[WCA Regulations 2014] Mean of N for FMC

[Erik]

Because having a different format would also mean that you have a WR for that format. i.e. there would be 2 rankings for FMC. All 3 results of a best of 3 can be submitted to the main rankings whereas a mean would mean that your mean would count for a seperate list which not many people would have the chance to be on.

It does not matter if it is preferred or not, the fact that it's possible means there would be another ranking system for FMC. Therefore they are not really different points, the effect would be the same.

 

[okayama]

Of course I understand this change makes another ranking for FMC, but is that itself so bad idea?
I think it's quite meaningful that WCA provides a better "possible" format to select a champion in each competition.

And I expect that if Mean of 2 is added, it will become popular, held in more competitions than at present, like Mean of 3 in 333feet.

 

[Erik]

Is it a bad idea? If people don't get the chance to compete in it: yes.
Not everyone is as FMC crazy as Sébastien, you and me. Most cubers think it's a fun event, but after one attempt they've had enough. I highly doubt that the mean of 2/3 would be used by organisers for this reason. I hope it will, but if I have to be honest I am pretty sure it wont.

 

[okayama]

Does anyone have data how many FMC events with "Best of 2" have ever been held in each year so far...?

AFAIK three-round FMC events in 2010 have been planned 2 times, and 8 times in 2011. Becomes popular.

 

[Erik]

Cubing in general is becoming more popular. Of course there will be more rounds of everything, there are simply more competitions organised every weekend. Saying that more attempts of FMC means that it is becoming a more popular event is not necessarily the right conclusion.

Of course I would favor a 'best of 2' as preferred format btw. This way all results go to 1 rankings list.

 

[kinch2002]

How about leaving everything as it is, and just encouraging the 'high-authorities' to hold best of 2 at major championships?

I think (as many people do) that mean of 2/3 wouldn't be held at many events and even when it is held most people will be fed up after 20 minutes of the first attempt. We'd end up with a ranking list of 20 or 30 people which makes it a bit of a farce given the number of official cubers. At least the only really short rankings list atm (5bld) is so because it's a very difficult event to get a result in, even when people try (which lots more people have).

 

[oranjules]

i think that making mean of 2 the preferred format isn't a big issue : i assume there will be 2 rankings : single and mean... So if an organiser wants to do only one try, competitors won't be ranked in mean. So ? There were best of 1 for master magic in the Blind Open 2011, and it wasn't a big deal... With the time limits, there are also a lot of people who are ranked for single and not average... There are 2 rankings for all events, so i don't understand why you (Erik) still disapprove.
This change is also one of the only ones that would not perturb the actual ranking... (not like the pyraminx's tips for example)

 

[kinch2002]

Too many people would have no average, so the average rank wouldn't mean as much as one might hope it would. At least in other events anyone who is vaguely decent will have an average.

There was a best of 1 at blind open for master magic, but that's a one-off thing that doesn't really happen often (99% of comps with single also have a master magic avg). Here, we'd get something like 5% having average, and 95% just single

 

[okayama]

I understand that not so many people may have a chance to get the average record.

But still let me remark: "Best of X (with big X)" increases the potential of win with a single lucky solution.
So essentially that is not an remedy to prevent such an unfavorable situation...

 

[Kirjava]

If one procedure is bad, that does not make the other good.

 

[cuBerBruce]

With a mean of 3 (or mean of 2), there is an issue that if someone makes a small mistake in writing down their solution on just one of the attempts, that person automatically will be in last place with a DNF mean-of-N result. I presume this is why someone said they would prefer median of 3 to mean of 3, as you can still possible to place well, even if you make a small error on one solve. That's seems to be why generally WCA prefers a best-of format (rather than average/mean) in events where there is significant chance of a small error resulting in someone getting a DNF.

Assuming a competition would use the mean of N format, I also wonder about if a competitor gets a DNF on one attempt, would he/she still be allowed to do the remaining attempts even though he/she would have no chance to win? (The competitor might want to do so for the possibility of getting a record or PB.) Would it be the choice of the organizer whether or not such further attempts are allowed?

 

[irontwig]

Yeah, I think the risk of DNFing makes Mean of 3 too harsh, then only Median of 3, Mean of best 2 or Best of 3 remains.

 

[Zarxrax]

I'm not really big into FMC, so maybe this is a really ignorant comment, but, do competitors REALLY need a full hour? This just seems excessive to me, and seems to just encourage lots of trial and error to look for lucky stuff. Wouldn't a half hour be sufficient?

 

[cubernya]

Wouldn't a half hour be sufficient?

Lots of times, no. Many of my attempts take 45-50 minutes, and I know people like Tim with the NAR is still putting stuff together in the last 2-3 minutes.

 

[Mike Hughey]

I'm not really big into FMC, so maybe this is a really ignorant comment, but, do competitors REALLY need a full hour? This just seems excessive to me, and seems to just encourage lots of trial and error to look for lucky stuff. Wouldn't a half hour be sufficient?

