How does your team scout?

[Xaviermw191]

nice job at the flr 191 too bad some of your alliances robots didn't work the last match.

Thanks, it definitely wasnt just them who broke, in fact 229 had to basically do all the top row hanging themself that last match. One of our robot's elevator motors smoked out so we wern't able to raise the elevator up, so we could only get to the bottom row that match.

[Koko Ed]

Originally Posted by Xaviermw191

Thanks, it definitely wasnt just them who broke, in fact 229 had to basically do all the top row hanging themself that last match. One of our robot's elevator motors smoked out so we wern't able to raise the elevator up, so we could only get to the bottom row that match.

Gotta love the Banebots (or not).

[johnmaguire2013]

We have 6 iPods which are set to view a page which I designed for viewing on mobile devices. Six players watch each match, and record data into the form, which looks like this:

http://k.min.us/ilKsrI.png

The data is inserted into a MySQL database, and later retrieved into an XLS file (using some XLS functions I found online and converted to a class,) in the correct format to fit this: http://www.chiefdelphi.com/media/papers/2450?

We copy and paste the data and the analyzer does the rest.

[AcesJames]

On 176, we do the following -

We have 1 student per robot, and 1 student per HP, in the years that they have a hand in scoring.

On Friday, 12 students fill out sheets for 6 robots and 6 HPs, each match.
Friday night, the scouts organize the sheets by team. We usually have about 7 sheets per team, 1 for each round they were in.
We look at the top 24 teams in rank on the FIRST site, and we disregard ourselves.
We re-rank the teams based on how good they are, according to our scouters.
We then have a list of 24 top teams, in order from best to worst.
We then try to run through mock alliance picks, and form alliances from the 24 teams we've selected.
We then finally look at where our team sits between that, and whether or not it's likely we'll be picking.
If we find it's likely we'll be picking, we look for robots around us according to our rank list, and then form an alliance.
We create a list of a few possible first picks, and a few possible second picks.
The robots we choose are custom-fit to the strategy we have established, and they are listed best to worst in order of how well they play/compliment us.

[XaulZan11]

Originally Posted by AcesJames

We look at the top 24 teams in rank on the FIRST site, and we disregard ourselves.

Do you not look at teams ranked lower than 24?

[AcesJames]

Originally Posted by XaulZan11

Do you not look at teams ranked lower than 24?

We look past the top 24 if we see that we cannot create a solid list of 24 just by reordering. If we find that a team ranked higher than 24 does not deserve to be there, we push them down to a "free fall" list, and if there are enough of the fallers in the top 24, we start pulling teams up who weren't initially in the top 24.

So, sometimes the original 24 makes a good list when it is simply reordered for us, but as you brought up, sometimes we search deeper than that. 24 is just a general starting cutoff for us.

[stundt1]

Originally Posted by Koko Ed

Gotta love the Banebots (or not).

I was rooting for your alliance to get to the finals and be the first to beat 2056 217 and 1533 didnt happen

[stundt1]

Originally Posted by AcesJames

On 176, we do the following -

We have 1 student per robot, and 1 student per HP, in the years that they have a hand in scoring.

.

you scout the human players? Havent heard of that do they make a big impact on alliance selections?

[Josh Fritsch]

I think a lot of teams underestimate the importance of scouting. We have been using the following method for a few years and it has worked out very well.

We have 6 scouts in the stands at all times, one for each robot position of each match (RED 1, RED 2, RED 3, etc...), each with an easy to fill out, intutitive scouting sheet. Each student on the team is on scouting rotations throughout the day so they do not get bored. Data slowly becomes worse the longer you have students scout we have found.

We also have 1 scouting student leader, in the stands the entire day. This person takes the sheets that each of the scounts complete after each match and inputs all of the data into a single home made database. This eliminates the need for multiple computers/electrical/battery etc.

We are then able to have a print-out summary of each of our alliance partners/opponents prior to each match for strategy discussion. Also, this makes sorting through data on Friday night much easier.

[stundt1]

Also does your team have a data entry database or some program to enter information and make your alliance picks be easier?

[stundt1]

anyone?

[ghostmachine360]

I'm using the Aluminum Falcons spreadsheet program as a secondary source.

http://www.chiefdelphi.com/media/papers/2450

Also, I'm using OPRnet in order to get the OPR score for the teams to add to my own research.

http://www.chiefdelphi.com/forums/sh...ghlight=oprnet

But I'm not really using any specific programs other than OneNote & Picasa.

[Dominator1619]

One place to look for information needed is on your team itself, specifically the drive team. Simply asking your teams coach about what info they need to know about teams in a good place to start. As my teams drive team coach, I usually put together a basic scouting sheet for the scouts to work with and develop to how they scout and still get the info we need for strategy. This info also then doubles as our alliance picking info which starts development Friday night but isn't finalized until noonish Saturday during lunch right before alliance selection.

[Bryany]

This year, in addition to pit scouting, we're using a custom program I threw together that is similar in concept to Aluminum Falcon's program:

http://www.chiefdelphi.com/forums/sh...ad.php?t=93841

There is only one person watching each match (although a second person is useful to ensure the first one doesn't miss anything). The user enters in the 6 teams competing in that match, and then whenever points are earned can attribute them to whichever team did the scoring. The program then compiles the data at the end of each match and shows the total points each team has earned in auto, teleop, and endgame, as well as ranking the teams by total points earned / matches played, with the option of disregarding select teams. We took the data and copied it by hand to paper for our captain to take onto the field for alliance pickings, which was pretty quick as there were only 28 teams. It doesn't make Excel docs; that's not my expertise. I opted out of internet databases, as internet is not always available. (it wasn't at our regional anyway)
The program doesn't go into bonuses or penalties or failed minibot attempts, but relies on the fact that if a team is consistent with their scoring they will get a higher subscore in that area. It's simple, but it works pretty well. The program will also give strategy advice based on what teams you are up against and what the strengths of your partners are, as well as what strategies you can expect from the opposing alliance.

[NullEntity]

We have 6 scouters (one for each robot) that fills out a form similar to a Scantron. Then after the match we load the forms into a scanner and let it process them and put the information into a SQL database for later analysis which is done through Excel. Previously, the alliance selection was done through the web system, but Excel offers more flexibility.