WCA rankings

[Noahaha]

The problem is that some competitions are MUCH easier to win than others. For example, the tenth best cuber at a New England comp might place in the top 3 at a comp in Washington. Also, not everyone has the same opportunities to compete. The current system seems the most fair, since you're competing against everyone, not just the people near you, and also you don't have to compete all the time in order to have a good ranking.

I don't see how inactive cubers skew the rankings at all. They still deserve the ranking, even if it was achieved a while ago. If you really want to see who the best "current" cubers are, just select this year on the rankings page.

 

[antoineccantin]

I think the idea here is similar to the distinction between world record and world leading in athletics. World record is the best time ever achieved, and world leading is the best time achieved in the current year. So Giovanni Contardi has the world record for OH single, and Gabriel Dechichi Barbar is the world leader. Since cubing records are broken so frequently, though, the world record and world leading are usually the same.

And why should Gabriel Dechichi Barbar be considered world leader? Because he has a faster average than the world record holder? Michal Pleskowicz has 6 averages better than Gabriel's best. Heck, even I have 2 averages faster than Gabriel's best. Should that make me second in the world?

 

[Torch]

And why should Gabriel Dechichi Barbar be considered world leader? Because he has a faster average than the world record holder? Michal Pleskowicz has 6 averages better than Gabriel's best. Heck, even I have 2 averages faster than Gabriel's best. Should that make me second in the world?

I never said I agreed with this idea, and I specifically said OH single. Michal Pleskowicz has the fastest average this year, so yes, he would be the average leader, and you would be second. I don't think this should be an official measure; it's just an interesting statistic.

 

[antoineccantin]

I never said I agreed with this idea, and I specifically said OH single. Michal Pleskowicz has the fastest average this year, so yes, he would be the average leader, and you would be second. I don't think this should be an official measure; it's just an interesting statistic.

I was just pointing out that there is no reason to say that Dechichi is the world leader and that Contardi is the world record holder. Gabriel might have a faster average but slightly slower single than Giovanni, but that still isn't a valid reason since Michal also has a faster average and an almost as fast single.

 

[Torch]

I was just pointing out that there is no reason to say that Dechichi is the world leader and that Contardi is the world record holder. Gabriel might have a faster average but slightly slower single than Giovanni, but that still isn't a valid reason since Michal also has a faster average and an almost as fast single.

What does the average have to do with anything? I was discussing the single. I'm not saying Gabriel is the best in the world at OH, only that he happens to have the fastest single so far this year.

 

[antoineccantin]

What does the average have to do with anything? I was discussing the single. I'm not saying Gabriel is the best in the world at OH, only that he happens to have the fastest single so far this year.

Ahhhhh, sorry I didn't see that part.

 

[UnsolvedCypher]

Maybe a ranking could be determined by averaging the last few averages together? Or perhaps ranking them a little lower if they haven't competed in over a year? I think this would give a pretty accurate representation of who is the fastest right now.

 

[Lucas Garron]

There have been proposals to have a separate ranking system for cubing, just like chess has rankings that determine, say, grandmasters. This is not inherently a bad idea, but here are some thoughts on it:

A lot of these other sports cannot be measured objectively. Chess players can only be compared one game at a time. Certain sports are judged subjectively (gymnastics, dance, etc.). In cubing, we have a simple objective measure. Perhaps the single solve is not that important.

LostGent mentioned Maurice Greene as an example of someone with a good result who should perhaps not be ranked as competitive. But does Track and Field even have such a ranking?

I don't think we need to introduce a mechanism to encourage "active" cubing. As mentioned, this can be unfair for competitors in isolated regions. We want cubing to remain a cooperative sport where cubers try to get the best times they can, not to beat as many others as possible.

We don't really need it, and it's not clear how we would go about designing a fair system.

 

[Dene]

I get what you're saying. I was thinking about having ranked tournaments, so obviously the bigger the event the more points go towards your ranking

So Feliks would miss out because everyone else in Australia sucks?

 

[LostGent]

LostGent mentioned Maurice Greene as an example of someone with a good result who should perhaps not be ranked as competitive. But does Track and Field even have such a ranking?

I don't think we need to introduce a mechanism to encourage "active" cubing. As mentioned, this can be unfair for competitors in isolated regions. We want cubing to remain a cooperative sport where cubers try to get the best times they can, not to beat as many others as possible.

