Score call outs, strategy tweaks on the fly, that sort of thing.

Quote:
Do you think game-to-game strategies are even worth it? Will other teams listen to our strategies say "this is what we think is best"?
Game-to-game strategies tend to be less important during early qualification matches or if you're a strictly offensive team, but goes up as you scout matches because you can start tuning your game against your specific opponents.

I imagine other teams would listen if you input on match strategy - nothing gets you on a DNP list like being uncooperative.
__________________
Team 2410 thinks KISSing is amazing! Keep It Super Safe!
  • "You know you've been in robotics too long when you start talking to your tools." "Well, you've been in robotics CLEARLY too long when they start talking back"
  • Theory is when you know everything but nothing works. Practice is when everything works but you don't know why. On our team, theory and practice comes together - nothing works and nobody knows why.
MMR 2410 Student (2010 - 2013) | MMR 2410 Mentor (2013 - Present)
FTC Game Announcer / EmCee (2014 - Present) | FRC EmCee (2015 - Present) | FRC Referee (2016) | FTC Referee (2017)
Academic Student (Forever)

Last edited by CalTran : 05-02-2017 at 21:14.
Reply With Quote
  #3   Spotlight this post!  
Unread 05-02-2017, 21:58
RyanSchwartz's Avatar
RyanSchwartz RyanSchwartz is offline
Registered User
FRC #4611 (Ozone Robotics)
Team Role: Electrical
 
Join Date: Jan 2017
Rookie Year: 2014
Location: Lewis Center, Ohio
Posts: 14
RyanSchwartz is an unknown quantity at this point
Re: Key Scouting Points?

CalTran,
Thanks for the input. I agree with much of what you're saying and I think your input on cycle times will be very beneficial to our scouting data.
Reply With Quote
  #4   Spotlight this post!  
Unread 05-02-2017, 22:14
Brian Maher's Avatar
Brian Maher Brian Maher is online now
Questionable Decisionmakers
FRC #2791 (Shaker Robotics), FRC #1257 (Parallel Universe)
Team Role: College Student
 
Join Date: Apr 2014
Rookie Year: 2012
Location: Troy, NY; NJ
Posts: 483
Brian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond reputeBrian Maher has a reputation beyond repute
Re: Key Scouting Points?

I think an important question is, what do you need to know about your partners to formulate a good strategy? This largely depends on your robot. Here are a couple examples:
  • If you are a shooter, data on where other teams shoot is particularly important for writing a picklist or planning a match. You usually want to pick a shooter who shoots somewhere else, if possible.
  • If you have a gear auto that only works for one of the pegs, you want to keep an eye on what pegs other teams use. You don't want to pick a team that needs that same peg as you to pull off their auto. If which peg you use doesn't matter to you, this becomes a much less useful data point.
  • If you don't have a floor pickup for fuel, watch out for teams who have floor pickups. If you don't have one and don't pick a team who does, guess what defenders are going to do in playoffs.
__________________
2016-present, Mentor, FRC 2791 - Shaker Robotics
2016: Tech Valley SF (5236, 2791, 3624) and Quality, Finger Lakes SF (5254, 2791, 2383), Battlecry@WPI Winner (195, 2791, 501), Robot Rumble Winner (2791, 195, 6463)

2016-present, Mentor, FRC 1257 - Parallel Universe
2016: Mount Olive Winner (1257, 5624, 1676), Bridgewater-Raritan Finalist (1257, 25, 3340, 555) and GP, MAR CMP Winner (225, 341, 1257), Archimedes SF (4003, 4564, 5842, 1257), IRI Invite

2012-2015, Student, FRC 1257 - Parallel Universe
2015: Mount Olive QF (1257, 1811, 1923) and Safety Award, North Brunswick Finalist (11, 193, 1257) and Team Spirit and Safety Awards
2014: Clifton Winner (1626, 869, 1257), MAR CMP QF (1257, 293, 303)
2013: TCNJ Safety Award
2012: Mount Olive QF (204, 303, 1257)

Last edited by Brian Maher : 05-02-2017 at 22:19.
Reply With Quote
  #5   Spotlight this post!  
Unread 05-02-2017, 22:17
GCarnes's Avatar
GCarnes GCarnes is offline
Registered User
FRC #5150 (Hybrid Hornets)
Team Role: Alumni
 
Join Date: Mar 2015
Rookie Year: 2014
Location: Michigan
Posts: 24
GCarnes is just really niceGCarnes is just really niceGCarnes is just really niceGCarnes is just really nice
Re: Key Scouting Points?

