Odd WCA stats/ Stats request Thread

[AvGalen]

Update Requested by Anders Larsson: http://hem.bredband.net/_zlv_/rubiks/stats/travel_ALL.html (Most distance traveled)

so fun to see this statistic. And I am very happy to see that someone did exactly 1000 marathons (42.195 km)
I guess "most traveled couple" is Hanneke and Ron, unless you count Adam and Shelley as a couple (but if you do that Lars and I should count as well!)
I also drop from nr. 4 to nr. 374 if you view it as distance per comp, so I can now rationalize that I really don't travel very much to go to a comp.

(this metric is of course extremely flawed because when I live in NL but go to a comp in Turkey and Portugal I actually do go home in between. But I didn't do that when I was travelling for 1.5 year so there is no good, automatic, way to do a proper calculation)

 

[Lid]

A question that came up this weekend: who has the greatest number of 10.xx 3x3 averages without a sub 10 average? I've got eleven now.

You're not alone up there:

19	[url="http://www.worldcubeassociation.org/results/p.php?i=2008LINH01"]Hao-Zheng Lin (林浩正)[/url], Taiwan
18	[url="http://www.worldcubeassociation.org/results/p.php?i=2007LUCH01"]Chia-Wei Lu (呂家維)[/url], Taiwan
15	[url="http://www.worldcubeassociation.org/results/p.php?i=2010CHEN20"]Lin Chen (陈霖)[/url], China
15	[url="http://www.worldcubeassociation.org/results/p.php?i=2007YOSH01"]Takumi Yoshida (吉田匠)[/url], Japan
14	[url="http://www.worldcubeassociation.org/results/p.php?i=2011ZVYA01"]Dmitry Zvyagintsev[/url], Russia
12	[url="http://www.worldcubeassociation.org/results/p.php?i=2010HIRA01"]Peerawich Hiranpaphakorn (พีรวิชญ์ พิรัญปภากร)[/url], Thailand
11	[url="http://www.worldcubeassociation.org/results/p.php?i=2008SIMO01"]Riccardo Simone[/url], Italy
11	[url="http://www.worldcubeassociation.org/results/p.php?i=2011KNOT01"]Wojciech Knott[/url], Poland
11	[url="http://www.worldcubeassociation.org/results/p.php?i=2007SUNT01"]Timothy Sun[/url], USA
11	[url="http://www.worldcubeassociation.org/results/p.php?i=2008CRAW01"]Simon Crawford[/url], United Kingdom
11	[url="http://www.worldcubeassociation.org/results/p.php?i=2008HASH02"]Wataru Hashimura (端村航)[/url], Japan
11	[url="http://www.worldcubeassociation.org/results/p.php?i=2006BARL01"]Thom Barlow[/url], United Kingdom
10	[url="http://www.worldcubeassociation.org/results/p.php?i=2007TAMA02"]John Tamanas[/url], USA

 

[10461394944000]

19	[url="http://www.worldcubeassociation.org/results/p.php?i=2008LINH01"]Hao-Zheng Lin (林浩正)[/url], Taiwan

apparently he has very bad luck

9.21 9.78 44.90 9.99 25.98

 

[kinch2002]

Robert Yau found the only(?) way to set multiple NRs in the same event on the same day
Nottingham Summer 2014 Skewb

 

[Carrot]

Robert Yau found the only(?) way to set multiple NRs in the same event on the same day
Nottingham Summer 2014 Skewb

how?

 

[kinch2002]

how?

By breaking a record twice with the same result (6.84 average in both rounds)

I'm sure people are aware of
9i2) All the results of a round are considered to take place on the last calendar date of the round. If a regional record is broken multiple times on the same calendar date, only the best result is recognised as breaking that regional record.

As there are 2 'best results', both are records.

 

[Robert-Y]

Off-topic: I hope that means I get twice the record points for fantasy cubing...

