Relative Solve Times for 2x2x2 - 7x7x7

[Logiqx]

It would be really interesting to see how those plots changed since 2018.
Does anyone happen to have the updated versions?

I'll look to do a refresh later this week.

 

[Logiqx]

Latest Stats (2020-02-18)

Updated charts - https://docs.google.com/spreadsheets/d/1ZnClqXZ1dHVvFIoBk0SCSl-5m34tV6mNtc2vldHhEXM/edit?usp=sharing

SQL - https://github.com/Logiqx/wca-stats/blob/master/sql/misc/relative_solve_times.sql

 

[One Wheel]

Are these based on average within vigintiles? I was thinking I was just exceptionally (relatively) good at big cubes, then I realized my official results reflect that, so if it was calculated the way I’d understood the Min for 4/5/6 should be lower.

Using my official results I come in at:
2x2/3x3: 0.409
3x3/4x4: 3.55
4x4/5x5: 1.62
5x5/6x6: 1.80
6x6/7x7: 1.50

My at-home global average is slightly better than that for 4x4 (3.30) but slightly worse for everything else. I haven’t checked other events.

 

[Logiqx]

Are these based on average within vigintiles? I was thinking I was just exceptionally (relatively) good at big cubes, then I realized my official results reflect that, so if it was calculated the way I’d understood the Min for 4/5/6 should be lower.

Using my official results I come in at:
2x2/3x3: 0.409
3x3/4x4: 3.55
4x4/5x5: 1.62
5x5/6x6: 1.80
6x6/7x7: 1.50

My at-home global average is slightly better than that for 4x4 (3.30) but slightly worse for everything else. I haven’t checked other events.

Yes. Averages are calculated within vigintiles.

e.g. Take the top 5% of 4x4x4 solvers and for each person calculate their ratio - 4x4x4 / 3x3x3. Calculate the average of all of those ratios to give the overall ratio for the top 5%. Repeat for every vigintile. Without exception the fastest 5% always have the best ratio (overall).

I do not report the minimum ratio of a single person either overall or within a single vigintile as it would be meaningless. For example, there is one guy with 5x5x5 PR average which is 35% of his 4x4x4 PR average.

 

[One Wheel]

Yes. Averages are calculated within vigintiles.

e.g. Take the top 5% of 4x4x4 solvers and for each person calculate their ratio - 4x4x4 / 3x3x3. Calculate the average of all of those ratios to give the overall ratio for the top 5%. Repeat for every vigintile. Without exception the fastest 5% always have the best ratio (overall).

I do not report the minimum ratio of a single person either overall or within a single vigintile as it would be meaningless. For example, there is one guy with 5x5x5 PR average which is 35% of his 4x4x4 PR average.

How much more work would it be to add error bars? If it would be tough don’t worry about it, but if it’s easy I would be very curious.

 

[Logiqx]

How much more work would it be to add error bars? If it would be tough don’t worry about it, but if it’s easy I would be very curious.

I'll look at that next week. It might show something interesting.