Hi all - I am a veteran member on team 88, TJ^2. Over the past few years our strategy group has been trying to ease away from paper scouting and try to develop a database / app to use offline during competitions.
Can any current members, alumni members, or mentors who work with their strategy / scouting group point me in the direction of what is best to use for collecting, transferring, and averaging certain types of data?
Also, I am wondering what other teams are currently using for scouting instead of filling in a paper with tally marks and notes. What are the pros and cons to the scouting system that your team uses?
For the 2016 / 17 FRC season, our team is hoping to be able to use a database that can be transferred into a spreadsheet that the strategy captain can easily access on their phone and / or a tablet.
So far I have looked into using Microsoft access to collect data, and transfer the data into an excel spreadsheet. I’ve also looked at various IOS & Android apps that I have used with data from this years competitions.
Our team made our own scouting app a few years ago and it has definitely worked for us! I’m not entirely sure about all the details on how the app was made (I can definitely get information for you if needed), but it doesn’t require WiFi which is usually scarce at robotics competitions. Instead, it transfers data to a master tablet using Bluetooth.
Some pros of our method, besides not needing wifi, is the fact that it compiles the data onto an excel type sheet so when we have strategy meetings at night before alliance selection, you can look at a variety of data including averages and raw data. Important data has been less likely to get lost when we use electronic scouting and is much easier to decipher.
Some problems that teams might run into is being able to get the tablets to take the data with. You’ll probably need at least 6 tablets for scouting if you are planning on scouting all robots, and then the master tablet. If you do pit scouting, you may need more for that as well.
If you would like to check out the app that we use as a reference, you can download it on the Google Play Store. It’s called FRCKrawler and here is a link with more information on it: http://www.team2052.com/frckrawler/
I hope this information helped at least a little bit!
My team (FRC 2485) used a HTML web form that put data directly onto a spreadsheet on Google Drive. Data is weighted such that more recent matches count more towards a team’s average stats. We have some documentation and a download at the following link, as well as instructions on how to set it up for yourself.
Pros – Multiple people are able to view/edit the data simultaneously, meaning it is quicker to create and allows every person at the Friday night meeting to look up teams on their own. It also allows someone on the drive team to have access to the data at all times, updated pretty quickly after scout submissions.
Google sheets also allows you to interface easily with other things on the web. For example, in order to see what defenses were up each match, we used The Blue Alliance’s API.
The HTML and JS of the web form is easily editable on the fly, just copy-pasting and changing names to add more fields. The interface itself is also very extensible.
Cons – A big con to this method is connecting to the internet. Cell phone data usage is a big deal over the course of a competition.
Another con is that Google limits the speed/size of each sheet. For this reason, our system this year needed to span several spreadsheets, all importing data from others, to prevent lag on the spreadsheet that looks up data caused by the spreadsheet that calculates it. This made the system pretty complex, but still manageable.
For the 2016-2017 season, our team is migrating to a Bluetooth-based system to collect data from scouts and send it to one person’s computer, who will then send the data to Google Drive so that only one person needs a hotspot. This alleviates our problem of data usage. I’ll make another reply here if that pans out.
Here on 2386 we had just recently moved to electronic scouting for the 2016 season. For the years before that, scouting was normally an afterthough and carried out by only a few students and mentors. Our system this year was comprised of a custom app writen by one of our senior students an an excel spreadsheet used to analyze the data.
We would input the data collected for each robot onto one of 6 android enabled devices in the stands. Such things included how many shots on each goal vs how many successful shots on goal, completed scales, what defences were crossed and (x) amount of times etc. From here we would transfer the data from 5 of the devices onto the 6th (the one we deemed the master device) via bluetooth. After this we would export a .CSV file (a file readable by excel) from the master device with all of the raw data collected.
This is where things got a little labour intensive and where we plan to improve next year. After we had the raw data that was collected during the day, I had to manually sift through the data and move all of the data pertaining to each team into their associated worksheet on our that primary spreadsheet. This can be improved upon by writing macros in excel to do this for us. From here we could assign different weights to each “catagory” as we deemed fit. These catagories were things such as number of high goals, number of low goals, number of crossings and number of successful scales.
Finally on our master list of all the teams, a score was calaculated based on the raw data collected during the day, and the weights assigned by our scouting team to help form our picklist. As this was our first year implementing the system, we still spent time manually checking our findings in the spreadsheet, however as time goes on we hope that we can rely more and more on this system to save us time while creating our pick list.
Attached is our primary spreadsheet which includes all of our raw data from Windsor.
