WCA Scramble Algorithm

[bcube]

Structure of the short solutions for the scramble matter MUCH more than the move count of absolute optimal solution.

I understand what you are saying and why you are saying it, and why Lucas said "I don't think we know right now whether basing this purely on the optimal solution move count is intrinsically guaranteed to solve more problems than it could cause". From your (method-dependent) point of view it makes sence, while I am operating on a method-independent point of view, hence the differences. As soon as "more fair" scrambles/rankings solution is introduced (if ever), I will be happy.

Say we have competitors A and B. Both are top level FMC solvers and thus use the before described method of DR/HTR. A goes to a competition X and B goes to competition Y. Both get an 18 move optimal scramble. The scramble A got has an optimal solution that is very easy to find with the method that EVERY SINGLE top level competitor uses. B gets a scramble where only 18 and 19 move solutions are practically impossible for any human to find. How is that fair in your opinion?

My opinion: they both had an 18 optimal moves scramble, they both could theoretically find an 18 moves solution. Just because EVERY SINGLE top level FMC-er doesn´t find an 18 or 19 moves solution doesn´t mean a total FMC beginner can´t find it too. Granted, chances are small, but they exist nonetheless in this scenario.

Of course, I would be happy to hear any "more fair" proposals than I made because I am not saying my proposals are the best ones we can come up with (I just made proposals which are fixing the issue from my point of view).

 

[stoic]

Just wanted to make the world a little bit better for everyone.

That seems a bold claim, and possibly disingenuous. Are you including the current WR holder in that?

 

[tseitsei]

I understand what you are saying and why you are saying it, and why Lucas said "I don't think we know right now whether basing this purely on the optimal solution move count is intrinsically guaranteed to solve more problems than it could cause". From your (method-dependent) point of view it makes sence, while I am operating on a method-independent point of view, hence the differences. As soon as "more fair" scrambles/rankings solution is introduced (if ever), I will be happy.

There is no good and proper definition for "more fair". Optimal movecount sure isn't one as I have demonstrated many times on my previous posts. So we just have to accept that luck is and always will be part of cubing.

My opinion: they both had an 18 optimal moves scramble, they both could theoretically find an 18 moves solution. Just because EVERY SINGLE top level FMC-er doesn´t find an 18 or 19 moves solution doesn´t mean a total FMC beginner can´t find it too. Granted, chances are small, but they exist nonetheless in this scenario.

Of course, I would be happy to hear any "more fair" proposals than I made because I am not saying my proposals are the best ones we can come up with (I just made proposals which are fixing the issue from my point of view).

You seem to be thinking this in some weird purely theoretical world. But in real world actual human beings are the ones competing.
Even if it is theoretically possible to find optimal solution for both those scrambles, it is no doubt SIGNIFICANTLY easier on the other one. Hence the scrambles are not equal and fair for both competitors.

I'm sure you tried to make proposals that would fix things. But as stated with lengthy arguments by me and several other people, we really don't think those proposals would actually fix anything or be an improvement over the current system.

 

[bldcuber314]

My opinion: they both had an 18 optimal moves scramble, they both could theoretically find an 18 moves solution. Just because EVERY SINGLE top level FMC-er doesn´t find an 18 or 19 moves solution doesn´t mean a total FMC beginner can´t find it too. Granted, chances are small, but they exist nonetheless in this scenario.

I don't understand. So the problem you're trying to solve is the hypothetical scenario where an FMC beginner tries some random moves, and despite the 43 quintillion combinations of the cube they stumble upon the optimal solution. I don't see that ever being an issue. And then it would be extremely unfair to that beginner because apparently they deserve the record just as much as people who have devoted hours and hours to FMC and got their solution through a smart, advanced solution. There are much bigger issues in cubing then this...

 

[bcube]

the WRC looks at mis-scrambles to figure out if they were unfairly easy compared to the proper random scramble.

Can you tell us more? How is unfairly easy scramble defined? Does it depend on the solving method used for such mis-scramble by the competitor?

I don't think we know right now whether basing this purely on the optimal solution move count is intrinsically guaranteed to solve more problems than it could cause.

From your perspective, what main problems could it cause?

 

[bldcuber314]

Can you tell us more? How is unfairly easy scramble defined? Does it depend on the solving method used for such mis-scramble by the competitor?



From your perspective, what main problems could it cause?

Um yes, it does depend on the solving method used because we don't live in a purely hypothetical world, and this is why this FMC issue isn't really an 'issue' in the first place! The problems it would cause-

Current results being disqualified- every result where the optimal solution isn't 18 moves.

It not actually solving any problems, because there aren't really any FMCers who actually see it as an issue. It wouldn't be very popular if suddenly half the results achieved were disqualified and it didn't actually solve anything.

Every time a record single hits 18, it would never be broken again but for example the WR single would end up tied by a huge number of people. This makes the event much less interesting if it is impossible to beat the WR, and there would be too many WR holders for 'holding a WR' to be a special thing.

The scrambles are still unequal because what decides the luckiest of an FMC scramble is not the optimal solution, but how easy and how good of a solution you can find with standard methods.

 

[quanloh]

I have been away for quite sometimes, glad to see people on the forum still actively discussing non-issues and try to convince each other with subjective opinion

 

[bcube]

How is unfairly easy scramble defined? Does it depend on the solving method used for such mis-scramble by the competitor?

Relevant and meaningful discussion on this topic (look for a discussion between users Stefan and qqwref, it ends on page 10 of that thread): starts here.

 

[Billabob]

bcube, you are the only person I have seen who thinks this is an issue, even after months of spamming threads about it and having many many top FMC solvers weigh in. At some point you have to give it a rest. It's disappointing when interesting FMC-related threads get bumped, I check on them hoping for an update, and it's just you dragging out the same arguments and misconceptions yet again.

Perhaps in the far future where solvers can find optimal solutions 80% of the time, the community will find it necessary to shift to some other metric, such as "time to find optimal". But as of now you're taking up arms against a hypothetical, and while it is an interesting discussion it's not one that merits such extreme dedication & volume. There's very little more that could be said, please could you leave it to rest.