On-topic: Another (very) dense 2x2x2 podium: http://cubecomps.com/live.php?cid=527&cat=2&rnd=2

 

[Carrot]

By breaking a record twice with the same result (6.84 average in both rounds)

I'm sure people are aware of
9i2) All the results of a round are considered to take place on the last calendar date of the round. If a regional record is broken multiple times on the same calendar date, only the best result is recognised as breaking that regional record.

As there are 2 'best results', both are records.

So you are saying that it's more impressive to tie a NR than beating it? makes sense...

 

[kinch2002]

So you are saying that it's more impressive to tie a NR than beating it? makes sense...

When did I say it was more impressive? I said it happened, and implied that it's quite an interesting/cool thing to happen, that's all

 

[cubizh]

When did I say it was more impressive?

Well, it kinda is, statistically only. Matching a given average is considerably harder than to break it (or go over it) as the number of reasonable solutions for (t1+t2+t3)/3 = z is far less than for (t1+t2+t3)/3 < z.
Of course, it is not, if you think of it in sporting terms.

 

[Sebastien]

So you are saying that it's more impressive to tie a NR than beating it? makes sense...

He did not tie the NR. He did in fact beat the NR twice and finished being tied NR holder with himself at the end of the calendar day when the records were evaluated.

 

[Carrot]

When did I say it was more impressive? I said it happened, and implied that it's quite an interesting/cool thing to happen, that's all

He did not tie the NR. He did in fact beat the NR twice and finished being tied NR holder with himself at the end of the calendar day when the records were evaluated.

There's something you two see that I don't see...

https://www.worldcubeassociation.or...ngdom&years=&show=100+Persons&average=Average
and
https://www.worldcubeassociation.or...ited+Kingdom&eventId=skewb&years=&mixed=Mixed

Either I miss something or it's inconsistency, if tied NRs (by same person) are handled as both being equally good (hence the record page), why is one of them ranking better on the ranking? so one place one is better, and the other place they are equally good? Yes, I find the double NR loophole stupid.

 

[TDM]

if tied NRs (by same person) are handled as both being equally good (hence the record page), why is one of them ranking better on the ranking?

Neither is better?

 

[Pedro]

so fun to see this statistic. And I am very happy to see that someone did exactly 1000 marathons (42.195 km)
I guess "most traveled couple" is Hanneke and Ron, unless you count Adam and Shelley as a couple (but if you do that Lars and I should count as well!)
I also drop from nr. 4 to nr. 374 if you view it as distance per comp, so I can now rationalize that I really don't travel very much to go to a comp.

(this metric is of course extremely flawed because when I live in NL but go to a comp in Turkey and Portugal I actually do go home in between. But I didn't do that when I was travelling for 1.5 year so there is no good, automatic, way to do a proper calculation)

That's me! I laughed a lot when I saw that. My average per comp is 1834, but I don't usually go that far. The 3 international trips actually inflated that number quite a bit.

Who did that statistic? How exactly did you calculate those?

 

[Lid]

Who did that statistic? How exactly did you calculate those?

That would be me

Your starting point is your first competition, then you "travel" the shortest distance between each competition.
Coordinates for each competition are given in the WCA_export_Competitions file (so there could be errors there).
The distance was calculated using the Haversine formula. Everything was done with a small Perl-script.

 

[vcuber13]

By breaking a record twice with the same result (6.84 average in both rounds)

I'm sure people are aware of
9i2) All the results of a round are considered to take place on the last calendar date of the round. If a regional record is broken multiple times on the same calendar date, only the best result is recognised as breaking that regional record.

As there are 2 'best results', both are records.

Is that a relatively new regulation? Because there are many results that do not follow that, such as Melbourne Winter 2011

Edit: It looks like it was changed last year.

 

[kinch2002]

There's something you two see that I don't see...

https://www.worldcubeassociation.or...ngdom&years=&show=100+Persons&average=Average
and
https://www.worldcubeassociation.or...ited+Kingdom&eventId=skewb&years=&mixed=Mixed

Either I miss something or it's inconsistency, if tied NRs (by same person) are handled as both being equally good (hence the record page), why is one of them ranking better on the ranking? so one place one is better, and the other place they are equally good? Yes, I find the double NR loophole stupid.

