[robocup-small] Tournament Format

Christian Lobmeier pardinensis at gmail.com
Wed Aug 30 18:27:03 EDT 2017


Hey everyone,

I hope everyone had a good time at this year's RoboCup and maybe a nice
trip across Japan as well. I know at least some of you did ;)

Anyways, since the large majority should be back home by now, I want to
discuss a topic with you all that I kind of was involved prior to my
nomination as OC: the tournament format.

For the first time, the small size league featured a novel format with a
lot more games played in total. In addition to the traditional single
elimination format with lucky loser games there were several
sub-tournaments for the dropped out teams. This was highly requested in
Leipzig and as far as I witnessed, also greatly appreciated this year.
However, because of the (unexpected) success of our team and me being our
main software guy, I didn't really have much time to speak with members of
other teams about this topic.

Therefore, I'd like to hear about the things that went well and the things
that need to be improved. I think I'll just start with my impressions:

- The increased number of games was a great improvement for the weaker and
especially the mid-tier teams. I still have strong memories of the
situation of NEUIslanders last year (They had 3 very strong teams and 1
team without moving robots in their group, placed 4th and dropped out
without having played a single game against a similarly skilled opponent).
Not even playing an interesting game should not be a thing on an event
where you have to spend thousands of dollars to participate.

- There were concerns that the increased number of games would be difficult
to schedule. I had the feeling that it was not really a problem (also
because of Toshi's great work). One minor problem was that the technical
challenges were played in the late evening when no audience was around.
Maybe this was a blessing this year, regarding the abysmal game play in the
mixed team challenge.

- There was also a concern that the large number of games may be too
stressful for the teams. I have to say that I only know that ER-Force had
some troubles getting the batteries ready for the finals because the semi
final and the grand final were almost immediately consecutive. Did you also
have some issues related to the high number of games or were we the only
ones?

- Prior to the event, I voiced the concern, that the sub-tournament games
would maybe not be taken as seriously as the main tournament ones, since
the difference between placing 12th or 13th for example is negligible. On
the RoboCup I did not feel that this was an issue at all. But I'd like to
hear your opinions.

- There was a minor flaw in this year's tournament plan: Two teams of the
same group faced each other again in the MSF games (MRL vs Tigers and
RoboDragons against STOx's). Situations like these can and will be avoided
for next year's plan, regardless of the format.

- One thing that did not change this year is that the seeding has a huge
influence on how far a team will advance in the tournament. Having to face
a rather weak RoboDragons (please don't take it personal) instead of
CMDragons like last year did make a significant difference for our team. I
already have some ideas on how to tackle this issue (like the already
proposed double-elimination format) but this is something for a future mail.

I'd really love to read about your opinions on the current tournament
format. My goal is to find a compromise that satisfies everyone: the strong
teams, the weaker teams and the audience. And since this became kind of my
hobby horse, together with you I want to develop the fairest, most
enjoyable way to decide who will be the next champion of the RoboCup SSL.

Best Regards,
Christian (new OC / ER-Force)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/robocup-small/attachments/20170831/f60ef564/attachment.html>


More information about the robocup-small mailing list