As a past driver, the most useful information for me during qualifications would have been any quick easy tactics to shut down the next opposing alliance (maybe all of the bots cannot intake balls of the ground, so go dump all the hoppers right away. Stuff like that.). Or ways to show off our own bot to the higher ranking teams if we aren't doing well.

If you're in a position to pick, find a strategy that you want to aim for for your alliance and fill that alliance to the best of your ability. Example: my team's 2016 bot could not shoot high goals, but we could play great defense. We HAD to pick two high goal shooters. Didn't matter if they weren't the highest percentage shooters. We had to have the two best that we could get. So if you know this strategy that you need for playoffs, scout to find those teams.

During playoffs, same thing about the shut down tactics as in quals. This info has to be fast and furious coming from the stands down to the drivers. Drivers don't have much time to analyze the data, so come to a conclusion and tell the drivers that, don't hand them a folder of stat sheets for them to try and scan through.

Feel free to ask questions if my jumbled mess of text wasn't clear enough.

TL;DR Scouters need to know the game; come up with an overall strategy; fill a planned alliance if you're picking; give drivers answers and ideas, not just data.
__________________

Reply With Quote
  #6   Spotlight this post!  
Unread 06-02-2017, 09:54
rick.oliver's Avatar
rick.oliver rick.oliver is offline
Mentor - Retired
AKA: Pap
no team
Team Role: Alumni
 
Join Date: Apr 2006
Rookie Year: 2003
Location: Liberty Township, OH
Posts: 249
rick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond reputerick.oliver has a reputation beyond repute
Re: Key Scouting Points?

I have found that there is no difference in the data you should collect based on whether you are scouting for Alliance Selection or Qualification Matches.

For Qualification matches, you need to know the capabilities (Strengths) and weaknesses of all 6 robots in the match. Yes, you should be scouting yourself.

In Qualification Matches, the strategy is based on the capabilities of partners and the tendencies and capabilities of opponents. You need to be honest about your own capabilities and you need to know your partners' capabilities. (Yes, some teams may exaggerate their capabilities).

For Alliance selection, you need to know what you do well and where you need help, establish a strategy which you believe will give you the best chance of success and select the partners which complement you and complete that strategy.

What you may want to consider is having a set of scouts who are focused on the upcoming partners and opponents who can provide the freshest perspective to the drive team for their match strategy discussion. These folks would be separate from the scouting team collecting the data for later evaluation during alliance pick list development.

Video of matches to confirm or correct scouting data is also helpful, too.
Reply With Quote
  #7   Spotlight this post!  
Unread 06-02-2017, 13:55
RyanSchwartz's Avatar
RyanSchwartz RyanSchwartz is offline
Registered User
FRC #4611 (Ozone Robotics)
Team Role: Electrical
 
Join Date: Jan 2017
Rookie Year: 2014
Location: Lewis Center, Ohio
Posts: 14
RyanSchwartz is an unknown quantity at this point
Re: Key Scouting Points?

Quote:
Originally Posted by rick.oliver View Post
You need to be honest about your own capabilities and you need to know your partners' capabilities. (Yes, some teams may exaggerate their capabilities).
I think one thing we're going to work on with pit scouting is verifying teams can do what they say they can do.

Ex: On our scouting app, we have an option for teams that have climbing mechanisms to say if they successfully climbed or not, but also one to say they didn't have the chance. I'm worried data will be skewed if a team says they can climb (but actually can't b/c part is broken or part is jammed), and scouters mark it as didn't get the chance to climb purely because they have the parts to do it on the bot.

Is this something I should be worried about or is this reading into the situation too much?

Quote:
Originally Posted by rick.oliver View Post
What you may want to consider is having a set of scouts who are focused on the upcoming partners and opponents who can provide the freshest perspective to the drive team for their match strategy discussion. These folks would be separate from the scouting team collecting the data for later evaluation during alliance pick list development.
Our current structure is to sign all attending team members (except drive team, team captains, etc.) up on a sheet - they are responsible for showing up to our scouting area during their assigned match number and scouting the team they are asked to look at. This way we don't have to task 12 unfortunate souls with scouting for 2/3 days straight. I, along with 2 or 3 others, plan on interpreting the data they submit and will formulate a strategy to give to the drive team before their next match.
I also heard someone recommend we have trained scouters that are watching most all of the matches (say it's a team of 24 so 12 rotate out every couple of games). Thoughts on that type of scouting? What does your team do?

Quote:
Originally Posted by rick.oliver View Post
Video of matches to confirm or correct scouting data is also helpful, too.
Has anyone had experience with this? I agree it would be helpful but watching possible alliance partners and our own matches is no short task. Any ideas to optimize this?

