Raise The Floor

There a have been a couple of interesting threads recently where concerns about the disparities between low and high resource FRC teams have been discussed. In particular, these two:


At times, it seems to be difficult to define the difference between low and high resource teams, much less identify and solve the disparity issues. I believe, this is primarily due to the inherent complexity of the FRC problem and solution spaces and the wide diversity of the teams working in those spaces.

Much of the discussion in those threads has resolved around ideas (real or proposed) that are intended to constrain the advantages of high resource teams, and if those ideas could work, be enforced, or perhaps be counter-productive.

I thought it would be interesting to start a thread dedicated to brainstorming ideas intended to raise the floor for low resource teams. There are already many resources that raise the floor to varying degrees of success: KOP, Ri3D, online tutorials, Thrifty Bot, CD, local multi-team workshops, and many more. However, I thought it might be helpful to have a place to safely discuss new ideas. While we can discuss changes that would require changes to the program from FIRST HQ, it might be better to discuss ideas that can be started locally and then spread if they prove fruitful.

I start by throwing one out for discussion.

How about a small group of experienced mentors (3 to 5) from different teams in a given area, offering a free off-season consultancy to local low resource teams. The process could be something like this.

  1. Teams apply. I think we would want “low resource” teams to self identify/select and not try to apply a metric and then try to go tell them they need help. These teams would be more motivated to improve.
  2. Local FRC leadership and the mentor group pick a manageable number of teams to work with. The number would vary with the mentors’ time availability and the possibility that some areas could have more than one consultancy team.
  3. The consultants attend a number of the team’s off-season meetings to get to understand the team and their constraints.
  4. The consultants make a set of team specific recommendations. The recommendations could be technical, process oriented, tools recomendations, or anything else. The value comes from the recommendations being tailored to a single team, its needs, and its opportunities. We may want some sort of review process here with local FRC leadership and/or the target team’s leadership before the full team sees the recommendations.
  5. Some form of follow up during the next off-season.

The primary benefit would be that the participating teams would receive team specific “raise the floor” recommendations from a group of experienced mentors. The main drawback would be more work/stress for some mentors but it would be voluntary and strictly off-season.

Has anyone tried this before?

Any other ideas floating around out there?

3 Likes

How would the teams know to apply? If they are really low-resource, they might never find out about it.

10 Likes

I love @Michael_Corsetto 's idea of making the kitbot a scoring-capable design that can successfully play the base objectives of the game. That + some documentation that explain why the engineering decisions were made the way they were for said kitbot would be a great way to raise the floor.

22 Likes

I’m with Gary. High-resource parts of first (HQ, mentors, local teams, regional teams) need to be the ones who take the initiative to find and work with teams that struggle.

One thing that I think would make a huge difference would be creating full generic field perimeters with annual field elements that are accessible for any team in a region. Find a location (warehouse, sponsor, empty mall store, whatever) and set up a field. Reach out to area teams that need it. Find out what would make it most useful for them. Transportation? On-site mentors? Whatever the barrier is, figure out how to lower it. Come to think of it, it would be great if the location of the field was set as close as possible to the low resource teams, and not the higher resource ones.

Anyway this has been near the top of my priority list for at least a year and it’s been a struggle, but I think it is a good idea and unless someone can show me why it’s not ill continue to pursue it

10 Likes

This is an idea that would raise the floor in the sense that, technically, teams would score a lot more points. Does it really raise the inspiration floor? For students who think they’re entering a robotics competition, is being on a team that puts together a kit robot that scores points going to be a vastly different experience than putting together a kit robot that just drives?

On the non-inspiration side, is it going to be a vastly different engineering experience? Yes, it’s easier to iterate on an existing design. But I don’t know that the 25th percentile team is going to be able to make substantial improvements on that kitbot.

7 Likes

I think the actual floor (the teams that go to competition with a clearly illegal/unfinished robot, struggle to drive at all, etc. – maybe 1-2 teams per competition) is different from what most people refer to as the floor (ineffective robots / teams that would be more competitively successful building Everybot and just practicing until comp – probably 50-75% of teams depending on year/area).

Things I think will help the first group the most:

  • Resources periodically sent out via email (e.g. links to Chief Delphi, Compass Alliance, and local offseasons/workshops in the fall, links to Everybot, Ri3D, and common FRC vendors during build season)
  • Everybot plans (or info + links) included in the KOP
  • Bumper build instructions + parts included in the KOP
  • Kickoff events with a KOP building/wiring/programming workshop after

And for the second:

  • Everything above
  • More scrimmage events or access to practice fields
  • Inspection workshops at scrimmage events (previous inspectors quickly run through the inspection checklists with teams)
  • More local offseasons
  • More local events
  • Cheaper second play / districts (second play included)

For mid/upper-mid tier teams (what a lot of the cost limit thread seems to be focused on):

  • Everything above, especially the cheaper/easier second play
  • Intermediate level between local plays and champs (i.e. DCMPs)
  • District qualifying (these are often the teams that barely miss out on a qualification spot)
18 Likes

I think the most insurmountable obstacle for noncompetitive teams is lack of mentorship. I’m not talking about someone who comes in three hours a day for four days a week, I’m talking about someone who spends an hour or two each week just doing a basic strategy/design review. In this vein, Senior Mentors should be reaching out to struggling teams and offering to help. Waiting for people to ask for help only reaches those who ask. If we want to inspire the hundreds of teams that never look beyond the schoolyard gate, then we have to be taking active steps to broaden their horizons.

