Mentoring at Kickoff - Strategizing, Clearing Ideas, Etc.

FIRST General Forum
1

As a second year mentor, I am still looking for the best ways that I can use my skills and knowledge to help not just in my subteams, but in the team as a whole. A way I’m looking to how others mentor is specifically on Kickoff. I feel like every team should have a general schedule of how the day is laid out, but I’m hoping to find ways that I can further promote strategic thinking and prototyping.

As a mentor, how do you aid in the initial strategizing of students for the new robot, whether it be coming up with the best strategy or helping find what prototypes should be looked at when considering subsystems?

Or if you’re a student, is there any methods mentors help with critical thinking that stick out to you?

2

It’s important to consider ALL possible strategies. After all, if you’re not playing it, it’s likely you’re playing against it. Keep in mind that you’re not really looking for the “best” strategy (that would be to do everything really well, which isn’t very feasible or useful), you’re looking for your best strategy. How can you play the game to the best of your ability, and no better?

I’d also suggest discouraging students from being possessive of certain ideas. It’s very easy to latch onto an idea because you thought of it and to take offense if someone disagrees. Every idea is “an” idea, not “xyz person’s” idea.

3

Make sure your team determines their strategy before they even begin to think about what mechanisms that would solve said strategy. I think everyone has fallen victim to jumping right into solutions before a strategy is determined. That is how you end up with a bot that is superbly mediocre at everything and not great at anything. Ideally the team will come to a consensus, but most of the time not everyone will get what they want. This is often where teams will vote for their strategy of choice.

In FRC, and in design in general, vote is a four-letter-word. By that, I mean voting does not objectively choose the best solution, it chooses the most popular solution. When deciding on a strategy, encourage your students to explain why they believe a particular solution is the best, or why they think other strategies are not as good. Encourage quantitative analysis rather than qualitative, or in other words, explain how one strategy will net your alliance more points (or less points for your opponents).

In regards to deciding on final solutions (after brainstorming and generating solutions), I will repeat that vote is a four-letter-word. Ideally you don’t want anyone attached to any particular solution, you want to choose the best solution for your team. This is where weighted decision matrices are incredibly helpful.

148 has an excellent write-up on how they use them. The basis is determine some criteria for which solutions may differ. Examples of criteria include but are not limited to speed, weight, complexity, maneuverability, and required driver precision. Ideally you will have a category for each factor that would vary from one solution to another. The team then collectively weights these criteria based on what is important (google sheets is AWESOME for this so everyone can do it at the same time). Make sure everyone uses the same scale (0-5, or 0-10, or whatever scale you want).

You now have your criteria weight. For each solution in this category, lets say the drivetrain for example, you rate the specific solution, lets say tank, multiply that rating by the criteria weight, and the sum the total for the solution. Once you do this for each solution, you now have a (relatively) objective rating of each solution compared to each other. Ideally the team will choose the solution that scores the best, but that is not always the case.

Hope some of these tips help you out a bit!

4

There must be a fun way to get the team to fully understand the new game/robot/etc rules on kickoff day. I wonder what it is?

5

In my two years as a student in FIRST, the best thing that always helped me and my teammates figure out any problem was going to a whiteboard or even just a piece of paper and listing ways to solve the problem and how each way benefited from and was worse from the other ideas. Have those who came up with the idealist pros and cons because that way they get to think more in-depth and their idea is better portrayed by others.

Otherwise, for a mentor, the best thing to do is let the students lead the discussion and you just stand back, intervene when say they get off track like at kickoff make sure they don’t go into designs and also input your experience to give more information to a topic.

This year I will be mentoring a rookie team with no prior first experience so I plan to do my best to keep them informed and educated about my experiences and how FIRST works.

6

I have lead several strategy sessions. We generally watch the kick-off and download the manual. We let everyone get some excitement and energy out, typically while a mentor or experienced strategy student read tournament section for any surprises. Ranking system is often the #1 thing to understand.

We then do the boring task of reading the game rules as a team. We will also usually read through robot rules, though more briefly looking for changes/updates.

When reading the game rules, if a question arises, we either try to answer it immediately, or if discussion or more information is needed, we have a “parking lot” which is really just a white board or post-it pad or… list that is kept until the end of the reading of the rules. Once we finish reading, a quick stretch, then we go through the parking lot to see if we can answer all the questions/concerns. Anything that cannot be answered, may eventually go to the Q&A, though we usually wait until after the first round of updates.