TL;DR What do your teams do that make scouting more unique or efficient than the next?
Reply With Quote
  #8   Spotlight this post!  
Unread 06-02-2017, 14:00
CalTran's Avatar
CalTran CalTran is online now
MST Rolla Senior
FRC #2410 (BV CAPS Metal Mustang Robotics)
Team Role: College Student
 
Join Date: Apr 2010
Rookie Year: 2010
Location: Overland Park, Kansas
Posts: 2,438
CalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond reputeCalTran has a reputation beyond repute
Re: Key Scouting Points?

Quote:
Originally Posted by rick.oliver View Post

Video of matches to confirm or correct scouting data is also helpful, too.
Quote:
Originally Posted by RyanSchwartz View Post

Has anyone had experience with this? I agree it would be helpful but watching possible alliance partners and our own matches is no short task. Any ideas to optimize this?
In my experience, you don't watch the match videos during the day - you rewatch them Friday night when you're making your picklist to help break ties (or confirm/correct scouting data if something seems outrageous or just plain off). IE it comes down between two teams that you're deciding upon, so you watch a few matches from both teams, rescout those matches, and then reevaluate both of them against each other.
__________________
Team 2410 thinks KISSing is amazing! Keep It Super Safe!
  • "You know you've been in robotics too long when you start talking to your tools." "Well, you've been in robotics CLEARLY too long when they start talking back"
  • Theory is when you know everything but nothing works. Practice is when everything works but you don't know why. On our team, theory and practice comes together - nothing works and nobody knows why.
MMR 2410 Student (2010 - 2013) | MMR 2410 Mentor (2013 - Present)
FTC Game Announcer / EmCee (2014 - Present) | FRC EmCee (2015 - Present) | FRC Referee (2016) | FTC Referee (2017)
Academic Student (Forever)
Reply With Quote
  #9   Spotlight this post!  
Unread 06-02-2017, 14:21
Jarren Harkema's Avatar
Jarren Harkema Jarren Harkema is offline
Dancing Drive Coach
FRC #4967 (That ONE Team)
Team Role: Mentor
 
Join Date: Sep 2013
Rookie Year: 2013
Location: Michigan
Posts: 171
Jarren Harkema is a name known to allJarren Harkema is a name known to allJarren Harkema is a name known to allJarren Harkema is a name known to allJarren Harkema is a name known to allJarren Harkema is a name known to all
Re: Key Scouting Points?

We are trying a lot of new and ambitious things with scouting this year, moving from a paper/digital hybrid to fully digital is going to enable us to pull some of our match scouting data for qualification strategy discussions.

The biggest data points we are looking for are pretty much the same for our partners and our opponents.

Where do robots prefer to shoot fuel from? Does that encroach upon our firing position? Is the firing position of an opponent robot predictable and susceptible to defense?

How many gears can our entire alliance place on average and/or max? How many gears can our opponents place on average and/or max? This info might tell us which number rotor to shoot for. There is no sense in going for the 4th rotor unless the data says the alliance can do it. Similarly, we can make a prediction of how far the opposing alliance might try to go, and make the decision if its worth playing defense or not on gear placement.

From past experience as drive coach and strategist, unless you have hard data to backup an alliance strategy, the chance that the match is going to go the way you planned it is very low.
__________________

Reply With Quote
  #10   Spotlight this post!  
Unread 06-02-2017, 17:59
RyanSchwartz's Avatar
RyanSchwartz RyanSchwartz is offline
Registered User
FRC #4611 (Ozone Robotics)
Team Role: Electrical
 
Join Date: Jan 2017
Rookie Year: 2014
Location: Lewis Center, Ohio
Posts: 14
RyanSchwartz is an unknown quantity at this point
Re: Key Scouting Points?

Quote:
Originally Posted by CalTran View Post
In my experience, you don't watch the match videos during the day - you rewatch them Friday night when you're making your picklist to help break ties (or confirm/correct scouting data if something seems outrageous or just plain off). IE it comes down between two teams that you're deciding upon, so you watch a few matches from both teams, rescout those matches, and then reevaluate both of them against each other.
I love the idea, however I guess I'm still wondering how the logistics of recording and storing works out. Did you have a "camera man" so to speak that stuck with the camera and clicked start/stop record at the beginning/end of every match? Did you record them to a micro SD card or did you stream it to a computer/external hard drive? How did you guys manage the recording of the matches?
Reply With Quote
Reply


Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is Off
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 13:57.

The Chief Delphi Forums are sponsored by Innovation First International, Inc.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi