Strategy Sub-Team

[Amit3339]

Last year we decided to open a new sub-team for strategy to analize the game in a better way. The goals were clear but it didnt go as plan.

Just wanted to ask how do other teams are running their strategy brainstroming, is it in a small group? the whole team? is there any format of discussion? and if a strategy group should have an influence on the mechanical desing(For Example: can burglarers or the feeder slides).

Thanks in advance

[MaGiC_PiKaChU]

In my opinion, there are 2 types of strategy...

1- Design strategy - this is your design team that try to find the best concept to maximize your team's performance. This year, they'd realize that the fastest can burglars win einstein, and that uncapped stacks are pretty much useless, so they will build a bot in consequence

2-Game strategy - This is your drive team, and people around it. They have to maximize the score by planning what they should do on the field. Who should fight who in the can battles, who should make hp/landfill, throw noodles or not, coop, play defense(not this year)

I don't think that a sub team doing only strategy is useful; it is already done inside other sub teams

[Amit3339]

Originally Posted by MaGiC_PiKaChU

In my opinion, there are 2 types of strategy...

1- Design strategy - this is your design team that try to find the best concept to maximize your team's performance. This year, they'd realize that the fastest can burglars win einstein, and that uncapped stacks are pretty much useless, so they will build a bot in consequence

2-Game strategy - This is your drive team, and people around it. They have to maximize the score by planning what they should do on the field. Who should fight who in the can battles, who should make hp/landfill, throw noodles or not, coop, play defense(not this year)

I don't think that a sub team doing only strategy is useful; it is already done inside other sub teams

The big problem with the mechanical team that I see is that they dont have enough time to sit and discuss on analyzing the game. Our can burglarer came up as a gimik but turned out to be very useful, there wasn't any special attention to how fast it should be. Were trying to avoid those mistakes by putting a special attention to the game elements and Priorities of mechanisems desings. Hope to hear some more replyes

[GeeTwo]

If you aren't already aware of Karthik's presentations on strategy, start there:

Originally Posted by Karthik

myself and Team 1114 will once again be delivering our seminar on Effective FIRST Strategies
.
.

Here are the threads discussing previous editions of this seminar:

www.chiefdelphi.com/forums/showthread.php?t=115843
http://www.chiefdelphi.com/forums/sh...50#post1156250
http://www.chiefdelphi.com/forums/sh...05#post1054705
http://www.chiefdelphi.com/forums/sh...ad.php?t=76778
http://www.chiefdelphi.com/forums/sh...ad.php?t=84958

Here's video from the 2013 version of this presentation:

https://www.youtube.com/watch?v=Apk_X-maRf8

Here's video of the TEDx talk that was inspired by many of the elements of this presentation:

http://www.youtube.com/watch?v=MfC3JdkEVgQ
.
.
If you're looking for a straight lecture on all the topics, I suggest you watch the webinars we did this fall as part of the Simbot Seminar Series.

As with any engineering project, the first step is to understand the problem. In this case, it means learning the game rules (including an analysis of scoring and the availability/opportunity/denial to score) and figuring out what actions are necessary for a robot to be successful. If you skip that step, most of the effort put into developing a strategy is going to be wasted.

[XaulZan11]

Originally Posted by Amit3339

The big problem with the mechanical team that I see is that they dont have enough time to sit and discuss on analyzing the game.

Additionally, there are students and mentors on the buildteam who just hate sitting down and talking strategy and just want to start building. Forcing them to spend a few days analyzing the game leads them to become disinterested and frustration for those who value strategic discussion and plannng. I found it's best to just send them to start prototyping something right away.

[GeeTwo]

Originally Posted by XaulZan11

Additionally, there are students and mentors on the buildteam who just hate sitting down and talking strategy and just want to start building. Forcing them to spend a few days analyzing the game leads them to become disinterested and frustration for those who value strategic discussion and plannng. I found it's best to just send them to start prototyping something right away.

