Team 27's Advanced Scouting 2014

[Josh Fritsch]

Team 27 wanted to share our scouting system for 2014 and show what’s possible with some of the new resources available to teams.

For this year, we are relying heavily on the use of Tableau, data visualization software available free for FRC teams as the end product. However, we have put a lot of work in this year on the front end to make the experience as easy on our scouts as possible while collecting all of the information we feel we need.

iOS App:
We've developed a custom iOS app to gather our match data in the stands. Using 6 scouts with 6 iOS devices they can enter the match data quickly and efficiently. The app then locally stores a CSV file with the data for each match recorded on the device.

From there we transfer all of the individual CSV files to a laptop and have a master Excel sheet. Unlike many excel based databases, this one is very simple. A single sheet with nothing but a button to run a macro which imports all of the individual CSV data files.

Alternatively, this process can be done much simpler for any teams interested by using a paper scouting sheet that we’ve developed that mimics the iOS app, and simply enter the data manually into the single excel sheet. I am not sure if we have the ability to upload the app to the App Store, but if we do I'll let everyone know here so they can download it.

Tableau:
This is where it really gets exciting. Using the data collected we have been able to use Tableau to manipulate and view the data in many creative ways, allowing us to have any and all information available to us that we need.

For example, our “Match Strategy Sheet” tells us general information that we would expect to see such as average points in teleop or auton. But we also are able to see overall auton shot percentages and even detailed shot percentage and averages based on starting location. In addition we have incorporated out pit scouting data such as drive train type which is displayed as well. We will use this sheet throughout the day for our drive team to develop strategies for upcoming matches.

For alliance selection purposes we use a different workbook to quickly review and analyze data. We can see how a team is improving throughout the tournament, detailed information based on starting location, defensive metrics and scoring stats. For every robot we can see their Max, Average, and Minimum scores for auton, teleop and total contribution.

Another metric that we will be using is our defensive rating, our only subjective data, can give us an important relative comparison on a team’s defensive ability. A component to this is that Tableau has the ability to filter different elements, such as our pit scouting drive train information. We have the ability to automatically sort and rank robots by the type of drive train if we wanted.

Anyway, we wanted to share what we will using this year and show how to stretch the capabilities of the software available to us.

Feel free to see for yourself at the below links to each of our workbooks. You can play around with the sample data that we have and see what is possible. If you have specific questions, feel free to post below and I can go into more detail on how to set something like this up yourself and provide our paper scouting sheet, Excel table, and give some tips on Tableau if you like.

Match Strategy Sheet:
http://public.tableausoftware.com/vi...splay_count=no

Strategy and Selection Workbook:
http://public.tableausoftware.com/vi...splay_count=no

Detailed Individual Team Data Workbook
http://public.tableausoftware.com/vi...splay_count=no

Screenshots:

[EricLeifermann]

Originally Posted by Josh Fritsch

Team 27 wanted to share our scouting system for 2014 and show what’s possible with some of the new resources available to teams.

For this year, we are relying heavily on the use of Tableau, data visualization software available free for FRC teams as the end product. However, we have put a lot of work in this year on the front end to make the experience as easy on our scouts as possible while collecting all of the information we feel we need.

iOS App:
We've developed a custom iOS app to gather our match data in the stands. Using 6 scouts with 6 iOS devices they can enter the match data quickly and efficiently. The app then locally stores a CSV file with the data for each match recorded on the device.

From there we transfer all of the individual CSV files to a laptop and have a master Excel sheet. Unlike many excel based databases, this one is very simple. A single sheet with nothing but a button to run a macro which imports all of the individual CSV data files.

Alternatively, this process can be done much simpler for any teams interested by using a paper scouting sheet that we’ve developed that mimics the iOS app, and simply enter the data manually into the single excel sheet. I am not sure if we have the ability to upload the app to the App Store, but if we do I'll let everyone know here so they can download it.

Tableau:
This is where it really gets exciting. Using the data collected we have been able to use Tableau to manipulate and view the data in many creative ways, allowing us to have any and all information available to us that we need.

