Alliance Selections - Not just picking numbers

This weekend 77WHO?! was at the Finger lakes Regional, where we placed 11th after qualifications. We were having issues on Friday which caused us to lose 3 of our match’s, if we were alive we could have had a better chance to winning them. But that’s beside’s my topic/concern.

Really what I want to talk about is how teams chose to Decline/Accept and offer and/or who they would be picking to be on their alliance. This year is not like any other year for alliance selections. Teams cannot go up to the alliance selections with a list of a couple teams of who they want to pick, because it just isn’t that simple anymore. Yes Last year you could go up (well sometimes) go up and just pick the next best team, but not this year. The issue this year is that there are SO many different kinds of robots. You have full court shooters that sit at the feeder station and launch frisbees constantly, close range shooters who run back and forth, 30 Point climbers that dumb (50 point). close range shooters that pick up off ground and “clean up”, pure climbing, pure defense AND MORE!

So how do you decide who to pick?

Here’s my Opinion: The person going out to pick MUST have a large knowledge of all teams strengths and weaknesses. They don’t need that knowledge for just only the people who you will be picking but for the people you will be playing against. You need to look at the teams you will be playing in the future. Look at how you will be able to score more points then them. look at how you will be able to play defense or IF you will even be playing defense against them. This year something I found was that you don’t need a defender, you can have 3 amazing offense robots together and be able to out score the other alliance.

Accepting or declining. If you are a team that is going to be a 8th seed or higher even if you decline you seriously need to look at the team that selected you. questions you should ask yourself:
Will they ever help you?
Will we be able to beat the alliance we are coming up against or would we be better off picker our own alliance?

These decisions will possibly decide if you are going to win or not.

This thread is really just a WARNING to all teams that will be competing in the next few weeks after we got to have the fun experience of being “test subjects”. Please voice your opinion as a structural comment, remember Chief Delphi is for helping.

  • Aaron

It all depends on what my robot is good/bad at.

For example, if my robot climbs for 30 points and dumps in the pyramid, but doesn’t have the strongest autonomous, I’m not going to pick another climber-dumper. I’m going to pick the team that has the best track record in autonomous and teleop, etc.

This type of picking strategy, while especially relevant this year (as there are many ways to successfully contribute to the game, strategy-wise) is ALWAYS relevant in FRC. Picking an alliance that complements the alliance captain robot’s strengths and weaknesses is one of the keys to success, no matter if the year is 2003 or 2013.

Also our team, among others, takes extensive data for every single team for every match. We sort the information using a spreadsheet which we use to rank teams based on their scoring effectiveness, etc. Armed with this knowledge, it’s easy for our scouting captain (anyone privy to the data we collect) to walk out on the field with our ideal alliance planned out, with several backups prepared in case we don’t get exactly who we want (which is pretty common).

Very insightful post! As always, I am going to reiterate that I believe the competition ends immediately after alliance selection.

Okay, not really. It’s just super important! My biggest/most relevant example happened this weekend at Kettering. I may be slightly biased, but I believe that 910 was easily one of the best 5 robots present, yet they somehow managed to the last robot selected, putting them on the winning alliance. Despite the fire power of the first seed/pick (70+862) I think that the eliminations could have gone much differently if another alliance had 910 instead of the first seed.

My hypothesis is that teams assumed 910 would be gone, and thus omitted them from a potential pick list. I made a similar mistake last year (forgetting about a certain Florida team) and that may or may not have been why our alliance didn’t do as well as we could have.

So the moral of the story is kind of related to Aaron’s post. Know how you want to play, and a pick list to accompany your strategy, but don’t forget to add key players to your list, or maybe even have multiple lists, because accidents happen (and can even win you competitions!)

To put it most simple and short, the best alliance is a well rounded alliance.

Methodology:

  1. figure out what my team is good at
  2. determine optimal strategy for playoffs based off of my team’s capabilities
  3. make multiple rank-lists of characteristics I’m looking for in teammates (long shooting, close shooting, defense… etc)
    4a) if choosing, select teammates to optimize predetermined strategy
    4b) if chosen 1st round, figure out what alliance captain’s plan is and share data to optimize alliance strategy and 2nd pick
    4c) if chosen 2nd round, you really don’t have any say in the matter

Edit: And for the love of mankind, cross off a team from your lists when they are selected by another alliance. Not only is it incredibly embarrassing to the person/team who is selecting teams already in an alliance, but it is just straight up unnecessary and ridiculous.

For the record, I believed this also. I tried to inform the 4th alliance captain, 1189 to pick 910, but to he chose differently. I’m not mad, just knew 910 was better than most thought.

The fact that this game isn’t a “one task wonder” was one of the primary motivations for us to switch from a linear list to a whiteboard. We still give the rep a linear list, with notes on the robots included, so they can communicate with the other captain in case they need to and as a backup if they can’t see the whiteboard, but being able to offload the decision to a group of people off the field really helps.

Wasn’t there a thing against using “wireless technology” to influence alliance selections? I recall a post from a little while back about a team thinking of taping a wire to a board in the same manner…

In the last few years, we’ve offloaded the decision to the scouts off the field, connecting to our team captain and representative on the field with a smartphone with data and a google doc. White board/team number written on a bit poster as a backup.

Yep, that was 1114 and featured in one of Karthik’s talks. (Or it might have been an EWCP cast…)

This was an isolated decision at a single regional that didn’t stick, not FIRST gospel. Smart teams have been “whiteboarding” for many years, we just finally joined the club.

Why would teams ever omit anyone from a pick list? I know it’s not easy to do this, but IMO a pick list should include at least the top 24 teams there (including the ones in front of you, which can help you decide if you want to possibly decline). Just cross off teams that are already gone…

I like most of your post but disagree with this part enough to comment. If you’re going up against a full-court shooter you better have a defensive plan; if you let it sit there you’ll be able to watch the remaining elimination rounds from the stands. :frowning: Block them, disrupt them, don’t let them get to the loading zone, something…but defend against them. The other alliance can use their other two bots to slow down your shooters’ load/shoot cycles and it’ll probably be more than enough to beat you. Maybe you use one of your offensive guys as a defender, but still; you’re going to have to plan on defense in this situation, imo. Even if you have one too; why take chances?

And…if you’re at an event deep enough to let you draft 3 amazing offensive robots, congratulations. Or condolences; I’m not sure which. :smiley: Although as another poster noted about 910 at Kettering, good robots can slip through the picks unnoticed. Our #2 pick at Hub City – 1801 – was well towards the top of our list but for whatever reason was available. They could have easily been someone’s first pick.

There’s a great story about that within The New Cool.

As far as everything else has been stated, while I agree with both Aaron and Jaxom, it really depends on what regional you’re at and what the state of the shooters are there. You could be at a regional knee deep in amazing cross court shooters, or you could be wading through puddles of decent shooters that could function as a cross court feeder. If you know that your next few alliances have strong cross court shooters, get a strong powered defender. If only one of your matches have a cross court shooter, just add a pizza box defender to a shooter and you’re good. (For the love of all things FIRST get the thing inspected (And ensure that it is indeed inspected) after you add it!). Alternatively, if you can get a shooter like 973 or 639, who are really really tall shooters, all the better! They can function as tall defenders but still with the ability to shoot.

Is there a specific reason why 910 was good? Were they just high scoring in general?

By their 12th qualifying match they had finally started to drain the three point shots from the feeder. They continued that all through elims.