Hi everyone,

I would also like to add my congratulations to the champions in the
various competitions (so like, 5 times to Matyas??) and to the
finalists. I very much enjoyed meeting many new (and fast!) cubers as
well as those from WC05 and 03.

From the results of the 3x3 speedsolve, I am compelled make some
observations. I am surprised that no one has made a post about this; I
know I'm not the only one who has been thinking in this direction.

Of the 16 finalists, I would say that half or more had at least some
chance of winning the competition (of course, some had a greater
chance than others). Yu Nakajima also led the second round in average,
but his 14.54 in the first round shows that he can have a relatively
slow average on a bad round. Andrew Kang had a 15.84 first round and
Mitsuki Gunji, too, finished the second round with a 14.92. With one
more second on any solve that counts in the average, Mitsuki wouldn't
even have made the final. At the other, unlucky end are Jean Pons, who
had a 12.48 in the first round but just missed the podium in the
final, and Edouard Chambon, Thibaut Jacquinot, and Harris Chan, who
have all shown themselves to be a better cuber than me but who did not
perform their best in the final. Had luck played out differently, we
could have seen a very different final ranking.

The problem is that being the best just means that you have a greater
chance of winning the competition than anybody else, not that you are
going to win it. Suppose Yu had made a few mistakes, and suppose I had
some U permutations and had gotten a sub-13 average, which is very
possible. I would have won the competition by luck. You would have had
a most unwilling champion, apologizing to both the Japanese and the
French and desperately explaining to the media that it was a fluke
(not that most of them would have cared). With there having been a
pretty realistic possibility of something like this happening, we
should be happy that Yu, a cuber we can agree is one of the very
fastest in the world, won the competition.

I like World Championships because they give us a good measure of the
current level of the top cubers, but I'm becoming more and more
uncomfortable with the idea of the title of "World Champion." Simply
because of the role that luck plays in speedcubing, that title means
just that the cuber who holds it won the World Championship, not
necessarily that he is the best in the world, as it is usually taken
to imply by the media. If no method that improves on Fridrich is
discovered, in two years there will be even more clustering at the
top. The difference between first and tenth may even be the luck on
the permutations. Are five solves in the final round of one
competition, world championship though it may be, enough to choose the
World Champion for the next two years...the one cuber that, thanks to
the media, will be best known by the non-cubers around the world as
presumably the best cuber in the world?

In any case, I know we have to choose a World Champion based on just
the results of the World Championship (or is there some sport where
the world champion is not the winner of the world championship?).
Somebody has suggested using all solves from a competition to
determine the overall ranking separate from the ranking in the final
round, and there are also other ideas that I know some people have. I
understand that it's difficult to change the system we already have,
but I find this important enough that it should at least be given a
discussion.

It's obvious that no single competition is enough to determine who the
best cuber is. The World Championship should not be thought of as the
competition to determine the best cuber but the competition to decide
the World Champion, no more.

How, then, can we best determine the best cuber during a certain
period? What I'd like to see is some average of all averages that each
cuber had during a certain year on the WCA database. I know there are
many problems to ranking by this approach--setting the minimum number
of averages, cubers can theoretically stop competing for the year to
give themselves a better rank, etc--but it would still be a more
accurate measure of a cuber than any that exists today.

On a completely different WC topic, are we not counting Stefan's
Megaminx solve as a world record? There's no way to tell from the
database which solve within the same round came first, but can't this
be entered in manually, if we do accept this as a world record? I have
personally encountered two other similar cases: Nathaniel Christian's
4.20 2x2 at Horace Mann 2005, which, if accepted, is very possibly the
world record that lasted the shortest time (about 10 seconds), and my
own 2:50.32 3x3 blindfolded at Caltech Spring 2005.

Finally, I apologize for getting third in 3x3 blindfolded. That was
possible only because of the easy solve, meaning because of luck.
Tyson, when we get something to eat the next time, I'm paying.

Best,
-macky