The truth is that it becomes more "luck" if you make it shorter. An hour is sufficient that a good person can almost always come up with a semi-decent solution.

I've had several competitions where I've wanted to compete in both fewest moves and something else (usually something BLD), and the overlap required me to do fewest moves in shortened time (20 to 30 minutes). I can always find something in that time, but my expectations have to go down. Normally I'm happy with anything sub-35 (because I'm not very good), but if I only have 20 to 30 minutes, I'll usually be satisfied with anything sub-40, or maybe even sub-45.

 

[okayama]

@cuBerBruce , @irontwig
I also hope like Median of 3, Mean of best X if possible. Such formats are clearly better than the simple Mean of X in view of that point.

 

[mycube]

Mean of N for official FMC?

~sorry for my englisch - I have many ideas about this and I hope you can understand what I wanted to say. If not, feel free to ask~

Because of personal reasons and current occasion I want to start this discussion about Means for FMC. So let‘s get started:

I know this is a result of personal frustration in official FMC but nevertheless I am interested in what you think about this because I am surely not the only one thinking about this.

I often compare FMC with 2x2x2 Singles in a higher level - you all know the discussion about easy scrambles and luck. I think I don‘t have to tell you more about this.

So what‘s about the idea changing the official rating system of FMC in competitions? What about a mean of 2 attemps for official Fewest Moves?
You do not just need luck (e.g. LL-Skips) but also skill to win. So one lucky single would not be the garant to be definite on the podium. Yeah I know it‘s true - getting lucky is not a crime, but what can we do to get an better balanced FMC-system without luck as a main factor? Today FMC is the only event beside BLD which has not the Best-of-system. So why not change this to a mean of 2 or 3?
With this change you can prevent winning a competition with pure luck. You really need beside luck also skill to find two or three good results in a row. So the better persons in FMC have to pay attention not to DNF an attempt because of already a good solution.

This - in my opinion - would lead to „fairer“ results in FMC.

I know there are some arguments against this:
you need much more time for two attempts, but just a little note: there are already many competitions who not just have one attempt of FMC but two or three
there would be less means because many people DNF one of their attemps - what about this: they have to pay attention not to DNF, if they do it‘s their own fault . I bet, if we will have the mean system, there would be much less DNFs than there are now
less people doing FMC: if you really want to do FMC, another system is not a reason to stop doing it. It is much more a motivation to be more effective and to find a solution in every attempt

BUT: I see there also a lot of pros for official FMC-means:
you really need skill to have more good solutions than just a lucky one
you have to do both (or three attemps) to win; you not just have one good solution and DNF or DNS the second one.
no more lucky solutions which will stay in the world ranking forever without a high chance to be better
the means would be much more clearly about skills:
you have to do another good attempt after you had a really good solution in the first one


my personal suggestion:
I would prefer to have the Single AND the Mean-system for WCA-Competitions:
both count for the WCA-statistics and results but it would be like in any other event: a Single is more meaningless than the mean. Why not change the score to mean of 2/mean of 3. So you can do two or three attemps in competitions and they both count in the same record statistic.
Or the second way to handle with this: you get two scrambles in the limit of one hour. The mean of these two is the final result. But I think this is not a good way of doing FMC. Everyone who does FMC and want to get a nice result need the full hour. So there would be less good solution.

Another example for means: in the german forum we are testing the mean-system for FMC in the weekly competition since last week (3 scrambles) - successfully. The competitors really try to find a solution for every scramble to have a mean. I am not sure if this will last for long or it is just a motivation at the beginning but I like it.

So what do you think about it? I am really interested in different opinions about this topic.

 

[JianhanC]

Dein Englisch ist sehr gut, keine Sorgen

3 rounds of FMC takes too long, and the number of participants aren't very numerous. These are the only cons that I can think of, but otherwise it's a pretty good idea.

 

[irontwig]

This has been discussed before, I stand by my conclusion that median of three (so basically just the second best result) would be almost as indicative of skill as a mean would, but without being so punishing if you DNF for what ever reason. But yeah, (sadly) few comps have 3 (or more) FMC attempts.

 

[tim]

A mean of N would require N hours of competition time to determine a winner for FM. I think that's neither practical nor should it be required for such a niche event (keep in mind that FM is also the only event you can't watch as a spectator).

Another suggestion I have:
Everyone gets two (or three?) scrambles and 60 minutes time to submit solutions for both of them. The mean of both solves determine the winner of the event.
This adds some interesting tactical challenges (how much time do you want to invest in cube x?) while reducing the luck factor considerably.

I really think this event needs to be changed. I've been to too many competitions where the winner of FM didn't "deserve" it according to some skilled people. (The hostility these "undeserved" winners get is still extremely unnecessary!)