A league or a ranking list

[TheBB]

Ok, so in the span of 6 years, the sport of speedcubing has developed from a bunch of geeks posting on a remote Yahoo mailing list and occasionally meeting in a coffee shop (or those that were fortunate enough to be able to, anyway), into large and vibrant community with 169 official competitions in 2009 (that's almost one every two days!), and a handful of elite cubers who are doing intense record-hunting in their respective events.

But that's all that's really happening, isn't there? We go to competitions, solve the same damn things over and over again. Eventually, someone breaks a record, and there is much rejoicing, except perhaps from the previous record holder, and then everyone goes back to solving, while the new record holder gets his (or hers, as the case may be) name on the top of a list, and in a news item on wca.org, and so on and so forth.

The point I am meticulously trying to make is that there are no long-term competitions, or rather, the only long-term competition going on is that of record-snatching, which rewards peak performance only. There are no reliable ways of telling who is the best cuber in the world now, unless you are prepared to sift through dozens of recent competition results and make up your mind (and argue your choice in front of others). There are no competitions that measure this. The World Championship tries, but does a poor job. Three averages of 5 aren't that much, and for the last two installments (I believe) the eventual winner did not lead the pack after the first or seconds round. I'm willing to bet a lot of money that the leaders after rounds 1 or 2 will not win the WC in 2011 either.

In most sports, or at least those I follow, the most prestigious competitions are those that reward long-term consistent performance. Here are some examples:

Football (or soccer if you are one of them...)
The most prestigious tournaments for most clubs are the national leagues, which can only be won through consistent high-level play throughout a season. For the select few that can compete in the Champions League, it carries great weight, but this is mostly because of the stiff competition, and the fact that participation requires qualification through the national leagues already (so consistent high-level play is a prerequisite). The World Cup is also popular, but this is mostly due to tradition and patriotism. The level of play in the WC is not really that high compared to CL.

Chess
Chess, like cubing today, consists mostly of disjoint tournaments with little greater structure. Thus, one of the most prestigious "competitions" in chess is the rating list which is published 6 times annually. This list takes into account every game played and has been a good indicator of future performance as well. The other competition is the world championship, an exceedingly strenuous ordeal involving winning a number of localized tournaments against elite players (20+ games), knocking out other super-elite players in a small elimination over some 6 games each, only to be able to challenge the current champion to a ~20 game match, which you have to win TWICE, since the sitting champion has a rematch clause. This system has been changed a number of times since the 90s under what is arguably a series of poor decisions by the governing body of chess, but that doesn't really matter - when this system was in place, there were few better indicators of the current best player in the world than the current world champion.

What I would like to see in cubing is some sort of long term competition, league or possibly a reliable ranking system, that is designed specifically to reward current, consistent form.

We do not have such a thing at the moment. What we do have is:

• A ranking based on best results, which is unreliable. You can't get #1 without being good, but you can certainly do it without being best.
• A bunch of more or less reliable competitions, which are again unsatisfactory because very, very few of them (one, to be precise) gather the world elite.

Things are usually easier said than done. There are a number of unknowns and challenges with such a league or a ranking system. What we want is a system that achieves the following:

• Peak performances are irrelevant, means and averages matter more. Thus, presumably, a large number of results are required for every participant.
• Results should be counted from official competition results, for the sake of credibility.
• Most of the world elite should participate, also for credibility.

A few problems: the "world elite" consists mainly of kids which can't necessarily travel far and wide to put up a specified number of results for a long-term competition. Thus, we need a large amount of flexibility. We can probably not, for example, collect the same number of results from each participant.

We also want to avoid the situation that a participant has put up good results in january to april (assume that this is a yearly thing), and also fulfilling his requirements for, say, a minimum number of results, and then deciding not to compete for the rest of the year in order to collect his almost certain win. I don't care if it's allowed or even accepted, we don't want to take away peoples desire to compete on a normal, regular basis either.

A possibility would be that the competitor specifies in advance which competitions should count for his result, but since this is long-term, this is seriously inflexible (and flexibility was desired).

Then we come onto the scoring system, which could probably be discussed to time's end itself. Is a simple mean enough? Remember that excitement is not prioritized.

Would a ranking list instead of a league be better? Ranking lists usually carry great flexibility and credibility, but the scoring system is, again, problematic. Tennis and chess use very credible scoring systems, but they assume a 1 vs 1 game format, which cubing does not have - nor will it ever.

So yeah, discuss please.

 

[guitardude7241]

you could've summed that up and said-

should cubing have power rankings? i thought about this a few days ago. i say no.

 

[Cyrus C.]