That makes sense, at least as far as prototyping subsystems. By the end of the Recycle Rush game reveal video, you could certainly have come up with some activities that you might want the robot to do:

• make a stack of totes
• put a recycle container on top of a stack of totes
• put litter into an RC

For Aerial Assist, it would have been:

• pick up the ball.
• project the ball to 6+ feet in the air
• catch the ball (truss catch)

It wouldn't have been a complete set for RR, and in each case, one of them was considered not worth the level of effort by most teams, but a few restless folks trying to figure out how to do these things would usually have been time well spent, presuming the strategy sessions were also taking place.

Another useful activity for these folks would be to begin building the practice field, or at least the key elements.

[EricH]

Originally Posted by GeeTwo

It wouldn't have been a complete set for RR, and in each case, one of them was considered not worth the level of effort by most teams, but a few restless folks trying to figure out how to do these things would usually have been time well spent, presuming the strategy sessions were also taking place.

Another useful activity for these folks would be to begin building the practice field, or at least the key elements.

I just want to expand on this a bit.

Sometimes, the strategy team will need something done to show how easy/hard something is. For example, is it easy or hard to pick a frisbee up off the floor? How about stack two totes? In those cases, a quick message from strategy to the restless mechanical team should be enough to keep mechanical interested (busy) for a few hours to a few days.

Example:
Strategy: "We think we want to do X. But we're not sure if it's a good use of time yet. Can we get a prototype?"
Mechanical: *a few hours of drilling, hammering, and sawing later* "We have a prototype for X, it's not working very well but has some promise."
Strategy: "Eh... we'll put that aside for now, what about Y?"
Repeat.

And any team that doesn't have key field elements (scoring platforms, chutes, and the like) is going to have trouble at some point. So getting those built should also be a top priority.

[MMmmDUDE]

How we do it on 2512:
We watch the kickoff video as a team. After the video we gather either at a team members home or a sponsors building. Our strategy department prints off multiple copies of the manual, studies it, and then gives a presentation on various aspects of the game such as every way to score, what fouls there are and the different aspects to the playing field. Team members can ask questions during the presentation so everyone has a good understanding of how the game works. After the presentation students and mentors are put into groups of 4 or 5 and they brainstorm robot designs. Then a representative from each group presents the ideas that the group came up with to the whole team. Designing the robot goes all through the day into the night and then more the next day until the build team has come up with a design they like.

The strategy department informs the build team about the important aspects of the game and recommends certain mechanisms but it is ultimately up to the build team to come up with the robot. The strategy department studies the game more and brainstorms possible scenarios. Also before competition they look for information online about robots that our attending the same regional as us.

At competition the strategy department is in charge of planning matches. They also take pictures of all of the robots in attendance, asks teams questions about their robot, and takes data on teams while they compete in matches. We use this information in a meeting that we hold the day before the playoff/elimination alliance selections. At the meeting we come up with a ranking list of who we would like to be aligned with and make sure the alliance selector knows what robot does what.

Hope this helps!

[MMmmDUDE]

How we do it on 2512:
We watch the kickoff video as a team. After the video we gather either at a team members home or a sponsors building. Our strategy department prints off multiple copies of the manual, studies it, and then gives a presentation on various aspects of the game such as every way to score, what fouls there are and the different aspects to the playing field. Team members can ask questions during the presentation so everyone has a good understanding of how the game works. After the presentation students and mentors are put into groups of 4 or 5 and they brainstorm robot designs. Then a representative from each group presents the ideas that the group came up with to the whole team. Designing the robot goes all through the day into the night and then more the next day until the build team has come up with a design they like.

The strategy department informs the build team about the important aspects of the game and recommends certain mechanisms but it is ultimately up to the build team to come up with the robot. The strategy department studies the game more and brainstorms possible scenarios. Also before competition they look for information online about robots that our attending the same regional as us.

At competition the strategy department is in charge of planning matches. They also take pictures of all of the robots in attendance, asks teams questions about their robot, and takes data on teams while they compete in matches. We use this information in a meeting that we hold the day before the playoff/elimination alliance selections. At the meeting we come up with a ranking list of who we would like to be aligned with and make sure the alliance selector knows what robot does what.

Hope this helps!

[pabeekm]

We’ve only been running the strategy team for the past two years, but it has been immensely helpful so far in letting us know where our decisions are coming from and what our biggest priorities are.