Eagle Imperium is also trying to move away from paper scouting, and we’re in progress with our system right now. The plan is to have an Android app that scouts can run on their phones that records data into .xls files on each phone. Whenever we want the data, we will plug all 6 phones into a master laptop with a USB hub, and then pull the data using ADB. Once the data is on the laptop, it can be combined into one big database in Excel, and we can analyze it.
This system has a few advantages. No internet access of any kind is required. It runs on phones and laptops people already had, so no need to buy a bunch of tablets. Also, we have already used Excel for analysis in the past, so the new system will fit in well with that.
Let me know if you have questions about our scouting system (paper or electronic). I’ll definitely be posting about it when it’s finished.
Team 107 used the SuperScouter app this past season at all of our districts, State Championship and World Championship and it worked very well. You are able to customize your own interface for collecting data on Andriod and iOS. After collecting the data we send the match data from each phone via QR code to a laptop in the stands. The data is put into a CSV file which we can open up in Excel. We then sort through the data using tables and graphs in Excel and also using Tableau. More information about SuperScouter can be found on this thread.
Hey Team 88,
I don’t know if you’ve seen our scouting process at any of the New England event but we didn’t do much real scouting up till we went to districts. The first year we made a concerted effort was 2014. We had 2 3" binders and had several teams joke about it. Last year at champs we were not picked and one of our local teams, 2836 BETA, had no scouting data and needed help. We meet with them for 2+ hours on Friday night and came up with a plan for them. As the morning progressed we revised our selections for them right up to the last match based upon current ranking and anticipated picks. They to be assured our info was correct and had two other teams that had “electronic” data sit in and they found our information was not only the same as the other two teams but we had more of it and valid reason to pick or not to pick certain teams.
RAW data and the process of compiling the numbers and using just that information does not always give you the best results. We have two students that want to develop an app to use for scouting. Our scouting captain and the drivers don’t want to give up all the valuable notes on each and every scouting sheet. It is also way easier to spot a trend or an anomaly looking at the paper sheets. The electronic versions can be prone to errors and you can’t easily filter out one badly scouted match. It becomes obvious when looking at each paper. One prominent NE team that uses electronic scouting has our team as scoring 2 high goals at the Hartford district event this year and we no longer had a high goal shooter. I asked about it and was told it happens sometimes. They had no way to back track and find out where that erroneous data came from.
We are planning on doing a scouting seminar at CT University day this year at CCSU. It will be based on our scouting experience and how well we have been able to use it. Many CT teams have told us we have some of the best scouting information in our area. It is surprising how many teams either do not scout or don’t take it seriously. It is one of the most important things the team can do to support the drivers and strategy team to be successful. My scouting captain was asked for contact information last year as CCSU and no one from your team ever contacted her. We would be more than willing to help you refine your process to make it more efficient.
I’ll go the direction of other people in the thread and promote 2338’s scouting app, GearScout. Our key features include:
-availability on Android and iPhone platforms
-access to our Amazon DynamoDB database, meaning you can do live updates between the pits and stands without anyone running around with a flashdrive
-flexibility in configurations, allowing you to take quantitative or qualitative data, and save your data in a variety of methods.
Many teams have found success with our app over the last two years; GearScout users have taken our app to numerous regional wins and two championship division wins. If you’re interested, PM or email me and I’d love to get you guys up and running with our programmers and learn how the app works.
At WPI, my team used a paper scouting system that consisted of a double-sided pit scouting and match scouting sheet for each team at the event. I got negative feedback from the match scouters about it being hard to organize.
At Boston, a mentor and his daughter had worked on a web app that allowed for pit and match scouting. It included a defense selector based on pit scouting data, as well as rankings based on statistics such as high and low goal shot and defense crossings. It was very bare bones visually and received many updates over the weekend, but those were mostly due to starting late on the project, and we hope to have an app that is better put together for 2017.
If you private message me, I can give you the link and login credentials so you can see for yourself. If we run into you guys at an event next season, I’m sure my team would be more than happy to work with you, as our main issues have had to do with actually having enough people actively scouting to put our systems to use. I will also be speaking with the team members about the specific makings of the app.
I’d check out 1678’s white paper. Their system is pretty impressive and they’ve been iterating every year and it has gotten them to Einstein repeatedly.
The 2016 version is much improved too, and most importantly the documentation. We’d love to help any team that wants to implement this type of system. Thanks much for the kudos!
we have a google form we fill for each robot each match and we move the data we collect to a microsoft access system where we can view each team and its performances, for example: every robot that shoot to the high goal, or what defenses each robot has crossed during the matches.