Making a scoring kitbot will certainly create a basic level of inspiration. I would much rather teams follow instructions to make something slightly worse than an Everybot each year, but that shouldn’t come at the cost of innovation. If the basic kitbot is “too good”, then innovating actually hurts your competitiveness. Kitbot should be good enough to complete the basic scoring task, but teams should be encouraged to improve upon it. My hope would be that building this basic robot would teach teams what good design looks like, giving them ideas on how to construct upgrades. A 2019 Everybot with no hatch panel mechanism, for example, would have been an excellent robot to compete with, without hurting the mid-tier teams that try to push the boundaries.

Every year, I expect the median team to be better than they turn out out to be. Scoring 2 gears, 3 cubes, or 4 cargo would outperform the average robot at regionals I’ve been to in 2017-2019. It’s a sad reality that inspires nobody.

15 Likes

I think this is the biggest hurdle to overcome. During build season it is hard but never asking for help happens at events too. I love it when teams walk up to me after a match and ask for help or something that I saw happen. I still go hunt teams down when I notice something happen but that can be hard to see sometimes (ex: climber is not working so we didn’t try to climb and instead played 15 more seconds of defense).

I think some teams have been trying to do this. Here is a link to the first CD post I could find: Team 4201 (Near LAX; LA Regional Location) Opens Full Practice Field. What do you think these teams could do differently?

1 Like

I’ve completed at events where the only meaningful points we’ve scored were in a singular practice match, both as a student and mentor. When I was a student, those kinds of events made me want to quit robotics for good. As a mentor I’ve seen students beyond discouraged from the frustration of seeing their hard work fall apart on the field. Playing the game at a base level is a huge source of inspiration, and I am confident that making that accessible to every team will raise the floor significantly in all areas, not just in terms of points scored.

15 Likes

I should have included a step 0 about advertising. A general email to the 2 registered lead mentors for each team in the local area would be a start. However, I agree this would be one of the trickier parts but something that could be overcome.

Edit: And for teams that clearly need help perhaps an email directed to just that team from a Regional Senior Mentor or similar that encourages participation would be a good idea.

Putting the 118 Everybot in the KOP is my current dream. Can someone come up with a catchy hashtag for this?

20 Likes

#everybot4every1

36 Likes

This is an interesting idea that could definitely help some teams. I would have a few concerns. First, it would increase cost for someone: higher registration fee and/or more being asked of KOP sponsors. Second, it would mean waiting for FIRST to act and I was hoping for ideas that could be locally actionable.

Maybe there could be a way to address these concerns. I have spent a bit of time sitting here thinking about it and all I have come up with is making it optional (similar to drive base opt out) to control some cost but that would not address both concerns.

1 Like

I liked this one because it is a nice tiered breakdown and some of it could be locally actionable (as much as anything in the midst of a pandemic).

1 Like

Yup, my thought was always to make it optional. I also have many ideas on how this could be done without adding too much (or any) cost to the KOP, but that’s a bit beyond the scope of this discussion

1 Like

Completely with you on this. For me, a big missing piece right now is a clear communications platform to make those connections. It’s a real shame that a high resource team can’t easily reach out to newer and lower resource teams nearest them who are in need or, even before those teams are formed to begin with.

We have no reliable contact information for the teams around us and we are 100% reliant on overworked (this might also come down to local politics with those individuals too as to if you can get emails forwarded) RDs/DCs to forward communications to reach out or else we have to make larger protracted efforts to find ways of communicating with those teams through social media, through in-person happenstance meetings, etc.

“BhUt MuH DaTa PRiVaCY!!!”… There are a TON of ways to make this work and reduce the communication gap while staying compliant with privacy laws. I chose " communication platform" as a generic term to describe what needs to happen - it’s not a prescriptive approach to this problem.

6 Likes

Coming from years of experience starting 100+ new VEX IQ and FLL teams, I can confidently say that VEX IQ having the claw bot design/instructions/stock code set itself head and shoulders above FLL. Being able to follow a set of instructions and get a robot that can “do the thing” is so huge for students that are just starting out in robotics. There’s a level of engagement right off the bat that just doesn’t exist when you are struggling to put anything that works on top of your chassis. And students are way more likely to “iterate” if they have something working right out the gate.

Currently, FRC is way more like FLL than VEX. Few instructions or resources directly given to teams in the kit. Long-term, I’d like to see the FRC version of the clawbot (everybot) make its way into the KOP.

TL;DR Score points, get hooked, be inspired.

21 Likes

What’s more inspiring and a bigger learning experience: building your own chair or putting together an Ikea table?

Hey I’m a new FRC student and I’ve never seen a chair before. I don’t even know what one should look like, what they’re made from, or where to buy the parts for said chair. What do?

20 Likes

Say FRC went with an Everybot with instructions instead of the KoP. Do you think that there would be many teams that would build Everybot year after year and not try to modify it to an Everybot+, or a 1stpickbot, or even a Captainbot?

Did you see growth in what the VEX IQ claw bot teams designed year after year?