Changing the competition ranking process

We like to run three fields for competitions, 3:30 - 3:45 match cycle, and 8 rounds per team, or more. We have at least four referees as we do not break for lunch.

Definitely getting lots of matches helps.

Also, SG3 is probably more of a problem for teams all season long than ranking system.

9 Likes

That is kinda your fault for not reading the rulebook

3 Likes

You want to expand on this for those of us that have no idea what you are talking about?

What are you talking about ecstaticPilot.

Summary so you don’t have to go back and read the thread:

Twice??? I guess you didn’t learn the first time? Most people will stop breaking the rule after the first violation…

7 Likes

Allow me to clarify. The first disqualification was for touching the alliances platform multiple times in the last 30 seconds. This was at no fault of our own though because we had a very light bot and we’re pushed by the other team. The other dq was on our alliance partner and we had not control over it. Just a note, can everyone please quote the statement they are responding too do I can understand what they are talking about, I’ve had like 6 comments in this thread and can’t keep track of all of them.

Maybe you should have read SG3b more carefully and designed your robot and strategy with it in mind.

2 Likes

SG3 changed multiple times over the season, and enforcement of it also changed multiple times. For end-of-season & Worlds, yes - it is pretty clear (if unfortunate) how SG3 goes, but early season it definitely was not.

5 Likes

This. We (roboteers, mentors, GDC) have all lamented the changes this season. It’s easy for March Foster to glance back at September Foster and snicker at the ā€œbadā€ choices made then.

I predict that the ā€œrings under the platform is OKā€ rule will change, I saw robots that could move rings under the edge of the platform to stop adversarial alliances from climbing the platform.

To stray back onto the topic: Early season AWP was a great thing, mid season if you had a partner you could bling out to get 1/2 of the AWP task was a great thing, at the end of the year not getting a yellow mogo in 273 microseconds predicted a loss. I did see a resurgence of "Well that’s nice that you got that mogo and now we are going to RIP IT AWAY FROM YOU activity

I’m still mulling over the average win point showings this year with the advantage that the AWP gave you. I think showing the average under played how much advantage that was early season.

It will be interesting to watch the Worlds rankings as the days go by.

6 Likes

Helpfully, I’ve graphed it over time:

Events in China that posted scores to RobotEvents really emphasized AWP, especially early season. The competition over Christmas break demonstrates a bit of an outlier, but it was not very many matches. Overall, AWP success rate rarely cracked 10% over the whole season.

Individual teams did seem to prioritize it; Freedom Gladiators were successful with their AWP in 66% of their qualification matches, and Centrifuge out of Kansas hit it more than 80% of the time (with 34 qualification matches played).

On the other hand, entire regions seemed to forego AWP almost entirely.

I’m guessing that teams at World’s that want to control their own destiny for Alliance selection will prioritize AWP, whether that means solo AWP or coordination with partners.

6 Likes

Thanks for the shoutout!

I completely agree with this statement. In think the top teams at worlds will be able to rush neutral goals and then be able to consistently complete the AWP task afterward. Maybe even solo the AWP after rushing a neutral goal.

8 Likes

My team didn’t make it to worlds but I’ll be interested to see the autonomous strategies there since we didn’t see much from the teams here in South Dakota.

So, I see that the ranking system could be ā€œunfairā€ from a certain perspective, but I honestly think that the system should stay the way it is.
First of all, correct me if I’m wrong, but I’m pretty sure the schedule randomizes the teams into their matches.
Secondly, qualifications don’t really mean that much, other than for scouting and alliance selection, in my opinion. I’m sure that there are some people that would disagree with me, though. No offense to those who do, it’s just my opinion.
Lastly, the actual tournament finalists and champions shine in the elimination matches.

Again, these are just my thoughts on the matter.

these are good points.

One thing that rankings does determine, is who gets to ask whom as alliance partner. Hence your points about scouting are critical. Indeed the scheduling software randomizes alliances during qualification rounds.

EPs can do their best to get more rounds in qualifying matches. Teams have their routes - win more than others (leave no doubt 1v2 or even, for conspiracy folks, 1 v3!. More importantly, do a lot of networking/scouting during event. Yes you can win all matches, but not sure your first choice for alliance partner will be willing if you have not made connections with other teams.

2 Likes

Actually qualifications matter a lot.
It will determine where you will be ranked and hence how much say you have in selecting your alliance for eliminations.

And for teams that are not seeded, the alliance selection is practically out of their hands, i.e. they can’t say no even if it is not their ideal alliance partner.

7 Likes

an extra way to get another wp for completing something during the match would be cool to reward teams for doing well or something impressive in quals (like rp in frc). like double parking in quals and getting rewarded for it with an extra wp would be sick.

3 Likes

Or maybe we should reward teams more for winning- for example, rewarding teams with 3WP for a win, making AWP matter slightly less.

Then what would you do in the case of a tie? Everybody gets 1 1/2 WP??

Why not? The number doesn’t have to be exact, and because WP is averaged on most tournament rankings as far as I know, this wouldn’t cause any major problems.

Also (and I know this isn’t soccer) but they use 3 points for winning, 1 point for a tie, 0 for a loss. Why can’t this work?

1 Like

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.