• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

All-Round Rankings (KinchRanks)

Coolster01

Member
Joined
Dec 24, 2011
Messages
2,342
Location
Near Deetroit, Michigan, United States
WCA
2011SBAH01
YouTube
Visit Channel
RankNameScoreminxpyramsq1clockskewb
1Jayden McNeill3.5130.6210.7070.6340.5501.000
2Rami Sbahi3.2800.6570.6810.5460.5080.888
3Evan Liu3.2690.5830.5240.4721.0000.690
4Daniel Wallin3.2630.4790.5410.5290.9050.807
5Robert Yau3.2480.7800.6480.6540.5810.585
6Yu Nakajima (中島悠)3.2220.7420.5800.6450.7150.539
7Louis Cormier3.1910.9370.6500.4630.6480.492
8Antoine Cantin3.1890.6480.6430.4710.7440.683
9Vincent Hartanto Utomo3.1720.6370.8150.5280.6560.535
10Daniel Sheppard3.1230.5890.5940.5440.8200.575
11Feliks Zemdegs3.0100.8410.5650.5500.5030.551
12Lucas Wesche2.9560.8370.4210.6540.5130.531
13Wojciech Knott2.9450.5700.5460.4100.9050.514
14Nathaniel Berg2.9390.4590.5800.2980.9570.646
15Nathan Dwyer2.9290.5980.4610.6930.6610.516
16Jan Bentlage2.9050.6010.5280.5150.5980.662
17Ainesh Sevellaraja2.9010.5910.6880.5350.6610.426
18Simon Westlund2.8670.8920.5510.4850.6160.323
19Mitchell Lane2.8450.6140.5550.4230.5840.668
20John Brechon2.8430.6870.6040.5080.5810.463
21Ivan Zabrodin2.8170.5560.4430.3530.7650.700
22Lee Chiang (蔣礪)2.7590.5030.6260.2620.7030.665
23Maarten Smit2.7400.6030.5240.5200.8200.273
24Walker Welch2.7380.5360.4770.4000.7170.608
25Chris Wall2.7350.8030.3800.3880.5880.576
26Dmitry Kryuzban2.7310.6150.6900.5170.6900.220
27Chia-Liang Tai (戴嘉良)2.7240.3420.6740.5130.6640.532
28Dan Cohen2.7100.5850.5240.7020.6210.279
29Oscar Roth Andersen2.7050.8480.8650.3070.1390.545
30Albin Xhemajlaj2.7010.4140.8080.2950.5730.611
31Rok Glinšek2.6840.4780.4270.4820.5150.783
32Bence Barát2.6180.6300.4380.6290.4390.481
33Jules Desjardin2.6080.3810.9550.5160.4310.325
34Brian Johnson2.5990.4630.5480.5590.4750.554
35Joshua Feran2.5850.4970.5250.2100.7960.557
36Emanuel Rheinert2.5680.3950.4400.8480.4270.458
37Justin Thomas2.5660.6330.4150.6100.5850.324
38Wilhelm Kilders2.5620.4420.4120.5000.6090.598
39Kim Jokinen2.5510.4420.6510.2840.6600.513
40AJ Blair2.5510.5560.3820.5500.6370.426
41Sébastien Auroux2.5500.3690.5570.3670.7670.490
42Carlos Méndez García-Barroso2.5330.5040.5260.4510.6820.370
43Paweł Kowol2.5230.3980.6260.5370.5380.425
44Jorge Castillo Matas2.5160.5840.5580.4180.4870.469
45Niko Ronkainen2.5030.3500.4820.2350.8200.615
46Anthony Lafourcade2.5010.2900.6630.1720.5780.797
47Michael Young2.4800.4970.3800.6870.4070.509
48Corey Sakowski2.4670.5020.5410.4210.3760.626
