Best Ways to Scout

[Michael_Rood]

As a first year scouting captain, I have had trouble finding a reliable way to scout. In the years before I became scouting captain, my team has been using paper scouting. This year we made the switch over to electronic scouting. We used Google Forms to create a scouting sheet that wee could compile into an Excel sheet. The one problem is, it requires a reliable wifi connection to work. I was wondering if there were other ways to scout electronically without having to rely on wifi. I have been talking with one of my mentors and have come up with some ideas. 1. Having a portable server that we can wire the phones into (would require making an app that doesn't need wifi). 2. Having the phones connected to a laptop via Bluetooth. If someone could help me build an app (in LabVIEW preferably) that could connect to a server (would also need to find a server that would work with the app) I would greatly ly appreciate it.

Thank You

[BenjiSG]

We made the decision this year to just stick with paper scouting. Electronic scouting is nifty and all, but the inconveniences of trying to get reliable WiFi and access to power are too much to make it actually worth it for us. Plus, lost paper is far preferable to lost electronics, and we don't see a great deal to gain from going digital.



Frankly, just printing out excel sheets, allowing them to be filled in by hand, and compiling all useful information digitally after Day 1 proves to be far easier and seemingly just as effective.



If you want to move to electronic scouting, by all means, but don't forget that the paper strategy remains not only an option, but in some ways a more preferable one.

[New Lightning]

Try looking through the myriad of threads on the issue in the scouting forum. There are a lot of really good ideas and tips on this issue.

Just click on the forums thread in the orange bar on top. Then go to the scouting sub-forum. Its all there for you.

[Boltman]

Best way to scout is "eyes on bots", excel sheets , pen and paper note taking. I use a cheap thin spiral binder and highlighters . My awesome scouts also take copious notes. I started the scouting department from scratch last year. Scouting is equal to engineering and drive team IMO and the entire team agrees with me they love it. Not only for winning games but building alliances to have the best chance at a win.

A field of 60 with 10-12 games is not indicative of a high level of need for data analysis and is highly susceptible to schedule pairings more than raw data from stats. Eyes first, notes second and data on first night to see if there are any you missed. Usually not the case...might be one or two but then you watch on day 2 and go "hmm wonder how they got that stat?" eyes don't lie like stats do.

I don't get alliance captains that don't already know their second picks instantly of a pick list, other than they must not scout at all and ultimately they exit in QF.

[Daws00]

This year we are using an html form we created on a local network.

We have a local server that everyone connects to with Ethernet cords. All the data gets pushed to an SQL database on that server. It looks like a mess, but it works really well.

[GreyingJay]

Originally Posted by Boltman

Eyes first, notes second and data on first night to see if there are any you missed. Usually not the case...might be one or two but then you watch on day 2 and go "hmm wonder how they got that stat?" eyes don't lie like stats do.

I didn't see this (so much for "eyes first") but I heard from several people that our robot accidentally scored a high goal once during quals. We're not even really high goal capable but apparently during one match we were driving forward and somehow a boulder got launched at just the right angle that it arced into the goal. Total fluke. I'm sure a bunch of teams were looking over their scouting data later going "wait, what? That's not right...!"

[sclancy1647]

Our team creates a scouting application to use on android tablets. This way we can write our own analysis program, to find what information we need about any team.

[logank013]

What devices are you using to scout?

[alextried]

If you want to maintain the electronic aspect of your scouting and have the phones/tablets to do it, I suggest SuperScouter. It's schema based, so it's customizable and I found the original thread to be very helpful.

http://www.chiefdelphi.com/forums/sh...t=superscouter

While I couldn't get the QR scanner to work, I'm planning on using micro-USB to Ethernet and lightning cable to Ethernet to sync everyone's phones to my laptop and then inputting that data into Excel. Then I input the data I want into Tableau, which is a visualization software free to FRC teams!

http://www.tableau.com/first-robotics

[bennettj800]

I have written my own database and hope it helps anyone looking for one.

i think a hybrid between paper and PC is the best way to go, with a projector for the meeting that night so everyone attending can see what the hay hay is going on.

[|HERE|] is that database
and if that ↑ didnt embed correctly, the link is here:
https://drive.google.com/open?id=0B4...FpFTC04d2doYjg

I STRONGLY suggest using this database (youll appreciate all the formatting)

I aso STRONGLY suggest using the sheets that come with it. If you do not you will lose more information than you think, because not all will transfer into the database, and not all of it will be available for the database... that being said, USE THESE PLEASE

Other than that, I just really dont wanna waste this exvel spreadsheet... its really nice... i hope you use it... I really really hope...

