Tournament Structure changes

This is a thread to act as an offshoot of this thread.

This thread is designed to discuss changes to the tournament structure to make morally questionable behavior less beneficial.

Initial ideas are (literally listing all the ideas I have heard over the years)

  1. Allowing teams to decline and accept alliance selection as they wish

  2. Overhauling the ranking system so that rankings accurately portray teams competitive ability

  3. Allow the first seed to pick both of the members prior to moving on to the second seed

  1. 8th seed gets first second pick then 7th, then 6th… ect

Edit: This makes 8th seed less of a death sentence. Also potentially ends people trying to be about 17th place. Sorta mixes everything up without taking much away.

First, I’m okay with how the picks currently work. It is difficult to arrive at a substantially different system that will still reward higher ranking, while allowing a realistic chance for the underdog.

That said, this suggestion (which has made the rounds for a while) is worth examining.

I like it, it would make for more unexpected results. As its stands the first place alliance wins almost always( @Robo_Eng_13 has the exact stats.), this would certainly disrupt that.

Dropping low score of every 4th match

Two robots per alliance and more alliances is another suggestion

Another suggestion I’ve seen is to alter the bracket or tournament structure. Currently, in the quarter finals, alliance 1 plays alliance 8, and so on, down to alliance 4 v 5. Restructuring that so the the alliances play their closest alliance each time would be interesting. You can get a feel for the likelihood of that changing things by comparing the 1 v 8 statistic with the 4 v 5 statistic. 4 v 5 is much more evenly spread.

There are other schemes worth trying. We have a tremendous number of events each year, It would be interesting to have RECF assign some tournaments to a different scheme, and look at the results after a season. Yes, I understand that means we’d all be participating in a grand experiment. But, as I’ve said before:

@Karthik is playing games with us. They could just be doing it in a different way.

So serpentine like frc?

Wouldn’t a team want to be in 24th place to get picked by 1st in the second pass.

maybe, but most tournaments are too small to make that a great idea.

So @Karthik, @nallen01 and I crunched the numbers and it worked that frc serpentine still had 1st seed win 60% of the time. This is a little lower but still pretty similar to vrc which was mid 60s.

do people sand bag to get in the 20’s for FRC?

Nope, because the top seeds are good with scouting and would notice. And that would go directly against how EVERYONE wants to play and be played with.

You should look up Karthiks FRC presentations on doing strategy, design, scouting, etc. It will do you an amazing amount of good for your teams.

To achieve this goal, just replace the elimination rounds with more qualifying rounds. After all the qualifying rounds are over the #1 team is winner. This will not be perfect or popular, but it eliminates any benefit in sandbagging or lying about your robot.

60% seems like a reasonable number, especially because the 1st seeded robot is the most likely to be the best robot at a competition. Do you know how often the other alliances win? Is the majority of the remainder won by the 2nd seed? Or is it more evenly distributed?

This makes sense to me. I don’t really understand the rationale between suddenly turning a 2v2 competition into alliances of 3. The third pick typically has very little effect on the results of the tournament. They also get a free qualification to the next level, even though they typically did very little to earn it and when there are often much more deserving teams (usually in the second seeded alliance).

We struggled all season to qualify for state. Tournament after tournament we were runners up. We had to watch all those third picks qualify before us, and we were a much better robot. Luckily our skills scores proved that, so we were still invited to state.

I really don’t like this idea only because that gives the 8th seed a major advantage. The whole purpose of trying to perform well in the qualifications is to get yourself that advantage of being in first or second place. In the end, the very best robots should win, and if 8th picks first, that isn’t really happening.

Again, the point of worlds isnt to see the best teams compete, its to reach the broadest audience. The more people that are exposed, the more future engineers there are, which is the point of vex

What if the tournament structure stays the same, but participation awards are given to each team (e.g. Skills participation, tournament participation, judging participation, etc.)?

How would that help? Then everyone would qualify for everything, and that ruins the point of qualifications

I believe the idea I propose below will help prevent most of the participants tanking on purpose during qualification rounds. It may also help to prevent students from lying about their robot’s operational status due to the importance of being in a higher seed.

I suggest the REC and VEX look at a 12 seed tournament with 2 teams per alliance. This will still allow 24 teams to participate in the tournament. However, the 1 through 4 seeds will have an automatic bye to the semi-finals. The 5-12 seeds will have a 1 match qualifier in the quarterfinals to advance to the semi-finals. This is similar to a wild card game played in the MLB. Not only does this save time with having 12 seeds, but will also help with keeping motors cool with only 2 teams per alliance.

Feel free to comment and make suggestions!