49Erik Akkersdijk2.4510.6670.5310.3860.5490.318
50Zijia Feng (冯子甲)2.4440.3030.3730.5390.8140.416
51Michał Halczuk2.4430.5900.2640.7690.4540.366
52Tim Major2.4410.3630.6840.3260.3890.678
53Wataru Hashimura (端村航)2.4350.5910.4080.2920.6230.521
54SeungBeom Cho (조승범)2.4320.5740.6480.4880.0000.723
55Matic Omulec2.4310.8500.4800.4540.3550.291
56Hunor Bózsing2.4270.5170.4160.3710.4360.687
57Ryan Jones2.4250.4550.4270.2850.7800.478
58Callum Hales-Jepp2.4220.6420.3930.3530.6790.356
59Kevin Costello III2.4200.6740.4480.3240.4380.536
60Alexandra Daryl Ariawan2.4020.4320.4090.4950.5220.544
61Kit Clement2.3990.3130.3680.3200.8230.575
62Rui-Jun Liu (劉睿鈞)2.3910.4100.4670.7590.3600.394
63Viktor Ejlertsson2.3850.4990.4370.2790.5760.594
64Christopher Olson2.3800.4230.4290.3720.6160.539
65Yinghao Wang (王鹰豪)2.3750.4770.4920.2800.6370.489
66Henri Gerber2.3640.7210.4590.2720.3510.561
67Michael Röhrer2.3590.6420.3360.5470.5760.259
68Jong-Ho Jeong (정종호)2.3570.7890.4420.4430.2290.453
69Mason Langenderfer2.3550.3770.4190.2980.5020.760
70Dmitry Aniskin2.3510.6430.4020.3230.5060.478
71Henry Savich2.3290.3780.7730.2710.4720.435
72Daniel Gracia Ortiz2.3200.4280.4720.5440.3540.522
73Michael Gottlieb2.3160.5740.3790.4380.5400.386
74Shao-Heng Hung (洪紹恆)2.3060.3600.4650.3060.6800.495
75Mats Valk2.3050.4680.4470.5680.4350.387
76Logan McGraw2.2990.3870.2760.2670.7650.603
77Drew Brads2.2980.4351.0000.2090.0000.654
78Daniel Cano Salgado2.2870.5730.3990.3070.6370.371
79Marcin Jakubowski2.2820.3600.4440.2410.7690.467
80Nikita Loyko2.2800.3680.4240.3210.7350.432
81Thomas Schmidt2.2660.3400.6050.2290.4780.614
82Yu Sajima (佐島優)2.2630.5140.4270.2560.7890.278
83Laura Ohrndorf2.2470.3180.4310.3630.7760.359
84Mattia Furlan2.2470.5730.4670.0000.7080.498
85Jure Gregorc2.2460.6030.3810.2690.5670.426
86Sungho Hong (홍성호)2.2450.4450.3470.3900.7110.353
87Filip Pasławski2.2380.4490.6370.2290.1960.728
88Alex Thielemier2.2360.3120.6810.4400.5770.226
89Chunyu Zhang 2 (张春雨)2.2290.4360.5360.4960.3610.399
90Artem Melikian (Артем Мелікян)2.2200.4690.5240.2480.3510.629
91Andy Denney2.2130.5610.3600.3120.3660.615
92Mharr Justhinne Ampong2.2110.6050.5580.2630.1640.621
93Hippolyte Moreau2.2070.5630.4550.2360.5260.428
94Brandon Lin2.2030.4520.3240.7960.3600.271
95Luke Hubbard2.2030.3190.6750.3690.5400.300
96Brady Metherall2.2010.4220.3910.6670.4260.295
97Nathan Azaria2.1980.2850.3070.5700.7710.264
98Blake Thompson2.1960.4240.4290.4510.3790.513
99Xiao Hu (胡霄)2.1840.4310.3770.4010.4420.533
100Evan Brown2.1830.4120.3280.3610.7520.330