If it was annually the results were tallied they would still be inaccurate because people can improve so much in a year. Also this would require many more competitions for a cuber to get enough solves in to qualify. Which probably aren't possible with the few amount of cubers we have. I think this would be better in the next 10 years when their are more cubers.

Although what we could do, is have a World championship, where there are state rounds, the top 3 from each move on to nationals, & then the top 3 move onto the world championships.

Another thing is, is that in chess, tennis, & soccer, you are competing against one another, which makes the want for something like this higher. Whereas in cubing you are trying to beat yourself/previous times.

 

[Lucas Garron]

Like this?

I think the community is there at the competitions. And once things get large enough, we'll have leagues and whatnot. The high schoolers at CA competitions already interact like some friendly interactive league.

Are you proposing that we implement something formal? Personally, I don't think it's time for that yet.

 

[qqwref]

It would be nice to have some kind of way to do averages at home and make them semi-official. Maybe instead of forcing it to be a specific competition we could have a bunch of trusted people who could be used as judges in any event, and anyone who averages fast enough at home would be able to submit an avg12 (or bigger?) judged by the trusted person every week. They could have their mom or a friend or something be trusted, so they don't have to travel anywhere. It wouldn't be a completely cheat-proof setup but it would provide a pretty good indication of who's the best current cuber. (Of course, a lot of fast people don't want their times known, so it wouldn't happen in practice.)

 

[Stefan]

Like this?

I changed that, btw. It used to consider only the solves in a moving 1-year window, but I switched to all-time now. One of the reasons was that the two versions were very similar, I believe the top 4 even were the same people in the same order.

(I must admit I only read about 1/3 of the original post... toooo loooong)

 

[JTW2007]

But that's all that's really happening, isn't there? We go to competitions, solve the same damn things over and over again. Eventually, someone breaks a record, and there is much rejoicing, except perhaps from the previous record holder, and then everyone goes back to solving, while the new record holder gets his (or hers, as the case may be) name on the top of a list, and in a news item on wca.org, and so on and so forth.

So?

I don't think that the current system needs to be altered at all. I doubt there can be a "best" cuber, because they would have to be absurdly fast at every event. I don't know many people who compete in every event, much less do well in everything. I'm fine with the way it is now.

 

[jms_gears1]

we dont honestly need to alter anything, i think the way its set up is the best way for this type of thing

 

[PatrickJameson]

You dragged this out way too long, to be honest. This isn't English class, all you need to do is get your point across, which still, even with this long manifesto of your's, was really not very apparent. What exactly do you want to do?

But that's all that's really happening, isn't there? We go to competitions, solve the same damn things over and over again.

Not how I see it. I see competitions as a chance to meet, be around other cubers and seeing other's skills in person, being able to discuss techniques and such with them. seeing, not really just for solving. Because of this, I find every competition fun and unique.

The point I am meticulously trying to make is that there are no long-term competitions, or rather, the only long-term competition going on is that of record-snatching, which rewards peak performance only.

If you're not in your peak performance, you should be. If you are talking about lucky solves and such, your way is quite worse than this(except maybe 2x2 single, but we shouldn't really count this in this discussion), which I will get to later in this post.

There are no reliable ways of telling who is the best cuber in the world now, unless you are prepared to sift through dozens of recent competition results and make up your mind (and argue your choice in front of others).

Well when is now? Now is at a specific competition, now is at a specific round, now is being able to out solve the others in equal conditions. Finding the 'best' cuber in the world is quite arbitrary and will always be, regardless of any scoring system. The simple fact that people get nervous during competitions is enough to prove this impossible to determine.

There are no competitions that measure this. The World Championship tries, but does a poor job. Three averages of 5 aren't that much, and for the last two installments (I believe) the eventual winner did not lead the pack after the first or seconds round.

As Michael Gottlieb puts it, competitions are to see who can get the best average of five at that very moment, not to determine who is best.

For the select few that can compete in the Champions League, it carries great weight, but this is mostly because of the stiff competition, and the fact that participation requires qualification through the national leagues already (so consistent high-level play is a prerequisite).

There have already been serious discussions about implementing strict qualifications for large competitions, but at this moment this is not needed. Next WC could, and in my prediction, will, be much different.

Chess
Chess, like cubing today, consists mostly of disjoint tournaments with little greater structure. Thus, one of the most prestigious "competitions" in chess is the rating list which is published 6 times annually. This list takes into account every game played and has been a good indicator of future performance as well.

Cubing is completely different. A few months of hardcore practice, at this early stage of the cubing history, could produce a world record. Chess involves years of studying just to get decent, so the rate of improvement is quite slow. With cubing, 'the best cuber' could change very very quickly.