For example, our “Match Strategy Sheet” tells us general information that we would expect to see such as average points in teleop or auton. But we also are able to see overall auton shot percentages and even detailed shot percentage and averages based on starting location. In addition we have incorporated out pit scouting data such as drive train type which is displayed as well. We will use this sheet throughout the day for our drive team to develop strategies for upcoming matches.

For alliance selection purposes we use a different workbook to quickly review and analyze data. We can see how a team is improving throughout the tournament, detailed information based on starting location, defensive metrics and scoring stats. For every robot we can see their Max, Average, and Minimum scores for auton, teleop and total contribution.

Another metric that we will be using is our defensive rating, our only subjective data, can give us an important relative comparison on a team’s defensive ability. A component to this is that Tableau has the ability to filter different elements, such as our pit scouting drive train information. We have the ability to automatically sort and rank robots by the type of drive train if we wanted.

Anyway, we wanted to share what we will using this year and show how to stretch the capabilities of the software available to us.

Feel free to see for yourself at the below links to each of our workbooks. You can play around with the sample data that we have and see what is possible. If you have specific questions, feel free to post below and I can go into more detail on how to set something like this up yourself and provide our paper scouting sheet, Excel table, and give some tips on Tableau if you like.

Match Strategy Sheet:
http://public.tableausoftware.com/vi...splay_count=no

Strategy and Selection Workbook:
http://public.tableausoftware.com/vi...splay_count=no

Screenshots:
Attachment 16395Attachment 16396Attachment 16397Attachment 16398

This is awesome!

The goal for our team was to do something very similar only with android, but life got in the way and I, nor any students had enough time to learn enough to get this done.

I'd love to have the opportunity to check this out next week in Duluth, and talk about everything you did.

[tim-tim]

This is great!

Something you may want to add or track are penalties. This is something that has bit us in the past and ever since we track it. Sometimes it's obvious, sometimes they are missed. Usually a well run event will announce the big penalties after the match.

[Pranit T]

I am absolutely impressed with the detailed yet clean output of data. The amount of information collected and displayed for the individual teams is to the point. (Although, how heavily are you relying on the subjective defensive/driver rating that each scout would put down?)

My personal favourite is where you output the trend-line to see how the team has been performing as they tournament goes on; makes life so much easier to disregard the outliers rather than go looking for them in a basic text based database.

(And Tim beat me to it, but tracking penalties, specially Tech Fouls, could come in really handy as you go later into the tournament weekend.)

I'd be interested to view your paper scouting sheet, and the Excel database as well, if you dont mind sharing it.

Brilliant stuff with these apps, once again!

[RamZ]

Not to be that guy, but could you possibly post a screenshot of your excel sheet and/or your paper scouting sheet. I'm having a bit of trouble wrapping my head around how I can make scouting this year anything more than manually inputting data, especially with having to both program the bot and scout.

Thanks

[Josh Fritsch]

Originally Posted by RamZ

Not to be that guy, but could you possibly post a screenshot of your excel sheet and/or your paper scouting sheet. I'm having a bit of trouble wrapping my head around how I can make scouting this year anything more than manually inputting data, especially with having to both program the bot and scout.

Thanks

Ill do one better. I've attached our paper scouting sheet as well as the Excel sheet that would be used as the input method based on the paper sheet.

Each paper sheet can be printed, 2 per page.

This Excel sheet is our actual sample data file and is all that is used to generate all of the Tableau data you see.

Thanks for all the great comments so far, we are excited to try it out in Duluth! If you happen to be there stop by and I can show you anything you want in more detail.

2014 Paper Scouting Sheet.xlsx 2014 Match Results Database_v1.xlsx

[Pranit T]

Thanks Josh. It's looking good!

Wish you the best for your competitions.

[RamZ]

Originally Posted by Josh Fritsch

Ill do one better. I've attached our paper scouting sheet as well as the Excel sheet that would be used as the input method based on the paper sheet.

Each paper sheet can be printed, 2 per page.

This Excel sheet is our actual sample data file and is all that is used to generate all of the Tableau data you see.

Thanks for all the great comments so far, we are excited to try it out in Duluth! If you happen to be there stop by and I can show you anything you want in more detail.

Attachment 16403 Attachment 16404 Attachment 16405

Thanks so much! I'll definitely take these into account when making my team's scouting forms!
Good luck at the competitions!