[EWCP] 10-30-11 Cast - Scouting

[Akash Rastogi]

I'm interested to hear how other teams end up making their pick lists.

[Taylor]

How much weight is given to pit scouting information vs. game scouting?
Do top teams scout themselves?
Is pit scouting even really worth doing, or is it used as an exercise for the rookies who have no other tasks? (kinda like the lead screw in the KoP)

[JackS]

Originally Posted by Akash Rastogi

I'm interested to hear how other teams end up making their pick lists.

Our team representative brings a metal bingo cage out to the field for alliance selection and picks the team whose corresponding ball comes out first.

[Chris is me]

Originally Posted by JackS

Our team representative brings a metal bingo cage out to the field for alliance selection and picks the team whose corresponding ball comes out first.

Actually, we pick teams in order of best performance on Einstein. Don't you remember 195, then 1124? 177, and 173?

[Frenchie461]

Can you guys let this one run a tad long if necessary, so nothing gets cut short, because this could be the most interesting one yet.

[Alpha Beta]

I'm looking forward to this cast.

One of the problems we seem to have is consistency in our data collection. Part of the problem is keeping the entire scouting team motivated. (Seems to be a common issue from the other comments on this thread.) One of the things that helps is to concentrate our data gathering on Friday afternoon only and double checking questionable data that night with video review. It is certainly easier to stay motivated when we're winning and the prospect of captaining an alliance is still in the picture. What do other teams do to ensure quality data?

When we have been an alliance captain, picking our first partner has been relatively easy from the list we send down to our representative. Often our list ranks exemplary defensive potential above mediocre scoring and a gem overlooked by others is still available for our second pick. We keep a copy of the list in the stands and if the top available pick is no longer a good fit for the alliance we do have ways of signalling our representative. What do other teams do here? Do they leave it entirely in the hands of the representative or is their some interaction on the final decision with the rest of scouting team? Do other teams have a static pick list, or in what ways do they adjust it based on the picks of other teams?

Pit scouting was critical last year in KC when hardly any practice matches got played due to a technical issue. No one had a good feel for the capabilities of the other teams in their first qualification match. Most years the only value it serves is a medium to say high to other teams and build relationships. We do just as well without paper and clipboard. I usually take a small group of our more mechanically inclined rookies around to each pit and ask them to look at each robot and tell me what they think it is capable of doing based on how its built. Then after we hang back and have our own discussion we approach the team and ask them to explain their design to us. It absolutely makes that group of rookies better at designing for the next build season. Every pit feels like Christmas morning. It is one of my favorite times at each competition. What do other teams use the pit scouting time for? Are mentors involved ever or just students? Do you base your scouting reports on what you see or what other teams tell you? (We all know some teams have hopeful delusions about what their robots are capable of.) Do you incorporate pit data into your pick list?

We do a little scouting report of each alliance partner and opponent for each match based largely on their previous match. The more participants in a tournament the easier this is to accomplish due to turnaround time. Our drive coaches have plenty of time in que to analyze this data and discuss it with our partners. We have found that the more organized you are the more influence you can have with your partners when designing the alliance strategy. If there is time the drive coach and chassis driver come to the stands after their match and the scouts give them a critique and suggest improvements for their next match. How do other teams develop specific match strategies and communicate to their drive teams specific data about their upcoming match?

We have had other teams ask to share scouting data to the benefit of both teams. Is this common? If another well respected team asked you for your data so they could combine it with theirs and then send you back the combined files would you allow it? Is this a good idea to build a relationship with another team, or does it lead to similar pick lists and run the risk of someone else sniping your sleeper pick?

I'm generally one of those silent guests in the chat, but I'll be there hearing every word.

[Jim Zondag]

Originally Posted by Dr Theta

How about how to effectively relay subjective and strategic scouting info to drivers for use in matches? Being a coach I know this can be huge.

We have a simple and effective way to do this which we have used for over a decade on our team. We will be sure to touch on this

[Andrew Schreiber]

Good news everybody, I'm finally learning how to use Talkshoe...

Call information:
TEL: (724) 444-7444
ID: 98466 You may need to register with Talkshoe if you haven't before.


http://www.talkshoe.com/tc/98466 should have a link to the cast once it goes live ~830PM EST

[Alpha Beta]

One more line of questioning for the cast. I've seen quality teams have mechanical failures or software/electrical issues in their final qualification match. Do your scouts generally cross them off the list or investigate in the pit the details of the malfunction? Do you take teams at their word when they say the problem will solved or do you ask to see proof? If you're team is the one with the malfunction what can you do to let the other scouts know you'll be ready to play?

Ok, 2 more lines of questioning..

If you're looking to get picked is there anything you can do beyond robot performance to move up someone's pick list? Does going to the pit of the top 8 teams and telling them how good your robot is do any good? Is it reasonable to show off your best feature in your last match or 2 (instead of trying to do everything) if you think you are on the selection bubble.

[Alan Anderson]

Originally Posted by Alpha Beta

Does going to the pit of the top 8 teams and telling them how good your robot is do any good?