What I would like to see in cubing is some sort of long term competition, league or possibly a reliable ranking system, that is designed specifically to reward current, consistent form.

As I said in a previous quote, cubing is anything but consistent.

Peak performances are irrelevant, means and averages matter more. Thus, presumably, a large number of results are required for every participant.

I seem to be going back to the cubing is not consistent thing quite a lot.

Results should be counted from official competition results, for the sake of credibility.

Wait I thought this list was for things that we don't have already.

Most of the world elite should participate, also for credibility.

How exactly do you plan on doing this? All the world's elite aren't able to afford to travel to a single competition.

A few problems: the "world elite" consists mainly of kids which can't necessarily travel far and wide to put up a specified number of results for a long-term competition.

Um, having the ability to travel somewhere is not always the problem. In the quote before this I brought up the cost issue.

Thus, we need a large amount of flexibility. We can probably not, for example, collect the same number of results from each participant.

We also want to avoid the situation that a participant has put up good results in january to april (assume that this is a yearly thing), and also fulfilling his requirements for, say, a minimum number of results, and then deciding not to compete for the rest of the year in order to collect his almost certain win. I don't care if it's allowed or even accepted, we don't want to take away peoples desire to compete on a normal, regular basis either.

Wait what? You don't even know how the scoring system is going to be like(I've read ahead), how are you deciding this already?

A possibility would be that the competitor specifies in advance which competitions should count for his result, but since this is long-term, this is seriously inflexible (and flexibility was desired).

Wait whhatt? This makes no sense to me. What could certain competitions possibly provide that others do not? If you are talking about possibly certain venues being better suited for a competitor because of some reason(you liked the chairs and table set up they had last year, maybe?). In this case, it would not be fair for others that don't have this royalty.

You are also forgetting that competitions are usually announced about 1-2 months in advance, how exactly would competitors specify which competitions they want to count if they don't know which competitions are going to be held?

Then we come onto the scoring system, which could probably be discussed to time's end itself.

And here's where I say, how can you propose this without having a way to actually rank people?

 

[TheBB]

you could've summed that up and said-

should cubing have power rankings?

No, I'm trying to think of something objective.

If it was annually the results were tallied they would still be inaccurate because people can improve so much in a year.

But I'm not trying to measure peak performance (or sudden improvement, if you will). Also:

I doubt there can be a "best" cuber, because they would have to be absurdly fast at every event.

I'm not talking about a competition encompassing all events. Maybe I didn't make this clear enough.

Like this?

No, those are records, not a ranking. Erik can never get worse than 239, no matter how bad he gets.

Are you proposing that we implement something formal? Personally, I don't think it's time for that yet.

Also this would require many more competitions for a cuber to get enough solves in to qualify. Which probably aren't possible with the few amount of cubers we have. I think this would be better in the next 10 years when their are more cubers.

Wait what? You don't even know how the scoring system is going to be like(I've read ahead), how are you deciding this already?

And here's where I say, how can you propose this without having a way to actually rank people?

I'm not proposing doing something now (although I don't think it's too soon). I just wanted to discuss, really, at first. These are ideas I'm having, not specific suggestions. Cut me some slack, will you?

It would be nice to have some kind of way to do averages at home and make them semi-official.

Yeah, it would.

If you're not in your peak performance, you should be. If you are talking about lucky solves and such, your way is quite worse than this(except maybe 2x2 single, but we shouldn't really count this in this discussion), which I will get to later in this post.

I spend much of the year out of form. I start cubing a bit about two months ahead of every competition I go to to get in form, but then I drop back again afterwards. Once we reach the saturation point, I don't think anyone will be able to stay in form at peak performance for ever.

Well when is now? Now is at a specific competition, now is at a specific round, now is being able to out solve the others in equal conditions. Finding the 'best' cuber in the world is quite arbitrary and will always be, regardless of any scoring system. The simple fact that people get nervous during competitions is enough to prove this impossible to determine.

"Now" is a sufficiently long enough time frame to have credible results, but also sufficiently short to allow for some turnover. This is experimentation. Nervousness is something that must be overcome. Good cubers must learn this, so that's not a problem.

As Michael Gottlieb puts it, competitions are to see who can get the best average of five at that very moment, not to determine who is best.

I may be in the vast minority, but I don't see how finding who can get the best average of five at a particular day very interesting, yet we're still organizing many and large competitions to do just that. Sure, yeah, there's the social aspect, but the objective of the WCA is not to have 169 parties yearly.

There have already been serious discussions about implementing strict qualifications for large competitions, but at this moment this is not needed. Next WC could, and in my prediction, will, be much different.

Good.

