Will Your Robot Be High Goaling, Scoring Gears, AND Climbing?

Based on posts by other people on Chief Delphi, it sounds like a lot more teams will be attempting to score in every way possible than in the past. I was a little startled when I saw that many teams will be attempting to score in the high efficiency boiler, score gears, and climb. I made this poll just to see how many teams are actually planning on doing all three.
I’m aware many teams will say they are attempting to do all three and that may not occur on their final robot, but I’m still curious to see the replies.

I always love the optimism of week 1.

This poll tells me that more than 60% of teams will have at least one mechanism on their robot that never works or works so poorly that it isn’t worth using (scoring 20 balls in the high goal all teleop). I will concede that a small portion of teams will succeed in having at least decent ability in all categories.

For everyone who voted the first option, please go watch Karthik’s Effective FIRST Strategies video, and then reconsider your goal.

This is about what it sounds like every year, it’s just that some of it fades off before it’s actually implemented. A lot of it doesn’t, unfortunately. It’s why we spend so much time begging people to consider what they’d pick as an alliance captain. You choose: Robot X with an inaccurate shooter, unreliable climber, and slow gear running; or Robot Y with fast gears and/or a slow but reliable climber, and no shooter? Because if you think you’re going to field a top-tier high goal shooter with the autonomous routines to match and a 7-cycle gear runner and whatever awesome climber, you are wrong.* If you know that’s what you’re going to do, congratulations on all your past success at the walk-before-you-run levels of robot design and I’ll enjoy seeing you on Einstein.

*If you think you’re going to build a sort-of-okay climber and gear runner and floor pickup and on and on, good on you for honest assessment, but why?

Karthik was “terrified” by the number of teams who said they’d do low bar last year. (330 ::ahem:: low bot ::ahem:: world champs)

CD is also populated to a large number of people on established, well funded, and competent teams.

That’s why I conceded a small number of teams will succeed. 330 is one of the elites that win multiple events almost every season, a past World Championship winner. More than 99% of FRC teams are not. I’ll also note that their first pick, the Roboteers, were one of the best high goal shooters in the world in part because they did not compromise their design for the low bar.

Karthik was correct last year, the vast majority of teams who limited themselves to fitting under the low bar while shooting were mediocre high goal shooters, or mediocre Boulder scorers period. You only needed to average 2.66 boulders per robot for a capture last year, yet a capture was rare outside of the top 2 - 4 playoff alliances in the first 3 weeks last year, and more than half the boulder scoring was being done by one robot typically.

If you are on one of those elite level teams more power to you trying it all. My advice is listen to the advice of many experienced FIRSTers and become elite in 2 out of 3 instead. You can win a World Championship without doing it all.

Please never look at how an Einstein robot (particularly an AC or first pick!) sets their design specs as the “correct” answer. Setting your specs to be a world champion powerhouse does not make you a world champion powerhouse. Meeting 30% of those specs is not going to make you 30% of a world champion. In the literal sense, it probably even won’t get you a third of the way there. Just because it worked for them doesn’t mean it will for us mere mortals–statistically speaking it almost certainly isn’t the correct choice. All the winning alliance captains of every other event are not looking for poor 330 knockoffs; they’re looking for your best effort to meet alliance strategy needs.

We have two that we’ve made much bigger priorities, and the third will be extra if we get to it.

This is a fairly close analogy to our decisions.


If you’re using 330 as an example of why Karthik was wrong then you completely missed his point

Personally the results of this poll are disappointing because I thought that one of the best advantages of this game was that there are a lot of creative and focused options for lower tier teams to pursue and be successful.

I would like to amend this with, if your goal is to make it to Einstein someday, someday you’ll have to reach for the stars. It’s just that the stars get a lot closer when you spent the offseason developing a rocketship.

Cheers, Bryan

You really didn’t take away the strategic point made there. It certainly wasn’t that an alliance didn’t need a bot that could go under the low bar to win the World Championship.

I don’t think 330 should have built 330’s 2016 robot

Teams should push their limits, a regional of AM14U3’s driving around playing defense is not one I want to watch.

Some of you are coming across very elitist. Will a team/student learn more from simply assembling kit bot or from trying something completely out of their comfort zone? Yes they should try to do it in the offseason, but if they only do FRC during build season, they should try something new.

Fail early and fail often, just learn from each one. It isn’t a robot right?