[XaulZan11]

Originally Posted by Boltman

Best way to scout is "eyes on bots", excel sheets , pen and paper note taking. I use a cheap thin spiral binder and highlighters . My awesome scouts also take copious notes. I started the scouting department from scratch last year. Scouting is equal to engineering and drive team IMO and the entire team agrees with me they love it. Not only for winning games but building alliances to have the best chance at a win.

A field of 60 with 10-12 games is not indicative of a high level of need for data analysis and is highly susceptible to schedule pairings more than raw data from stats. Eyes first, notes second and data on first night to see if there are any you missed. Usually not the case...might be one or two but then you watch on day 2 and go "hmm wonder how they got that stat?" eyes don't lie like stats do.

I don't get alliance captains that don't already know their second picks instantly of a pick list, other than they must not scout at all and ultimately they exit in QF.

I could agree with the "eyes over stats" idea IF you have experienced scouts* that know what to look for and IF you have enough experienced scouts to get eyes on a majority of matches for every team. You realistically need at least four experienced scouts to literally watch 90% of matches. Since most FIRST teams do not have these resources, putting together a stats-based system (where any freshman can contribute) is best for most teams.

* By experienced scout, I mean a scout that can remember basic information about nearly all teams and put teams in tiers or rough picklist. In the roughly 10 years I've scouted, there are probably 3-5 students who I'd consider capable of doing this for an entire 60+ team event.

[CBG]

My team uses an electronic reporting form, but we use phones and tablets that connect directly to the internet. We also value paper scouting. One of our mentors always takes paper notes on each robot's performance across one or two matches and we rearrange these cards when making our pick list Friday night. I also want to point out the advantages of making a pick list of Friday--it may seem smart not to make these kinds of decisions until just before alliance selections, but you will be pressed for time at this point and it's best to have an idea of the order of teams that you can rearrange when teams either increase or decrease their level of performance.

[trumpthero786]

STORM Robotics also has created an Android application! We are particularly proud of this!

More info and the link to the Play Store is in this following thread.

http://www.chiefdelphi.com/forums/sh...hreadid=142649

It is easy to use, very thorough, and completely offline, utilizing QR codes and outputting .csv files which can be opened in excel for easy management and analysis. There is a User app for data entry, and then a Master app for collection and creation of the CSV files. I hope you try it out! We, along with several other teams, have had great success using the app at our respective competitions.

[Roboshant]

You can always use google sheets offline on devices. Simply set up an excel form on google sheets, and people can edit the excel on their device. To compile simply connect the phones to a computer (android only) or go outside at the end of the day and send it with 4g. I have some google sheets forms handy if you want them.

[Citrus Dad]

Originally Posted by Boltman

Best way to scout is "eyes on bots", excel sheets , pen and paper note taking. I use a cheap thin spiral binder and highlighters . My awesome scouts also take copious notes. I started the scouting department from scratch last year. Scouting is equal to engineering and drive team IMO and the entire team agrees with me they love it. Not only for winning games but building alliances to have the best chance at a win.

A field of 60 with 10-12 games is not indicative of a high level of need for data analysis and is highly susceptible to schedule pairings more than raw data from stats. Eyes first, notes second and data on first night to see if there are any you missed. Usually not the case...might be one or two but then you watch on day 2 and go "hmm wonder how they got that stat?" eyes don't lie like stats do.

I don't get alliance captains that don't already know their second picks instantly of a pick list, other than they must not scout at all and ultimately they exit in QF.

We've successfully set up a scouting system that can use relatively inexperienced students to keep quantitative counts. We have to work to find the other 2 superscouts who make qualitative judgements. Comparing our system to what we see at regionals and even Champs, we believe having a more rigorous system if more fruitful. For example, we were able to get our 10th overall pick (even accounting for alliance captains) at Sacramento as our 2nd pick. 971 and us use a similar rigorous system and our draft lists agreed quite well. (5274 should thank 971 for jumping them up a spot on the list.)

We know who we DON'T want by the end of the second day, but our mid-field list where our 2nd picks reside is much more cluttered and subject to revision. (We've gotten by the QFs every competition since 2013... )