Cubing is completely different. A few months of hardcore practice, at this early stage of the cubing history, could produce a world record. Chess involves years of studying just to get decent, so the rate of improvement is quite slow. With cubing, 'the best cuber' could change very very quickly.

But here, you are implicitly assuming that "the best cuber" is the one with the world record (or if you're not, it sure sounds like it). This is not necessarily correct, at least not the way I see it.

As I said in a previous quote, cubing is anything but consistent.

I think this is just because the methods with which we measure cubers are not good.

Wait whhatt? This makes no sense to me. What could certain competitions possibly provide that others do not?

Nothing whatsoever. Let me try to clarify. A participant in the proposed league must put up, say, 6 averages to count. If he goes to more than 6 competitions, we must find a way to choose which 6. To avoid this, the competitor could say in advance of a competition that this result should count. He might, for example, be particularly motivated or find that he is, for whatever reason, in form or whatever. It doesn't matter.

You are also forgetting that competitions are usually announced about 1-2 months in advance, how exactly would competitors specify which competitions they want to count if they don't know which competitions are going to be held?

I'm not forgetting it, that's part of what I mean by inflexibility.

 

[PatrickJameson]

Nervousness is something that must be overcome. Good cubers must learn this, so that's not a problem.

I'm pretty sure I will never get over being nervous in an event I'm good at, no matter how long or how good I get a cubing.

I may be in the vast minority, but I don't see how finding who can get the best average of five at a particular day very interesting, yet we're still organizing many and large competitions to do just that.

We're getting off topic slightly. Your plan proposes long term rankings and has nothing to do with individual competitions. How exactly do you plan on moving away from this in a single competition?

But here, you are implicitly assuming that "the best cuber" is the one with the world record (or if you're not, it sure sounds like it). This is not necessarily correct, at least not the way I see it.

No I'm not. I'm saying that 'the best cuber' through your definition and any definition can change very very quickly. There could be some guy who when off to war for a few years, unable to attend any competitions, but practicing hard in the meantime. When he comes back and is able to attend a competition, he could destroy everyone, and would possibly be 'the best cuber', using your definition and mine.

As I said in a previous quote, cubing is anything but consistent.

I think this is just because the methods with which we measure cubers are not good.

See above.

Wait whhatt? This makes no sense to me. What could certain competitions possibly provide that others do not?

Nothing whatsoever. Let me try to clarify. A participant in the proposed league must put up, say, 6 averages to count.

What if they can't go to 6(or any proposed number) of competitions a year? What if they live in an area that gets one annual competition, and they can't afford to go to other competitions around the world? Would they be completely excluded from this ranking system?

If he goes to more than 6 competitions, we must find a way to choose which 6. To avoid this, the competitor could say in advance of a competition that this result should count. He might, for example, be particularly motivated or find that he is, for whatever reason, in form or whatever. It doesn't matter.

I just realized how nervous this would make me for the entire competition, knowing that I have to do better than a previous competition or else my 'ranking' would go down. And don't say, 'then don't compete', because that would just completely screw up the way cubing is done. I could also be just having an off day, and completely fail everything(see: my MIT results <_<). To be honest, if it was like that, I would probably quit.

 

[Erik]

It would be nice to have some kind of way to do averages at home and make them semi-official.

Yish?....

 

[Stefan]

Like this?

No, those are records, not a ranking. Erik can never get worse than 239, no matter how bad he gets.

Wow, you *really* did your research.

Edit: Ok... here comes the spoon: That's a ranking based on rankings based on averages, and everybody's score and rank can change in both directions (except the current leader's, of course).

 

[TheBB]

Uh yeah, I misinterpreted that list. I thought it was the best sum of placements in a single competition.

But it's also not what I'm looking for, because those ranks which you sum up are based on the best average in an event (amirite?) which does not change. So again, the only way you can get worse is by everyone else getting better.

 

[Sin-H]

just a big +1 for PatrickJameson

Track and field does it exactly like cubing too, btw
It's just how it works in a sport where you measure time.

Not everyone can go to so many competitions, at least, at the moment, it wouldn't make that much sense to switch to a ranking where your long-term performance counts...

and, srsly: at the moment our records improve so fast that you have to go to a lot of competitions and improve fast anyway if you wanna keep up.

It would be nice to have some kind of way to do averages at home and make them semi-official.

Yish?....

he said "some kind of way"

like, maybe, a WCA delegate visiting you (we'd need more delegates then, though). and filming it, too. you'd still be nervous but at least it's the environment you're used to.

 

[Stefan]

So again, the only way you can get worse is by everyone else getting better.

That's not "again". And if you do want people to drop ranks because they don't compete anymore or start sucking, what about the aforementioned version with a moving 1-year window?