[FRC Blog] - Regional Advancement Task Force Update

[bdaroz]

@jtrv

This reminds me of conversations many mentors have that “2 champs isn’t inspiring,” which I always found nonsense. The students don’t know what 1 champs is like. They are still massively inspired by 2 champs. My first championship experience was 2 champs and it pulled me back into the program.

And you came around to what I was talking about when I said HQ was pulling the wrong lever too far.

Forgive the excessive hypothetical, but lets say this system doesn’t get put into place, and Dean Kamen’s wish comes true and every FRC team recruits at least 1 more FRC team and we get… 125% growth (we know there are overachievers out there).

That growth means at least doubling the number of events, and for the sake of argument lets say it’s even between district/regionals.

You’ve doubled the number of winners, EI winners, Inspire winners, maybe RAS winners too… Regional advancement has to get very weird. (I couldn’t find exact numbers, so take these with a large grain of salt). If we go from 3300 teams to 7425 teams, you’re going to go from ~18% of teams attending champs to ~8%.

FIRST’s reasoning for keeping champs in Houston included some verbiage on the size of the facility (might have been the announcement before last, I can’t look it up at the moment) or struggling to find appropriate facilities. 600 FRC teams + who knows how many FTC and FLL teams in a single venue is a challenge.

I’d largely argue that the team count at a single (program wide) champs event can’t go much higher. So how do you grow the program and maintain the reach of the champs experience?

Expand Champs to multiple events.

Gee, sure would be nice for the IB teams if another Champs event was held the week before Houston… Would be nice if the LGBTQ community had another option…

Lots of issues and problems get solved or severely mitigated by bringing back 2 champs.

HQ said it was hard on staff doing two champs back to back. You know what it’s like doing 2 events back to back, and so do I. It’s… hard.

Robotics is hard too. Getting teams together, funds together, designs and ideas together, it’s hard too. We teach the kids to adapt and overcome - to think outside the box, and to tackle the hard tasks.

I’m just asking the same from HQ.

[GKrotkov]

@bdaroz

No one has challenged my position that giving something to only a part of a “winning” alliance diminishes the contributions of those that don’t get it. That’s true whether it be points, an automatic bid, or anything else.

Assuming no backup bots are called, all members of the winning alliance earn the same 30 advancement points for winning the event.

[BordomBeThyName]

Right now, in non-imaginary 2025, there are about 294 champs slots and 68 regional events. Each regional gets no more than 4.3 invites. Full stop.

I would love if we could find a system where the entire winning alliance, the finalist alliance, and the winners of Impact, EI, and RAS all get to attend champs, but that’s 9 teams. If a unique team won each of those then we couldn’t even fit them all into two champs of the current size. The only move we have available to us under the old system is to keep fiddling with wildcard orders and deciding which of the above groups do and don’t deserve to attend.

A big benefit of a points system is that instead of wholly including or excluding any of the above groups, we can invite a meaningful percentage of each group.

Another big benefit is that it can scale to any size or number of championships. If the number of champs invites gets bigger, the number of direct event invites and regional pool invites can also get bigger. Those invite counts getting bigger will also increase the percentage of RAS/EI/2nd pick teams that get to attend, and those percentages will scale up together.

A final relevant benefit is that it’s never impossible to re-scale the points to balance invites differently.

[Lidor51]

In order to uphold one FIRST’s core values system, it’s essential that the points systems of districts and regionals are (very) highly correlated.

[waialua359]

@jtrv

There isn’t and can’t be a metric for this, but you can evaluate it holistically. 5419 built an immaculate robot this year for example. It should have been at champs. It was better than 90% of the program. I don’t need a metric to know this. They drove it better than 90% too. But they simply exist in the wrong part of the country, so they didn’t get to go to champs. Inspiring.

There is a part here that is missing in these latest conversations.
Teams that are in a regional system are free to sign up for any regional, the same as any other team (outside of waitlisted preferential treatment by RDs). 5419 went to the Canadian Pacific regional in 2023 outside of their local area.
Expenses are a hurdle. Yes, we know. There are inherent issues in every region that are both unique and similar depending on what lens you are looking at. We also “exist in the wrong part of the country.”
Just today, I come back from Chezy Champs, and find out our batteries are stuck in SFO and they wont send it back home. So unless I find other options, we are out of luck on $800 worth of batteries, an additional couple hundred on connectors, and a special battery box we made last month using $200 more of materials and fasteners. Its unfortunate, but we are stuck with yet another cost.
The Championship model has not changed over the years, just the size and venue. If the current model continues and with FIRST growth, we need a larger capacity or more than 1 Championship event. Its a pretty simple solution, but at what cost to FIRST, time and effort?

[GaryVoshol]

@Lidor51

it’s essential that the points systems of districts and regionals are (very) highly correlated

They need to be similar - they are - but not necessarily the same.

A big difference is FI and EI award points. In districts, the points are low; in this regional system they are high. But DCMP’s have the capacity to invite all their district event FIA winners directly. Post-2025, the regional system will not do that, so the points need to be much higher to make up for that.

[billbo911]

First off, it is obvious I didn’t read all the details, MY BAD!
Thanks for linking this section, and yes, it does answer my concerns. THANKS.

[Zaque]

I’m gonna be honest, your interpretation of these changes makes me think we read completely different documents. If you don’t want 2nd picks singled out and receiving less than the captain/1st pick, this new point system perfectly solves that. All teams on an elims alliance receive exactly the same reward for their performance (baring backup bot changes). Winning an event has no inherent connection to advancement; just ask district teams, where event winners still have to qualify for their DCMP based on points.