After my last comp: 2.56/3.28 = 0.780 for pyraminx, 42.89 / 64.63 = .664 for mega, 5.94 / 11.28 = .527 for clock now.
.78+.664+.527+.546+.888 = 3.405 now. Yay, I'm getting closer to Jay. :p
 

kinch2002

Premium Member
Joined
Dec 22, 2008
Messages
2,504
Location
Guildford! UK!
WCA
2009SHEP01
YouTube
Visit Channel
Rob and I briefly discussed changing the magnitude of the scoring system. Describing your scores as 0.xxx is a bit weird and cumbersome.
How about if we multiply everything by 100? It seems more obvious what it all means, and everyone will start speaking to the same level of accuracy (i.e. 2 sig figs) :)
The final all-round score would be an average of all the scores (sum them and divide by 18). This gives it a more comprehensible score than the current way.

Instead of
Feliks has a score of 12.2 (out of possible 18), with a score of 0.86 in 5x5, 0.21 in multibld and 1 in 3x3
It would be
Feliks has a score of 67, with a score of 86 in 5x5, 21 in multi, and 100 in 3x3.
I think the latter may be better. Any thoughts?
IMkUHpc.png


Further thoughts on the system generally:
- Over time, the depth in events will increase, and I think this will result in higher Sum of Ranks scores in the future. However, I think the best KinchRanks scores should not get worse, as they do not depend on the depth of the event. This is a positive imo.
- In KinchRanks, a second gained when you're near the WR is worth a lot, while a second gained when you're not near the WR is not worth so much.
In Sum of Ranks it's generally the opposite way round. You gain a lot from improving your not-so-good events a tiny bit and aren't rewarded for the super difficult improvements at the top end of the event.
Again, a big positive for KinchRanks imo.
 

Kit Clement

Premium Member
Joined
Aug 25, 2008
Messages
1,631
Location
Aurora, IL
WCA
2008CLEM01
YouTube
Visit Channel
- In KinchRanks, a second gained when you're near the WR is worth a lot, while a second gained when you're not near the WR is not worth so much.
In Sum of Ranks it's generally the opposite way round. You gain a lot from improving your not-so-good events a tiny bit and aren't rewarded for the super difficult improvements at the top end of the event.
Again, a big positive for KinchRanks imo.

I was just thinking about this myself -- and while it's great for those among the top, it makes it very difficult to distinguish middle-of-the-pack cubers, especially for events like 2x2 where the WR time is so small. Maybe some form of log transformation would be interesting to apply.
 

not_kevin

Member
Joined
Jul 23, 2008
Messages
495
Location
California, USA
WCA
2008YOUN02
YouTube
Visit Channel
I really like this system - many props for formalizing your thoughts like this, Daniel!

Rob and I briefly discussed changing the magnitude of the scoring system. Describing your scores as 0.xxx is a bit weird and cumbersome.
How about if we multiply everything by 100? It seems more obvious what it all means, and everyone will start speaking to the same level of accuracy (i.e. 2 sig figs) :)
The final all-round score would be an average of all the scores (sum them and divide by 18). This gives it a more comprehensible score than the current way.

Instead of
Feliks has a score of 12.2 (out of possible 18), with a score of 0.86 in 5x5, 0.21 in multibld and 1 in 3x3
It would be
Feliks has a score of 67, with a score of 86 in 5x5, 21 in multi, and 100 in 3x3.
I think the latter may be better. Any thoughts?

I also like this modification, because as you said, it's easier to visualise the output. In particular, I like the averaging of the overall score, because it's easier to compare across some of the other modifications proposed so far (such as the ones limiting events (side events only, BLD only, etc.)).
 

kinch2002

Premium Member
Joined
Dec 22, 2008
Messages
2,504
Location
Guildford! UK!
WCA
2009SHEP01
YouTube
Visit Channel
I was just thinking about this myself -- and while it's great for those among the top, it makes it very difficult to distinguish middle-of-the-pack cubers, especially for events like 2x2 where the WR time is so small. Maybe some form of log transformation would be interesting to apply.
I don't see why 2x2 would be any different to other events. In fact, it should appear to separate people who are close in shorter events more than in longer events.
The system reflects the fact that it's easier to improve from e.g. 8.1 to 8.0 at 2x2 than it is to go from 2.1 to 2.0. That's why the top people appear to have larger separation - their skill difference is much bigger imo. It's perfectly natural for there to be a spacing out at the top of any ranking (cubing or not).
The logarithm, while interesting perhaps, would go against the basic idea of the system. Not many people can interpret logarithmic scores - pretty sure we all work best in linear.
Sorry, that was the same thing explained many times badly :p
 

kinch2002

Premium Member
Joined
Dec 22, 2008
Messages
2,504
Location
Guildford! UK!
WCA
2009SHEP01
YouTube
Visit Channel
I also like this modification, because as you said, it's easier to visualise the output. In particular, I like the averaging of the overall score, because it's easier to compare across some of the other modifications proposed so far (such as the ones limiting events (side events only, BLD only, etc.)).
Thanks - you make a good point about subset rankings.

