Proposal - Average of 5 for 6x6

[kinch2002]

I don't agree with people mentioning that time shouldn't be an issue because FM mean of 3 is being discussed. The change is FM is highly useful for major championships. The single solve format would still be totally worthwhile in most competitions and I expect the majority to stay using best of 1. However, average of 5 for 6x6 would become the standard for all comps and that's why this is a different time issue.

 

[cubeninjaIV]

Scrambling the puzzles isnt the issue in my experience, as a decent scrambler could get through 3-5 cubes in the time it takes most people to solve it meaning that with two or three people they can easily keep up with the average of about 8 timers at a competition and while not everyone jumps at the opportunity to scramble 6x6, if the event is being held at all it means that there are people willing to scramble.

I don't think it would be at all practical to have a MO3 and AVG5 at the same time so that said, if an average of 5 was made standard, the only way to not greatly increase the time it takes to complete the event is to make harder cut-offs and by making it an average of 5 event, you will almost double the number of solves, and therefore double the time spent on the event and the only way to keep the time down is to make cutoffs a lot tougher

I'm all for making 6x6 an average of 5 event, but doing so will make it a much longer event that will be harder to fit into schedules that already sometimes are hard to keep up with, meaning that holding the event comes at a cost of other events or rounds not being held, and probably making 6x6 a much less common event

 

[Dene]

Could someone with powas (Mr. Pochmann?) please find out how many competitions host 6x6? And how many host multiple rounds of 6x6 I think would be useful too. It would be nice to have evidence to support or refute some of the claims being made in this thread.

 

[Sa967St]

The Feet discussion has been moved here:
http://www.speedsolving.com/forum/showthread.php?43591-Remove-3x3x3-with-Feet-as-an-Official-Event

 

[cubizh]

Could someone with powas (Mr. Pochmann?) please find out how many competitions host 6x6? And how many host multiple rounds of 6x6 I think would be useful too. It would be nice to have evidence to support or refute some of the claims being made in this thread.

As of last database update, there have been 413 competitions that hold 6x6.
They were devided into Combined Finals (235 competitions) and Finals (178).
Of these, 3 competitions had Qualification rounds; 12 had First round and 27 had Combined First round

For more details about 6x6 compared to other events check here and here.

 

[Dene]

Thank you very much, these statistics are very telling, specifically in reference to FMC. It seems that almost an identical number of competitions have held FMC, which is guaranteed at least one hour spent. That's not to mention the extra time spent marking solutions. I can't speak for other competitions, but usually here in Australia for 6x6 we have a cutoff around 4 minutes and we expect the round to take up roughly 30 minutes of competition time. Looking at the database (only quickly because the wca site is going really slowly at the moment) around 1000 people have completed a mean in 6x6, and 1400 people a single. So about a third of people that compete in 6x6 aren't getting more than one or two solves anyway. I suspect this means other competitions take advantage of cutoffs, and quite possibly use up a similar amount of time for 6x6 as we do in Australia (although perhaps we are just lucky because me and Feliks are highly competent scramblers).

My point being thus: Some people are objecting to the suggested change, claiming that an average of 5 would add a disproportionate amount of time to run 6x6 in competitions. Based on some basic stats and intuition, I don't believe there is any basis to these claims.



NOTE: I am not arguing for changing 6x6 to average of 5, in fact I am still on the fence regarding this issue (I'm waiting for a good argument to persuade me). I'm just going to play devil's advocate best as I can.

 

[ryanj92]

How much of a compromise would median of 3 be? On initial inspection, it seems okay - can get away with one bad solve, no need to spend extra time doing two more solves, still need two good solves to do well.

But there's something to be said for reducing averages to one solve only, I'm sure...

 

[AvGalen]

27 posts and nobody made a calculation to show how much extra time it would cost. So here I go
Assumption: Currently there is a cutoff at 4 minutes, time limit of 6 minutes, slow solvers need 5 minutes on average, fast solvers need 3 minutes on average, 1 out of 3 solvers is slow. Let's say there are 15 solvers and 5 timers

3 solves: 5 people needing 1* 5 minutes, 10 people needing 3 * 3 minutes = 25 + 90 = 115 total time, so 23 minutes pure timer-occupying time. time limit of 6 minutes turns out to be irrelevant for this calculation but it is necessary to prevent extremes.
5 solves: 5 people needing 2 * 5 minutes, 10 people needing 5* 3 minutes = 50 + 150 = 200 total time, so 40 minutes pure timer-occupying time. Same comment about the time limit.

Result: (the "fun"-formulas are now AvG-patented)
Theoretical increase in fun: (5-3)/3 = 67%
Practical increase in fun: ((10+50)-(5+30))/(5+30) = 71%
Practical increase in time: (40-23)/23 = 74%