Build strategies for a post-bag era in FIRST Robotics

The point is, it already expands to fill the time allotted. Some teams just are not taking advantage of the time, or cannot afford to take advantage of the time. Removing the bag removes the “can’t afford a second robot” from the equation.

We meet 4 days a week pretty much through the “World Championship”

The value of eliminating bag day is not in the extra time it gives teams before they compete… it’s access to the robot AFTER they have competed and BEFORE they compete again to fix small issues, etc…

the more kids that get to play in matches where they are truly valuable the better.

But it’s not about winning, failure is also inspirational!

I’m fairly confident there are teams that ‘meet’ fewer than 6 days per week, but the key mentors and students are still working (cad, discussing strategies, ordering parts, planning meetings…) the other non-meeting days.

Anyone who claims to know definitively how teams will act in a post-bag schedule is talking out of their behind. It would certainly change the dynamic greatly, and many teams would react very differently from one another. Many teams do need the hard deadline to stop and take a breather. Many teams need to be “saved from themselves.” Some teams are already working at a fever pace past that deadline. Some teams may find a way to pace themselves better over a long period, but some other teams will be left in the dust by those who can maintain a build season schedule over 13 weeks instead of 7. Teams that build practice bots and attend multiple events are almost certainly going to have a very different viewpoint on this than teams that build one robot for one event. Teams led by a group of engineers will have a different viewpoint than those led by one or two teachers or parents. Community teams will have a different view than school-based teams. Teams with access to practice fields will have a different view than those who do not. A team with their local competition in week 1 will feel differently than a team with their local competition in week 5. There are many different scenarios that will impact various teams in unique fashions.

Using FTC and VRC as an analog is flawed. Not only do those competitions have a different perspective of “stop build day,” they have a wildly different perspective of “start build day.” VRC has a game out literally 365 days a year. They announce the next competition during the championship of the current one. There are months between the game announcement in April and the earliest official tournaments in August. Teams can opt not to compete for the first time until after the calendar turns into 2018, if they so choose, giving them 8+ months to work on their robot before competing. State/provincial championships don’t occur until February/March (10-11 months after kickoff) and the VRC championship event is literally a full year after the previous game kickoff. FTC is more condensed, as their kickoff isn’t until September. The very earliest FTC events (October 22?) are roughly the same gap between game reveal and competitions as FRC, but many don’t really start ramping up until November, December or after the new calendar year. And once again, we’re still talking a 7.5 month span between kickoff and FTC Championships.

I have these all the time. We have to rigorously test prior to delivery, and that involves scheduled use of our test facilities. While it’s possible to reschedule that test time, it’s a hassle that impacts multiple programs and is saved for a last resort. As a result, we have hard cutoffs established by our testing time, and barring any show stopping errors being found, its’ hands off once it goes to into testing.

I think this is only partially true. For teams with access to a nearby practice field, sure. For teams that don’t have a practice field or realistic field elements, it will be dramatically less beneficial. One of my major hang-ups with the “Stop the Stop Build Day” paper was the implicit assumption that competition time could be used as a proxy for any robot access. In reality, many teams see huge boosts the first time they can test their robot on a full-sized field with realistic game elements. Driver practice and autonomous testing on wooden proxy elements inside the limited space of a classroom, hallway, or library can only go so far at preparing a team for competition.

Build Season Schedule Questions

Spreadsheet/Heat Diagram linked from above

The mode (15 of the 48 teams reporting) of that poll is 20-24.99 hours. Range 11-51 hours.

Edit: median 24 hours, mean 26.3 hours.

My impression is bag and tag isn’t trying to mimic real world engineering, it is trying to mimic real world product cycles. Bag and Tag is the point in the real world when the end product is released to the customers for a given product cycle. At that point engineering is too late to develop a new product for the cycle and can only make revisions to the existing product.

Whether that’s a good thing or not for FRC to be doing is open for debate but that is what I think it is trying to do.

IF there is a change to B&T rules, when do you think teams will be notified of it? Would it be at kickoff or earlier?

Earlier. No question.

And highlighted at Kickoff.

Last year there was a survey on what to do from HQ during the fall, as I recall. They notified teams early that there would be no change.

They have already released bag day for 2018, so I don’t think it will happen this year, I could be wrong.

I would hope teams would be notified of this change before registration begins. I think there are some schools who would be more likely to take part in FRC because of the open build season, and I also know there are some teams who would opt out of participating because of this change. Letting teams know in the summer would allow teams to make an informed choice at registration.

Last year’s discussion on the FRC Blog about it said, in September and October, that an an announcement would have to be made early if they were going to change anything. Specifically, on September 7, Frank said:

We would want to publish information relating to any substantial changes to the system many months in advance, hopefully not later than early- to mid-summer of the year before. So, for example, if we were intending to make major changes for the 2018 season, we would want to announce them in early- to mid-summer 2017. We know that teams order their lives around the build season, and it would be unfair of us to make any significant changes with less notice than that.

And he reinforced this on October 21:

Any significant changes we would make would be for 2018 or later, as we want to give this topic very careful consideration, and announce changes (if there will be any) long before the season starts so teams can make plans.

Based on that, the clock is ticking… midsummer is, technically speaking, the week-ish around the summer solstice in June. We’re well past that at this point. I’m not saying changes can’t still be announced, but I also think it’s highly unlikely to happen after event registration starts next month. The more likely scenario, I think, is for an announcement like this to happen at or soon after champs.

That statement wasn’t about top teams, but teams with low $$ resources and relatively high human resources. This may describe a few top teams, but not many.

Do you also think that if they cleared the track that the fastest hurdlers wouldn’t go faster because they don’t break stride for the hurdles now?

The teams with the resources and capabilities to quickly iterate and reverse-engineer designs already do so. Opening up the robot post-bag to every team will not allow these top teams to do anything they weren’t doing already.

Absolutely!

Why?

No “defined end date”.

It’s a lot easier to interest some folks in doing hard/stressful/tough stuff if you can convince them that there is, in fact, an end date. And believe me, I know what I’m talking about.

At the risk of over-stretching the metaphor, imagine if the top 25% of runners were 15 feet tall. They don’t even notice the hurdles, or they do but they barely have to adjust for them. So for the top teams it’s a straight footrace, everyone else it’s a hurdles event. At that point, why not get rid of the hurdles? Then it’s a footrace for everybody.

Because the lead teachers/mentors involved with these teams don’t want to commit any additional time to the team beyond stop build day. The current format allows their season to stop in late February and only resume for a couple of days of competition(s). By moving to an unlimited build season, these adults would feel pressure from their teams to continue working beyond any self imposed deadlines designed to replicate the stop build day experience. It’s not something they want to deal with.

Sure there is, it’s the end of competition season.

Nothing is stopping a team from defining their own “end date” if they wanted to stop sooner (or later), we’re basically on the honor system right now as it is so it’s not like there’s really anything preventing a team from working for more or less time. Your teams schedule is what you make it.

Don’t teachers already get this in the form of “we should build 2 robots so we can keep working and be like successful team X”?
I’ve worked with teams that have started their build season in week 3 and only meet 2 days a week (and their resulting robots were about what you’d expect for that kind of schedule), but it was a decision they made to limit their own schedule (or more specifically, the teachers), not one imposed by FIRST. I don’t know, that reasoning just seems like a cop-out to me.