Picking for eliminations alliances

I saw another post recently about alliance selection and I had a different proposal that could work in both the district and regional models.

So you watch and event and more often than not, seed 1 pairs up with seed 2 causing the lower seeds to move up and become captains, then you have two “power house teams” together to win an event, and sometimes this leads to something like seed 8 getting immediately knocked out.

I propose 2 changes. A change to picks and how the finals bracket works.

  1. With picks we make it so a captain cannot pick another captain, that way the top teams can’t rig the bracket so only they have a chance, and the lower seeds like 7 & 8 have more of a chance and aren’t screwed from the beginning.

  2. I propose we do a different bracket, such like what Magic the Gathering does for its major events.

This bracket benefits both the top and bottom of the alliances. The seeds that didn’t do as good have to do more of a gauntlet, but they aren’t against the top seeds immediately and can get a chance to make it to a semi final or even finals matches. It also helps the top seeds, because they get rewarded for doing good and making a good robot by not having to go through the quarterfinals and essentially getting a “round 1 bye.” They also lose the risk of losing in the quarter finals with this.

Then the question arises about district points for the “every match you win, you get 5 district points.” I haven’t thought about this one much, but one quick suggestion if we use this bracket style would be that a default 10 district points are given being seed 1 or 2, since they get to play 1 less round, and quarter final stage 1 gives none, that way it also make district points more scarce in playoffs and pushes team to try harder or “show more of their strategy” in qualifier rounds.

Thanks for listening to my TED Talk

  • Matthew McInnis
1 Like

Can you explain further why the change to alliance selection has to be linked to this change in bracket construction? How does this system disincentivize teams from intentionally ranking outside the top 8?

3 Likes

I wouldn’t say that they are linked, just from what I see to be good ideas. Penalties could be put in place if it looks like a team is clearly throwing a match and not trying their best, such as in sports.

Change number 1 is a huge problem when seeding doesn’t accurately reflect the top robots of an event, as it very often does not. It will incentive teams strongly to throw matches to avoid being a #8 captain instead of a #1 pick, and it generally does way more harm than good.

Change number two is an enormous benefit to the top seeds of an event and a huge penalty for the bottom seeds. It doesn’t matter that they are facing theoretically worse opponents (ignoring for a sec that it is often not the case) - having to do twice as many matches as the top seed, with faster turnaround times, will be an enormous disadvantage.

These are solutions searching for a problem.

18 Likes

I see the problems with change 1 and I’m thinking of ways to work around it, maybe it is just a bad idea. And with change 2 that is the point, it benefits teams for doing well, and it makes a bigger challenge for teams that didn’t play so well. To combat turnaround times give an extra timeout to the lower seeds, or just don’t rush people to get through as many matches, give teams an extra couple minutes in the pit.
If the team isn’t prepared to play their next match then they either need to figure out how to get ready for a turnaround time. Whether that is build a more rigid robot or add a couple extra air tanks so they don’t have to compress as often between rounds. Keep a “small pit” close to the field during playoff matches so you don’t have to run around to your pits after every match. Keep a spare battery on hand.

The only I would say if you are a higher seed like 5-8 you may not have enough batteries and time-out would likely be more present to try to get full charges on batteries before matches. I personally support the model but I have issues with that and events would definitely take much longer
Also first alliance and second alliance have more of a chance to work out kinks and perfect autos so I guess that’s just an advantage

If you are a top seed you most likely already have your kinks finished from the quals matches you’ve played.

The battery issue does not wash away though

false. the QF is a time for the alliance to gel and get going on their strategy. giving the #1 and #2 seeds a free pass to the semis allows them to play less which is nice but it also disadvantages them in that the teams they face will have had several matches to get their strategy down and meld as an alliance.

6 Likes

A one-match turnaround time for an entire event is brutal on any team. It’s not just a problem for unprepared teams - even highly prepared teams will be in a stressful rush if absolutely anything goes wrong, if there is any change to strategy, etc.

I don’t see why lower seeds need to be given twice as much work to do as top seeds to win an event. I know there is some benefit with extra mesh and gel time, but I would much rather have a half hour between match 1 and 2 (and match 2 and 3) than a bunch of extra high pressure practice, wear, and tear that #1 didn’t have.

1 Like

It seems like most of the people that don’t like the current elimination format are areas that have a few teams that are a clear step above everyone else.

Changing the elimination format seems like it’s addressing the symptom, not the root cause.

8 Likes

As a 1 or 2 seed I would LOVE to watch the QF’s and see my eventual opponent “get going on their strategy.” I would know exactly what to expect, and they would never have seen my alliance play. If I picked my alliance wisely I should have drive teams that don’t need 1 more match to work out kinks.

Am I wrong or did 525 not work in your 1st QF match in NL this year… How’d that match turn out for you?

2 Likes

Lol you are correct and it was our only loss, but that isn’t really applicable to your post or my reply. No where in your post did you mention mechanical/programming failures of a robot. Your post was about strategy and having time to “meld a strategy”. When 525 fixed their code issue the following match and was able to function we did just fine. Whether that happened in QF or SF is irrelevant. In actuality if we didn’t have to play until SF they probably would have had the time to fix it before the match.

I’ve said this before, but more offseason events need to mess with the playoff structure as a proof of concept before FIRST jumps into big changes.

1 Like

IRI or Cheezy Champs would be perfect places to test these most likely

1 Like

Then the second powerhouse team would just do nothing and derank themselves as best as they can to prevent being a captain for alliance 1 to pick them.

This ruins the incentive to be any rank but #1. If two teams which are best friends for life go to the same event and one ranks #1 and the other one ranks #2, then I don’t see how it’s fair to force them to be split up into two different alliances for playoffs. How is it fair to pit them against each other? They both worked just as hard as any other team to build good robots, and yet you’re punishing them.

4 Likes

It not necessarily intentional throwing, but teams would definitely prioritize having good scoring numbers for scouting over getting more rp, and with how random match schedules are, the second best robot at the event could likely finish outside the top 8

With the current system, it’s possible that the 8th seeded alliance is actually the second best alliance in the entire bracket, but they get knocked out in the first round by the best alliance who is seeded 1st. The same thing could happen to whoever the winner of the 4th/5th QF match is.

Before we propose any changes, I think we need to answer the question of; “Do we want to design a bracket that avoids that scenario or it just the disadvantage of being seeded low?”.