I thought of a downside - it's not as easy to figure out how your overall score will change when you improve at something.
But I still think I'll change it anyway.

ANOTHER IDEA:
I mentioned that KinchRanks uses FM single and 3bld Single because overall they reflect abilities better than their respective Means still. However, I think we can improve on this by using the better score of a person's Single and Mean. Given that in the ideal world we'd use Means for these events, I think a good Mean should be rewarded with the option to use it, as it is a truer reflection of ability.
In the future I'd like to fully shift to Means only, but that is still over a year away imo.
 

Keroma12

Member
Joined
Jul 28, 2010
Messages
656
Location
Vancouver, BC, Canada
WCA
2010MATT02
YouTube
Visit Channel
I like this. Is it possible for this to be on the WCA page along with sum of ranks? If not, it would be really nice if it were hosted on the web somewhere. I would also like to be able to see ranks within each country (with respect to NRs instead of WRs).

Edit: http://hem.bredband.net/_zlv_/rubiks/stats/kinch_ranks.html (thanks Kit) (still waiting on by country though!)
 
Last edited:

Kit Clement

Premium Member
Joined
Aug 25, 2008
Messages
1,631
Location
Aurora, IL
WCA
2008CLEM01
YouTube
Visit Channel
I don't see why 2x2 would be any different to other events. In fact, it should appear to separate people who are close in shorter events more than in longer events.
The system reflects the fact that it's easier to improve from e.g. 8.1 to 8.0 at 2x2 than it is to go from 2.1 to 2.0. That's why the top people appear to have larger separation - their skill difference is much bigger imo. It's perfectly natural for there to be a spacing out at the top of any ranking (cubing or not).
The logarithm, while interesting perhaps, would go against the basic idea of the system. Not many people can interpret logarithmic scores - pretty sure we all work best in linear.
Sorry, that was the same thing explained many times badly :p

Consider this simple example -- to get a 50% score in 2x2, you need to have a 3.20 average. That takes quite a bit of skill, and likely takes knowing some EG, at the very least CLL. To get the same score in 5x5, you need to have a 1:48.40 average. That's a considerably easier level to attain -- and it's just because the numerator for 2x2 is so small. It's even sillier for FMC -- getting that score takes a solve of 40, or a mean of 50, which someone can do having never tried an FMC attempt, assuming basic 3x3 knowledge.

I do agree though -- It is rather desirable that the score function for an event is convex, but I'm just arguing that it may be a bit too convex. I do see the appeal for ease of interpretability though, and I'm starting to see how applying logs is just getting arbitrary.
 

tseitsei

Member
Joined
Jan 12, 2012
Messages
1,374
Location
Tampere, Finland
WCA
2012LEHT01
ANOTHER IDEA:
I mentioned that KinchRanks uses FM single and 3bld Single because overall they reflect abilities better than their respective Means still. However, I think we can improve on this by using the better score of a person's Single and Mean. Given that in the ideal world we'd use Means for these events, I think a good Mean should be rewarded with the option to use it, as it is a truer reflection of ability.
In the future I'd like to fully shift to Means only, but that is still over a year away imo.

3bld mean really just means how safely you want to play it... if I get "good enough" single on first solve then I will rush the remaining two attempts probably leading to DNF. But if I get reasonably bad times on 1st and 2nd attempts I'll probably just attempt to get asome kind of a good time without rushing it on the final attempt. After all winner is decided by the single and not mo3..

So mo3 3bld doesnt reqlly represent skill. So pls don't use that in your ranks...
 

not_kevin

Member
Joined
Jul 23, 2008
Messages
495
Location
California, USA
WCA
2008YOUN02
YouTube
Visit Channel
3bld mean really just means how safely you want to play it... if I get "good enough" single on first solve then I will rush the remaining two attempts probably leading to DNF. But if I get reasonably bad times on 1st and 2nd attempts I'll probably just attempt to get asome kind of a good time without rushing it on the final attempt. After all winner is decided by the single and not mo3..