At this point, we will do a functional decomposition. IE, what are all the things a robot/team/human player can do in a match. If students get stuck, we will use some thought starters like: things that score points, things that inhibit scoring, things that descore, penalties you can get on accident, any penalties you might get on purpose?, odd things to do with field elements, odd things to do with scoring pieces…

Another bio break and some stretching/running around. We will then revisit scoring and objects and do a scoring analysis. Goal is to figure out if there is a maximum score, what a likely early score will look like, what likely high score will look like, what we think it will take to win a district, state championship, world championship…

Usually then we break into smaller groups (4 to 8 students) and develop a conept of operations or basic match strategy per robot and what sort of general systems would be needed to implement the strategy.

Once the groups have written up a basic design, a presenter presents from each group.

We will then go back and revisit scoring analysis and look at a minimum competitive concept. IE, if you have low resources, what must you do to be considered as a partner, what must you do to get into position as a captain. A lot of discussion is spent here talking about ideal strategies and types of robots vs. effect solutions that require less resources.

Usually around this time, the KOP shows up, and we play around a bit with the game piece.

We will then revisit what we think is a good strategy, and what sort of concepts need to be prototyped, what need some experimentation, what need to be designed.
For instance, if a piece is compliant, you often need to figure out what sort of ideal compression to move the piece.

If the overall concept is contingent upon a prototype, then we will work that prototype until we have a good proof of concept. I will say, this was a bigger deal in the past when concepts seemed “newer”, now frequently, those experienced in first will hear an idea, then you can actually attach/associate that with a team/robot in the past. Want to collect balls using bristle brushes? Let’s watch 179 from 2012. Want to use a wheel shooter to shoot 1-3 precise shots of a foam ball, let’s see what the best in the world did in 2012 and 2016. Handling 1 large ball, reference 2014 or 2008. Picking up an inner tube, how about 2011 and 2007… Stacking totes… see 2015 or 2003.

This will often involve trying to find matches or images of old robots, or pulling out Behind the Design books.

Back to the OP, part of what you do as a mentor depends on your teams philosophy as a mentor. Some engage more than others. As a mentor, keeping the group productive is a big challenge. Let the environment be fun is important, but ensuring it is a chaotic recess is also important. A tough challenge is being supportive, without letting the teams go on wild goose chases. This is an area that is also important to the team philosophy. For instance, if the team is choosing a concept that you are fairly certain they cannot execute. Some mentors/teams believe it is important to real the team in and pursue a more realistic approach. Others believe that failure in and of itself is an important lesson, and thus you let the students pursue the crazy. If your team has core values or an agreed upon philosophy, it can help. If not, this will likely be an area for conflict where personal values of different team members do not balance with your beliefs. This is an extremely common source of conflict on projects (not just FRC teams).

7

Sarah - one thing that we typically don’t have any mentors want to do the first day is create a bill of materials of what we need to build field components. We’ve been fortunate to have a parent or two that will go through the team field components document and make a shopping list and compare to inventory. I like to be able to purchase that before the end of Saturday so we can start building on Sunday. We usually also have a disagreement on how much of everything we need to build (for instance, in 2015 I should have realized we didn’t need to build all the platforms or steps, and in 2017 I had to push hard to come up with a design of our own for building the airship for practicing since it wasn’t in the team drawings).

Pearadox has been hosting other teams for kickoff. This can lead to a very large group of people, and this year I’m anticipating around 100+ students from 6 teams of varying skill and experience.

My plan is to:

  • After kickoff, pull up the game animation a couple more times and point out some potential illegal maneuvers (as is common in the game animation). Point out things we may want to get immediate clarification on
  • Watch the field tour videos again and point out anything that we may want to get look at closer or that may be trickier than it appears at first glance
  • Allow students to search the rule books on their own and eat lunch.
  • Come back together and highlight major rules, identify methods of scoring, etc as a large group. This is usually done in a format of question and answers (how big can the robot be, how much is this action worth, are bumpers included in perimeter, etc)
  • split up into small groups (5 or 6 students) to come up with different strategies they think might be seen at competition (not necessarily what we would/could build).
  • “play out” some of those games on our field with people as robots.