Touting your robot to another team's pit crew probably won't help, for two reasons. First, the pit crew generally doesn't have a lot of influence over the scouting group's analysis. Second, the scouting decisions are generally made based on observation of robot performance rather than advertisements, and it might even hurt to be seen as "bragging".

That said, you might get some benefit out of explaining both the cause of a particular performance issue and how you've corrected it.

[Akash Rastogi]

Our head scout and drive coach always go to investigate any teams with mechanical failures. Never dismiss a team without figuring out what went wrong and how they are dealing with it.

[Chris is me]

Originally Posted by Alpha Beta

One more line of questioning for the cast. I've seen quality teams have mechanical failures or software/electrical issues in their final qualification match. Do your scouts generally cross them off the list or investigate in the pit the details of the malfunction?

We'll talk a lot about Saturday scouting in the podcast (it's something I personally think 2791 is very good at), but personally it depends on where they are on the list. If they're near the top, it takes a lot of thought. We'll go down to their pit and figure out what's going on and depending on their response we'll adjust them relative to other teams of their caliber.

Many teams are on the list for "flashes of greatness" and thus need to be thought about differently. Definitely marked down as high risk, but if the reward is high enough then we might want to take it. At CT in 2010, 173 had technical problems all of Friday that stemmed from a defective classmate PC, but they had a strong autonomous mode and consistent hang, setting up us for a relatively easy 4-6 points with their presence alone. Autonomous from the back zone was our priority for a second pick, and they had a hanger that we thought we could use. The position we were in called for a pick like them, so despite their faults we took the plunge. I'd certainly make that same pick again, as despite some ill timed failures they were a huge asset to the alliance.

So basically, you have to weigh their potential over the particulars of their failure and rank them relative to your teams on the board. This is a lot of "gut" work; once Saturday rolls around numbers can only do so much for you.

Oddly enough, sometimes we learn more about the team doing this than if they didn't break. I remember at some event a while back, one team took the time to explain to me exactly what went wrong and how they were planning to fix it for elims. We weren't even in an obvious position to pick them if I remember correctly, but the extra effort put in to communicate with us reassured me that they would be a good pick broken or not.

Do you take teams at their word when they say the problem will solved or do you ask to see proof?

This depends a lot on how vitally you need something. A good example would be an autonomous mode. If you're picking someone for their autonomous and their arm snapped off, I'd ask to see it if there was time.

If you're team is the one with the malfunction what can you do to let the other scouts know you'll be ready to play?

Be proactive in selling yourself. Find teams you think you would work well with and have a frank, down to earth discussion. No bragging.

Thanks for the great questions, and I'm sure all of us will go in depth this Sunday. Scouting is more or less my favorite topic in FRC.

[XaulZan11]

Originally Posted by Alpha Beta

If you're looking to get picked is there anything you can do beyond robot performance to move up someone's pick list? Does going to the pit of the top 8 teams and telling them how good your robot is do any good? Is it reasonable to show off your best feature in your last match or 2 (instead of trying to do everything) if you think you are on the selection bubble.

As Alan already stated, going to teams' pits rarely works. Even if the scouts are there at the time, its unlikely you telling them how good you are will change what the see on the field. If you have been having mechanical issues, it may help to explain what the issues are.

In terms of being on the bubble of getting picked or not, its a very good idea to learn what teams are looking for in partners and have an idea of the depth of the field. In terms of knowing what teams are looking for, this year a common elimination alliance consisted of 2 scorers and 1 defense/herder. If your hoping to be one of the last few teams picked, you probably shouldn't spend the entire match strugging to score 0-2 tubes (like so many teams did this year). Instead, you should showcase your ability to play defense and herd tubes.

Its also important to understand the depth of the field at your event. This was big in 2010, but not as much last year. If your a bubble team that plays each zone fairly equally well and you see there are, say, 18 adquate zone 1 robots at your regional, but only 6 that can play zone 3 decent, you should play each match in zone 3.

As least at the regional level (especially at the shallow ones), its rare to see a team on the bubble that 'gets' strategy to take advantage of these two things.

[Chris is me]

Originally Posted by XaulZan11

Its also important to understand the depth of the field at your event. This was big in 2010, but not as much last year. If your a bubble team that plays each zone fairly equally well and you see there are, say, 18 adquate zone 1 robots at your regional, but only 6 that can play zone 3 decent, you should play each match in zone 3.

This is tricky, though, because most of the time your best skill might not be what's needed to win the qual match. Even if we were a great back zone defender, what if we were the only team with offensive quality?

[XaulZan11]

Originally Posted by Chris is me

This is tricky, though, because most of the time your best skill might not be what's needed to win the qual match. Even if we were a great back zone defender, what if we were the only team with offensive quality?

Entering every event, the goal is to win every match and secure the top seed. But after you are unable to seed high (typically after your 4th or 5th loss) and you know your likely on the bubble, then I think you should consider strategies that put your team in the best position to get selected. Obviously, you never go against what your alliance partners want to do, especially if they are in the race to seed high. It's a lot like the dynamic of the 6-0 matches in 2010, when teams would intentially lose matches to help them seed high to win the event.