Forming of starting groups (Closed for comments)
Comments about this discussion:
Started
I took part in the X-Style competition at UNICON 17 and I did not really like how the starting groups were formed. The groups were determined by the organizers prior to the competition day by putting the good riders (according to their opinion) into different groups and randomly assigning everybody else to the groups. Moreover, there was one group for riders under 15.
I don't see the point of putting all riders aged under 15 in one group. Either we need a separate competition for juniors (which I wouldn't suggest because most of them can't do that many tricks yet - it would be really boring to watch) or the young riders have to be treated like everybody else. There are some really good young riders that can definitely make it into the finals and the ones that are not that good yet do not need to be in the 2nd round.
I definitely get why the organizers wanted to put the good riders in different groups. There might be a chance that they will all be in one group if riders are randomly assigned to groups. On the other hand the probability of that event is quite low. Since three riders make it into the 2nd round, the chance that really good riders get eliminated in the first round is not that high. So maybe it's not necessary to put the good riders into different groups manually. But if it's done then it should be done according to rules - for example putting the three riders that won medals at the last UNICON into different groups or putting the riders that made it to the finals at last UNICON into different groups (and applying the same principle to nationals and regionals).
Moreover, I observed that many riders signed up for the competition and did not show up at the competition day. Therefore, the size of the groups was really different. This was especially a problem regarding judging. Since the riders judge each other, we lacked judges for some of the groups because people did not show up. I think it would be best to get all riders together before the start of the competition and then to determine the groups (either completely randomly or by applying a reasonable principle to get the best riders into different groups). Everybody who is not there should not be able to enter the competition.
That are just my thoughts - please comment ;)
Comment
I agree that it should be random. I don't think that there will be issues with top riders getting eliminated too early because of what Marie said.
Also, since I was the person who used the computers to randomize for freestyle, I can assure people that it's super easy to do on the spot and would take no time, you just press a button. So from a logistics stand point it should work to wait until the day of.
Comment
For Unicon we separated the young riders into one group, so they could compare themselves being the juniors (within their group) while still have a chance to become champions (and oko made it to the final round). The rest was randomized, we only swapped Nina Herzog, because she asked for it, so she was able to start in another competition.
However, there is still a chance that all the best can be randomed into one group, so the problem of a niveau distribution persists.
We do have some other solutions:
1) You assign your level group before: A, B, C, D and distribution takes those into account. For each competition it must be defined what A - D means (giving skills that fall in either group)
2) Use something like an ELO ranking - but is hard to establish at an internal level, would work in germany... maybe. Needs to be tested once we have enough data
Regarding distribution at competition day. This whole process takes tooks too much time, means loosing too much time for the actual competition (unless you want to wait ~2 hours). To have a plan for the first round, who is in which run and judges which round is super efficient for riders and organizers (We couldn't have done that last unicon, because basically X-Style was unknown to people).
Riders not showing up is sad and a problem. Some have been reported injured, which is the only acceptable excuse. However, how to deal with riders not showing is a topic for the main committee.
Comment
Note: X-Style was designed to be resistant against the problem that good riders can be eliminated early. This is achieved by letting the same number of riders advance to the next round as we want to award in the finals (currently 3 riders). This means that the end result CAN NOT be affected by a different order of riders in the starting groups. It ONLY makes a difference for the audience. This means that a random starting order is not really more fair than a chosen one by the organizers. I personally want to avoid age groups or senior/junior competitions. Everyone should have the chance to become an overall winner. Maybe there is another solution to still award younger riders.
Comment
Felix, "the end result CAN NOT be affected by a different order of riders in the starting groups" is only true theoretically with an important premise. That premise is that each rider performs equally well each time.
If the top three best riders all start in the same prelim round, all three of them must perform well each time in order to move on. If they all start in separate groups, the stress to perform at the rider's best in each round is not as high.
I'm not sure how I feel about setting up the finals but I wanted to make sure that you understood that it does make a different. Perhaps X-Style is resistant to early elimination but it is not immune.
Comment
Additionally, while it theoretically doesn't affect who gets the top three spots, it does affect who makes finals. If you 5 best riders all start in the same starting group, then 2 of the best riders won't make it to the finals, it's impossible.
Comment
Good point, Scott. Patricia is also true, but can we do anything about that. We don't know who will make it to the finals in advance. That's why we need the competition. I'm also not a big fan of riders assigning levels to themselves to affect their starting group. There is too much room for manipulation. ELO ranking is also not suitable. First, we don't have 1 vs 1 competitions and second, If not everybody is competing regularly against the same people, we have another room manipulation. In the end, randomizing gets attractive again.
Comment
Audience will come for the finals and we should get the best riders there to please them, what alternatives we haven't thought of?