Our strategy team is a small group who practice modeling games in the preseason (I'm happy to say that it’s been getting more popular; as of this offseason we already have 4 underclassmen who have been meeting every week since champs to practice). While the strategy team is at work the first few days of build season, the rest of the team is either reading the rules and brainstorming their own strategies to pitch to the strategy team, handling basic administrative tasks (getting shipping orders in, making supply runs, etc.), or preparing to build field parts.

On our team we've been doing things a little differently; we strictly separate mechanism design and “strategy design”. The strategy team brainstorms, analyzes, and refines sets of game objectives the whole team will later debate. The group does not discuss mechanisms much beyond basic requirements because we want to focus on identifying the best sets of objectives before becoming invested in particular mechanisms that may not meet the best gameplay goals. After we narrow down to the top 3 or so options using mathematic models, the factors that make one strategy a better option than the other cannot be easily weighed (presence of certain partners, mechanism feasibility, etc.); such factors require the full experience of the entire team to effectively judge. After debating the top options as a full team, we hold a blind vote to determine our favored strategy. We then get to focus our prototype building around meeting the top strategy's specific gameplay requirements.

It can also helpful to not just set clear goals, but to also set your design/gampeplay objective priorities in various tiers. When you need to make a tradeoff, these priorities and their tiers should make your defining factors obvious. This is very helpful because when things do go wrong, it’s much easier to pinpoint what factors you were weighing and where your design considerations can be improved for the future (ex. for us this year we made several poor tradeoffs primarily because we kept overestimating weight; weight-based decisions will thus be more carefully made in the future).

[Citrus Dad]

Originally Posted by XaulZan11

Additionally, there are students and mentors on the buildteam who just hate sitting down and talking strategy and just want to start building. Forcing them to spend a few days analyzing the game leads them to become disinterested and frustration for those who value strategic discussion and plannng. I found it's best to just send them to start prototyping something right away.

We require everyone to participate in the build strategy at the beginning of the build season. After working on the overall concept for 1 or 2 days, we start working on individual components which satisfies the hands on folks for a while.

[Citrus Dad]

We have two presentations (slides + video) on our season strategy and scouting strategy. We are trying to get them up on our website.

We don't have "separate" strategy subteams. They are overlays from other subteams. We have the entire team in on the Kick Off design strategy for several days. This lasts about a day and an half. They we're off to prototyping.

Our scouting app team spends time developing methods to capture desirable traits for our first and second picks. In that process we find other strategy nuggets.

At competition, we have a student specifically tasked with setting match strategy based on analysis before the competition and updated with our scouting data on our smartphone app. Our scouting system kicks out our draft pick lists. We then have a smaller group of about half a dozen get together to set up our draft lists. That team is made up of mentors, lead scouts, match strategy and a pit scout.

[Amit3339]

Originally Posted by Citrus Dad

We have two presentations (slides + video) on our season strategy and scouting strategy. We are trying to get them up on our website.

We don't have "separate" strategy subteams. They are overlays from other subteams. We have the entire team in on the Kick Off design strategy for several days. This lasts about a day and an half. They we're off to prototyping.

Our scouting app team spends time developing methods to capture desirable traits for our first and second picks. In that process we find other strategy nuggets.

At competition, we have a student specifically tasked with setting match strategy based on analysis before the competition and updated with our scouting data on our smartphone app. Our scouting system kicks out our draft pick lists. We then have a smaller group of about half a dozen get together to set up our draft lists. That team is made up of mentors, lead scouts, match strategy and a pit scout.

Can you give more information about scouting strategy? I'm not sure I fully understand it's meaning

[Citrus Dad]

Originally Posted by Amit3339

Can you give more information about scouting strategy? I'm not sure I fully understand it's meaning

The scouting strategy is about in-competition strategy. We use it for both match strategy and draft picks. We need to put up our Powerpoint to explain it more fully. I'll see if we can get it up on the white papers here.

[Citrus Dad]

Originally Posted by Citrus Dad

We have two presentations (slides + video) on our season strategy and scouting strategy. We are trying to get them up on our website.

As promised.