I was never talking about those rankings pages, although it's an interesting point that you bring up. It surely applies to anyone who has tied their personal best? Even if they don't appear on the records page, they will only have one result on the rankings page despite having a tied PB. I don't have a problem with it personally.

The "double NR loophole" is only to do with setting NRs on the same day, not to do with those rankings pages.

For what it's worth, I find the regulation of records being at the end of the day to be quite silly, especially for National Records. I mean, has there even been a case of the same NR being set at 2 different comps on the same day? I can see the logic for World Records, but National Records could easily be defined at the end of each round.

Here's a weird situation: Comp A has 3 FM attempts across Saturday and Sunday. Person A sets a WR 19 moves on Saturday.
At Comp B there is 1 attempt on the Sunday. Person B gets 18 moves.
That's going to hurt hard for Person A, because he doesn't even get it for a day.

Here's another weird situation: Comp A has 3 5bld attempts to be done at any time during Saturday and Sunday. Person A set a new WR of 5 minutes on the Saturday.
Comp B has the same format and Person B sets a WR of 4 minutes on the Sunday.
Person A again will feel very hard done by. Although, the organisers of Comp A could choose to force the round to end on the Saturday so that Person A gets his WR for a day.

As well as the harshness of losing records in later rounds, this rule of "records at the end of the day on which the last attempt of the round falls" causes multiple strange situations. I haven't seen evidence that it has helped in any situations compared to just saying that records are set at the end of each round, and in the very rare situation that records are set in different competitions, the absolute time should be used to determine who set it first. I'd like to see an example where that old rule was actually a problem, if anyone has one?

p.s. This post is straying off the topic of the thread, so should maybe be moved to a new thread if it goes further.

 

[Lucas Garron]

As well as the harshness of losing records in later rounds, this rule of "records at the end of the day on which the last attempt of the round falls" causes multiple strange situations. I haven't seen evidence that it has helped in any situations compared to just saying that records are set at the end of each round, and in the very rare situation that records are set in different competitions, the absolute time should be used to determine who set it first. I'd like to see an example where that old rule was actually a problem, if anyone has one?

Is there an example where the new rule has been a significant problem? In particular, I haven't heard of any cases where this has happened for big/multi BLD attempts or FMC. (Those events are also weird because we schedule and think of each attempt in the same round the way we think of an individual round in other events – in the US it also used to be common to refer to them incorrectly as "rounds".)
Some people have commented on the change, but no one has really *complained*.

The new rule was meant to simplify things for the results team (they have more than enough work), which now has to deal with a lot of simultaneous competition results coming in at different times. I don't know how many of those have their rounds labeled with dates, but at least in theory deciding who has the final record is trivial once all the results are in.
It also takes away a bit of the visceral unfairness that comes from happening to do your solves before/after someone one the same round/day, or in a different time zone.

Back in 2007, Dan Dzoan thought he set the WR single in California, with 10.08. However, Thibaut Jacquinot had set a 9.86 in Spain earlier. If the time zones had been reversed, Dan would have had the world record. Different people will have different intuitions about the best way to handle this, but almost any procedure will have cases that will feel weird or unfair to some people. Until we've thought about this thoroughly, we might as well pick something simple.

Anyhow, the Regulations aren't fixed in stone. This could change, but that would be a separate conversation, and not something that would be a high priority for the WRC.

Trivia: also recall that in 2008 Yu Nakjima tied a WR single on the same day (after setting the average).

 

[Antonie faz fan]

That would be me

Your starting point is your first competition, then you "travel" the shortest distance between each competition.
Coordinates for each competition are given in the WCA_export_Competitions file (so there could be errors there).
The distance was calculated using the Haversine formula. Everything was done with a small Perl-script.

But what if someone moved to the other Side of the world?

 

[Bindedsa]

But what if someone moved to the other Side of the world?

That's interesting. I live in South Africa and have been to one comp here, but will be moving to the US sometime soon. Does that mean every competition I go to would count as that distance(over 7500 miles)?.