So mo3 3bld doesnt reqlly represent skill. So pls don't use that in your ranks...

Yeah, I kinda agree with this - in particular, as long as competitions continue to use bo3 instead of mo3 for BLD, I don't think we should switch to using it, because the current times aren't optimized for it.

I think that we can eventually switch to FMC means, 'tho, since it's getting increasingly popular to hold mo3 competitions rather than best-of competitions.
 

obelisk477

Member
Joined
Aug 26, 2010
Messages
1,144
Location
Raleigh, NC
WCA
2009BATT01
YouTube
Visit Channel
Sorry if I missed this, but why wouldn't you divide everyone's total score by the first place holder's total score (and then multiply by 100, if you wanted it to look nicer)? So its more obvious where you are and who is at the top. So Feliks would score 100, Yu Nakajima 99.7, etc.

EDIT: I of course mean this for the sum of the kinch ranks only, and not the individual events themselves
 
Last edited:

Tim Major

Platinum Member
Joined
Aug 26, 2009
Messages
5,381
Location
Melbourne, Australia
WCA
2010MAJO01
Sorry if I missed this, but why wouldn't you divide everyone's total score by the first place holder's total score (and then multiply by 100, if you wanted it to look nicer)? So its more obvious where you are and who is at the top. So Feliks would score 100, Yu Nakajima 99.7, etc.

This should tighten up the graph a lot, so if you've missed something I've missed something too
 

kinch2002

Premium Member
Joined
Dec 22, 2008
Messages
2,504
Location
Guildford! UK!
WCA
2009SHEP01
YouTube
Visit Channel
3bld mean really just means how safely you want to play it... if I get "good enough" single on first solve then I will rush the remaining two attempts probably leading to DNF. But if I get reasonably bad times on 1st and 2nd attempts I'll probably just attempt to get asome kind of a good time without rushing it on the final attempt. After all winner is decided by the single and not mo3..

So mo3 3bld doesnt reqlly represent skill. So pls don't use that in your ranks...

Yeah, I kinda agree with this - in particular, as long as competitions continue to use bo3 instead of mo3 for BLD, I don't think we should switch to using it, because the current times aren't optimized for it.

I think that we can eventually switch to FMC means, 'tho, since it's getting increasingly popular to hold mo3 competitions rather than best-of competitions.
Thanks for the input. I've changed it for now so that you use your average if it gives you a better score than your single - I think that's still fair right? Will help those who haven't had a lucky single yet but have demonstrated their skill with an average.

Cool site. I'm 181 for Kinch-rankings as opposed to 92 for normal sum of single ranks.
Also, could we have average Kinch-rankings?
KinchRanks uses mostly averages already, apart from 4bld, 5bld, multibld. 3bld and 3fm uses whichever is better for you.

Sorry if I missed this, but why wouldn't you divide everyone's total score by the first place holder's total score (and then multiply by 100, if you wanted it to look nicer)? So its more obvious where you are and who is at the top. So Feliks would score 100, Yu Nakajima 99.7, etc.

EDIT: I of course mean this for the sum of the kinch ranks only, and not the individual events themselves
The reasons not to do this are
- The final score would no longer be an objective measure of how good you are overall
- Your final score would depend on the top person's overall score, which means that you can't track your own improvement properly.
 

not_kevin

Member
Joined
Jul 23, 2008
Messages
495
Location
California, USA
WCA
2008YOUN02
YouTube
Visit Channel
Thanks for the update on the main post - cool to see how the countries compare to each other. All the more confirmation that Poland is a huge cubing powerhouse - even the huge populations and competitive depth of the United States and China can't overcome Polish talent :p

Mostly as a personal request (getting something like this set up would take quite a few more resources, I'd imagine), but I'd love to see how people compare within their regions (like how sum of ranks is set up on the WCA site). We now know how strong the regions are compared to each other - but how strong am I within my region? I can calculate my own regional KinchRanks without too much difficulty (not including the 3bld and 3fm mean change, my score is 45.38 in NA), but how does that rank against the rest of NA?
 
Top