We then break up into groups and reconvene a couple of different times.

Goals of the first day:

  • identify things that affect rank (this is important for deciding robot function priorities)
  • identify all methods to score points
  • identify all methods to prevent points from being scored if applicable (examples could be can grabbers in 2015 even if we can’t score cans)
  • identify most/all robot functions (2017 examples could include robot speed, shooting speed, shooting capacity, floor gear pick up, floor fuel pick up, human player pick up, climb, etc)
  • Begin to hone in on strategies achievable by our team. Usually this can’t be finished on day 1 because everyone needs to digest the game more. But this gives us a starting point on what we really need to prototype and build good field models for us.
8

One of the things our team does is make a mock field and play the game as the robots. We tape out the field or at least half of a field, use boxes and totes as field elements, and watch the strategy develop.

9

This has been a strategy I’ve wanted to do for a few years! Would you say that it helps people understand the gameplay better?

10

We’ve done this every year and I’d say the answer is “it depends”. It can be good if you have students that generally know how a robot ‘behaves’

You have to make sure that the people that are being robots try to be robot like - which usually means not running or using their hands. Sometimes it also means extrapolation - for instance in 2015 I wouldn’t have been asking any student to carry 6 totes with a garbage can on top.

11

Kick off is all about management.

Check out Citrus Circuits’ videos:

The first 26 min are great, too, but at this link point Mike is talking about kick off.

12

It completely changes how the game is looked at when it comes to understand the game. Highly important that the game play rules be read first though.

13

Definitely important to read the rules first. I know we’ve taken a little while to understand some field elements in the past and I bet doing a mock game would help with that.

14

Second year expect improvement!

Here is my advice to help focus a team as a mentor:

First, what are the “MUST DO’s” what I mean by this is what will the scouts need to pick you?

Secondly, " Can you score alone ?" , there are some games you will be the only bot that scores, what does that score look like during all three phases (auto, tele, end) also how will you earn a Ranking Point (if available)…alone. This part is all about ranking/showing day 1, being in scout discussions that night.

Third, “Are you reliable/robust” , that boils down to “are you an asset or liability to an alliance ?” Being unique helps as well especially as the third pick.

If you focus on those items, you will either be selected or selecting in just about any competition you enter.

I also do scouting with my scout team, my lead scout makes our 28 deep pick list. We do that with “eyes on bots” and make a list of those that help us even if we are not selecting teams know we scout well they see us. That helps to become an alliance partner as well.

In the end we want to be an asset and not a liability, that works.
Doing a few things great is way better than a lot of things not great…not much time to do many things.

15

I’ve tried to distill our method down to easy to follow steps for teams. The methodology is a hodge podge of researching other teams styles, trial and error and determining what works best for us over a ~15year period. Everyone does it slightly differently, but many of the core guiding principles are the same.

We have a recorded seminar here: https://youtu.be/tNI6Lp_gveY

And the seminar itself is up here: https://www.thecompassalliance.org/single-post/2017/12/07/Game-Breaking

Good luck to you guys!
-Brando

16

^ This is one of the fun parts! ^

There is some more on this topic here…
https://www.chiefdelphi.com/forums/showthread.php?p=1715671#post1715671

Our mock game play tends to be the most memorable part of the day…
well okay not MOST…:yikes:

It is a great way to flush out designs and strategy too.

We go out in the driveway and mark off the “field” and go…
People play the alliances, pick a function they want to emulate, etc.
You walk through the game… do your auto… how can I get here to there. can you even see that!!

loosely time it… score it

The first several matches mentors call out to the players that their chosen movement/play is too complicated and such.

After a few matches everyone chimes in on what is going on…

We have several teams attend so this can take some time. So worth it though.
It is great for the teams to play through the game together.
A nice break from sitting and reading, some times a great idea comes from out the blue as someone runs by… “Hey my next match I have a hooked pole…”

Good thing is we get some fun outdoor times playing a game we are all excited about… Team bonding… nah family bonding!

Have fun,

Aloha!

17

Thanks for the great info! It sounds kind of similar to how we usually visualize the matches going, just a bit easier to grasp. We usually lay it out on a whiteboard and have people move around to estimate cycle times, so this method seems a bit more visual which sounds effective. Good luck to you!