I think a team can feasibly achieve 2 out of the 3 tasks rather simply. (in order of simplicity)

  1. Passive gear mechanism
  2. Velcro (if ruled legal) or other type of drum/winch climber mechanism

The aforementioned 2 tasks could probably be manufactured and put on a robot in 1-2 weeks or less by a good amount of teams. This leaves a good amount time to develop a somewhat effective shooter. Now that time could possibly be better spent improving other function of the robot along with driver’s practice. At the end of the day I think many teams find themselves with the resources to check off all three tasks, but I’m curious to see how this translates to competition.

I am on a similar wavelength with Dikshant. A passive gear mechanism can be integrated quite easily. The climber is a bit harder but really again if velcro is ruled legal (Which I am of the mindset this is almost a definite) than all you need for a climber is a drum covered in hard velcro. With all the COTS solutions for Drums from WCP and Competition Robot Parts this isnt hard at all and could be extremely effective.

My team personally is going to be prioritizing balls over gears and gears over climbing. However, I do expect to be able to accomplish everything. Atleast by our 2nd regional ;).

However, we will not even be attempting to build a climber until our shooter as well as gear mechanisms are all atleast somewhat working. Karthiks Golden Rules are definitely something which should be taken to heart. I hear he knows a thing or two about FRC.

Last night, Team 4272 Maverick Boiler Robotics decide we would only be focusing on only two aspects of the game and not all three. As a team, we decided our efforts would be best put towards two high quality (relative to our past) systems rather than three below average systems.

I’m not bashing on teams that try all three, it’s just our team did not think WE could do it and do it well.

There are no wrong ways to learn in FRC (there are ways that align less efficiently with one’s goals). However, I have spent years trying to teach folks that the learning experience that comes from optimizing a simpler system is just as valuable (and often more so because it’s a less common experience) than taking on a bigger design spec. Teams can certainly learn a lot from building shooters and loaders of all types. Teams who don’t invest time in a high shooter can also learn huge amounts about more detailed design optimization that is often drastically undervalued.

Personally, I’ve always found the FRC tendency to value the learning experience of attempting more “elite” bots versus the experience from building simpler ones to be ridiculously elitist. It’s also a culture that generates a lot of common learning gaps like this. I admit we tend to overcompensate in an attempt to get some teams to reassess. I do believe this is a year on which more robots than some other years can do all functions (certainly more than full-color, 30 point climb, floor pickup of 2013), but that doesn’t mean everyone who might be able to automatically should.

Before I start - every team needs to do what works for them, which may or may not be what works for my team. My opinions are my own, I don’t speak on behalf of my team here.

You seem to be setting up a pretty harsh binary set of options here - either building a box on wheels or going way outside a team’s comfort zone. My experience has been that my students seem to get the most learning and inspiration from doing something just a little bit outside of their comfort zone, but still within their reach if they push for it.

Up until about 2013, 1257 was a shoot for the stars sort of team. Every year, the goal was to do everything. Every year, that goal led to chaos. Students bit off way more than they could chew, and that turned into frustration, which turned into anger, which turned into burnout. By the end of build season, nobody wanted to be around anybody else on the team. Nothing much happened in the offseason, because nobody wanted to be there. We were a terrible team - not just in terms of competitive performance, but even in terms of being a group of people who worked together.

In 2014, we started to listen to some of the mentors you’re referring to as sounding elitist. We realized that we weren’t ready to build an Einstein bot - we needed to focus on being able to build a robot that worked at all. We simplified, and we didn’t try to do everything. We weren’t a captain, or even a first pick, but we got picked. Given our team’s recent history at the time, that was a big deal to us. The students had a much better experience on the team when they weren’t overwhelmed by trying to reach too far out of their comfort zone, and they learned a lot more when the robot was simple enough that more of the team could understand what was going on and work on it.

Our goal since then has not been to build outside our reach, but to expand our reach a bit each year. If the KOP chassis starts to become trivial, we can modify it. If modified KOP chassis becomes trivial, we can try designing a custom chassis, starting with one in the off-season. If we know that we can absolutely, without a doubt, get one task done well, we can try for two.

Also, please keep in mind that for some teams, a basic box-bot truly is a challenge. I see multiple teams every year that struggle simply to have a legal robot that they can drive around the field.

Perfectly stated.