[Skyehawk]

@Zaque

Winning an event has no inherent connection to advancement; just ask district teams, where event winners still have to qualify for their DCMP based on points.

But it does (for districts)… I mean it’s not an auto bid or anything, but winning the event (as cap/first pick) means you just need to basically show up to your other event (if you aren’t already over the threshold) for states (at least in Michigan)

[Zaque]

@Skyehawk

as cap/first pick

So it’s not inherently related

[Skyehawk]

Well i guess… Looking at the alliance holistically you are correct, 1/3 of the teams are not effectively at the states threshold.

[Herringbone]

@jtrv

How can we accept that you can build a better robot than 90% of the world but say, in the same breath, “but you’re not good enough to go to champs” or “your team is simply in the wrong state, if you were 20 miles east you’d go to champs”?

pain.

(We were the 10th best bot in California and only lost out on going to champs because our bot got flipped at San Diego in the losers bracket of the playoffs)

[bduddy]

In what way? A format with lower variance almost always benefits the stronger participants in any competition. Maybe you have a different idea of the “middle tier” than I do but for me the “middle tier” just had one of their main ways of making championships, being the second pick of a winning alliance, taken away.

[waialua359]

@Herringbone

(We were the 10th best bot in California and only lost out on going to champs because our bot got flipped at San Diego in the losers bracket of the playoffs)

Lower bracket.

[MARS_James]

@Herringbone

We were the 10th best bot in California and only lost out on going to champs because our bot got flipped at San Diego in the losers bracket of the playoffs

You were the 10th best by EPA but that does not inherently make you the 10th best robot.

I watched said match and though it is hard to see it does not look like you got “flipped over” it looks like you tipped over, there is a big difference in these terms as one implies someone else’s movement and action made you fall the other implies your action and design did it.

An example of flipped over:

Vs tipped over:
link

Both are 1902 both result in them being unable to continue but one puts blame on someone and one does not.

The fact a red card was not awarded by the referee who had a very clear view of the interaction tells me this was also a case of tipped not flipped

This also assumes you would be able to beat Alliance 6 in the rematch which is also not a guarantee as when left undefended way over performed EPA predictions.

You also went to another regional after San Diego and did not qualify. Meaning the “flip” in San Diego did not end your season.

Be very careful making flat out statements like this, always quantify them and think about the terminology you use and the implications behind it.

[waialua359]

I saw this occur live while waiting in queue. I think Eric was the referee on that side, and he used the “its safe” baseball umpire motion when the interaction occurred.
If it’s not Eric, excuse my bad memory due to old age.

[MARS_James]

If we ignore for 2 of the last 3 years the 2nd pick of winners already did not qualify and you are looking at all tiers being the same size than you are correct.

The average regional is size last year was slightly above 45 teams so 5 tiers would put the middle tier as the 19th-27th best robots which in an ideal world would be 6th alliance 2nd pick through 3rd backup robot. For teams of that caliber they did lose their best way to try to qualify

However I have never thought of the tiers as equal size and the fact that alliance 1 has such historical dominance implies they should not be as they are not the same as the other alliances (in an equal tiered system Alliance 1-5 should all be roughly the same)

My 5 tiers have been:
1st & 2nd best robots (Alliance 1 Captain & First Pick)
3rd-6th best robots (Alliance 2 & 3 Captain & First Pick)
7th-14th best robots (Alliance 4-7 Captain & First Pick)
15th-24th best robots (Remaining Alliance Selections)
25th best & On (Does not make Elims)

My logic has always been that a 3-2 advantage the tier below could win but 2 tiers below would not:
An alliance of 1st 2nd and 14th best robot most likely loses to 4th 5th and 6th best robots
However
1st 2nd and 24th beats 12th, 13th and 14th.

In my type of system the “middle tier” in an ideal world (ranks are perfect, alliance selection goes by rank and better alliance always wins) used to have no chance of qualifying for worlds, and even the second tier had a 50/50 at best of qualifying (these would still be part of your “top tier”). Meanwhile my 4th tier (roughly equivalent to your middle tier) used to have an 88% chance to qualify if they were picked late enough. This seems wrong.

To put it another way:
Under 2024 rules the range of chance to qualify by each tier

Equal tiers:
Tier 1: 0%-100%
Tier 2: 0%
Tier 3: 0%-88%
Tier 4: 0%
Tier 5: 0%

My tiers:
Tier 1: 100%
Tier 2: 36.7%-54.1%
Tier 3: 0%
Tier 4: 0%-88%
Tier 5: 0%

While under 2026 Rules:

Equal Tiers:
Tier 1: 13.6%-100%
Tier 2: 0%
Tier 3: 0-36%
Tier 4: 0%
Tier 5: 0%

My tiers:
Tier 1: 98.1%-100%
Tier 2: 65%-95.1%
Tier 3: 0%-36%
Tier 4: 0%-36%
Tier 5: 0%

This to me thinks that FIRST vision of tiers more closely aligns with mine than all tiers being equal

[nobody5690]

I’m going to add on to what you’re saying here. This is a new thing and it may not be perfect yet but whats to say that in a few years we won’t realize a certain group is underrepresented and adjust the points to match?

The good thing about this system is that it offers much more granularity. We actually can turn the dials to adjust what’s fair. What we have now may or may not be the right balance but it sure beats what we had before, and it has the framework to be improved iteratively.

[Herringbone]

I wrote an edit too fast and didn’t reread it. So it is my fault for poor wording but there is a deeper story behind me saying flipped which I’m not gonna get into right now.
Let’s hope I don’t have to edit this one