We don't really need it, and it's not clear how we would go about designing a fair system.

I live in Ireland, we've only had one competition here so I can completely understand where you're coming from. I just thought it might have been more interesting to see the rankings work in a manner similar to golf or tennis but then this is an amateur hobby so it's very different. As for Maurice Greene, it was just to indicate how if we applied WCA rankings to Track and Field it would lead to skewed results. Our community is very small and can be isolated so perhaps it's not the best idea.

Thanks for all your feedback everyone

 

[Stefan]

In cubing, we have a simple objective measure.

Programming competitions like TopCoder or CodeForces have a fairly simple fairly objective measure as well, and they're doing nice ratings. You can see mine on the bottom here:
http://community.topcoder.com/tc?module=MemberProfile&cr=263379

Works something like this:

1. Take the ratings of a round's participants before the competition.
2. Calculate which position they're expected to end up.
3. Somehow adjust ratings according to how much better/worse than expected they actually end up.

Of course, they have the advantage of frequent global competitions.

I live in Ireland, we've only had one competition here so I can completely understand where you're coming from.

Really? He's from California. They've had about 97 competitions.

I just thought it might have been more interesting to see the rankings work in a manner similar to golf or tennis but then this is an amateur hobby so it's very different.

I don't know how golf/tennis rankings work, but if they're interesting, we could maybe add some semi-/unofficially in the misc section.

As for Maurice Greene, it was just to indicate how if we applied WCA rankings to Track and Field it would lead to skewed results.

They would not be skewed. They would be absolutely correct all-time rankings.

 

[scylla]

Its a good idea as an extra, not instead off.

Let's look at this example

Cuber A did 6 averages at 2 Competitions: 5 of them where between 11 and 12 sec, but one in the first round was 9 sec.
Cuber B did at the Same competitions 6 averages of 10 sec and has besten A in 5 of THE 6 rounds including both finals.

Who is THE beter cuber? I think its B. Current rankingsysteem says A is better

 

[antoineccantin]

Its a good idea as an extra, not instead off.

Let's look at this example

Cuber A did 6 averages at 2 Competitions: 5 of them where between 11 and 12 sec, but one in the first round was 9 sec.
Cuber B did at the Same competitions 6 averages of 10 sec and has besten A in 5 of THE 6 rounds including both finals.

Who is THE beter cuber? I think its B. Current rankingsysteem says A is better

It really depends. Perhaps cuber A is just extremely nervous in competition, and got less pressure in the first round so achieved a normal average, but failed badly due to pressure in the other rounds. Also, maybe cuber B does well under pressure and got good averages in every round. He also got more pressure in both finals, so achieved better times.

 

[Stefan]

Current rankingsysteem says A is better

No it doesn't. It says A has a better record.

 

[scylla]

No it doesn't. It says A has a better record.

That's correct, so I think this idea is an interesting option to find a ranking which put THE best cuber at the top.

When there isnt one, people will percept the current rankings like A is better than B

It really depends. Perhaps cuber A is just extremely nervous in competition, and got less pressure in the first round so achieved a normal average, but failed badly due to pressure in the other rounds. Also, maybe cuber B does well under pressure and got good averages in every round. He also got more pressure in both finals, so achieved better times.

Haha, are you serious?

 

[antoineccantin]

Haha, are you serious?

Yes...

 

[JonnyWhoopes]

Haha, are you serious?

That's actually a very common scenario. Pressure to do well in competition is real, and some people deal with it better than others. If you go to my official competition history, you'll see that I always do worse in the second round of competitions because of the pressure.

 

[scylla]

I know pressure has impact and the impact is not the same for everyone. However, its part of the game. It makes no sense to try to filter the impact of pressure out of results.

 

[qqwref]

It also makes no sense to try to filter people's best times out of results. Cubing is about getting the objectively best time, not about beating people. Note that even final rounds of big competitions are not done in a "head to head" fashion, but instead, each person does their solves individually and the times are compared later.

 

[KongShou]

It also makes no sense to try to filter people's best times out of results. Cubing is about getting the objectively best time, not about beating people. Note that even final rounds of big competitions are not done in a "head to head" fashion, but instead, each person does their solves individually and the times are compared later.

Maybe we should start doing knock outs for big competitions. Would be fun to watch.