# How the 3.47 WR changes cubing

#### Matt—

##### Member
I agree that this WR is far from unbeatable, but some of this math is pretty far off.

Let's eliminate the probability of nerves since that's so hard to estimate, just so we can get an idea of how many solves have WR potential. Given your other probabilities, we end up with a 0.000946 chance of a potential WR solve. I'm very skeptical of your 29.6% estimate for 2 three-move pairs in a row (and @Sajwo brings up some good points above), so let's be conservative and estimate a 0.0001 chance of a potential WR solve.

You can't add probabilities like you did in your post; the correct formula for the chances of a WR-potential solve is:
$1-(1-P)^n$ where P is the probability of a given single solve having WR potential, and n is the amount of solves done. Let's say we want a 50/50 chance of a WR-potential solve. This would require: $(0.9999)^n=0.5$

Solving for n we get approximately 6931 solves.

This means, assuming each solve has a 0.0001 chance of having WR potential if done by a world class cuber (which is a very hard number to accurately estimate), 6931 official solves would need to be done by world class 3x3 solvers in order for there to be a 50% chance of one of these solves showing up. Additionally like you mentioned, they would have to control their nerves, which is definitely quite difficult to do.

I'd be surprised if this record is broken before 2020, but it's totally possible for it to be broken. These calculations make too many assumptions no matter what, since we can't account for improvements in hardware and solving methods. There's always a chance of someone coming up with a method that gets 30-move solves on average, and in that case, a 10TPS average solve would beat the record.
This made me think of something.

I forgot to mention that the solve was a Petrus solve. That’s really cool even though it was meant to be CFOP. If he had meant to do Petrus and didn’t get an EO skip, it still could have been a sub 5 time if he was good at EO. This just shows that CFOP is not always the best method.

#### Cefe origol

##### Member
Well, his point was that it's impossible to see whole <20 solution in inspection time. Besides that your probability calculations for last 2 slots are wrong , you can't also assume that there is only one possible way to achieve fast solve (for example xcross and 2 easy pair with LL skip could also result in ~3s solve) and you don't have enough data from top level speedcubers about their solves to even calculate the probability of ~3s solve
NO, you just onelook the cross and 2 pairs. I also know that not every solve is the same but I'm giving an example.

Last edited:

#### alister

##### Member
So it wouldn't be very rare to see a sub-25 move WR
But has there ever been a sub-25 move WR? Seems kinda rare to me

The UWR of TPS in a solve is of 12.68.
Feliks has had a 13.4 TPS solve from 2 years ago:

And it would not surprise me if one of the top cubers has already managed to get a 14 TPS solve off camera. But yeah, I completely disagree with your theory about a sub 1.5 second solve though. If it ever happens, I don't think it will be soon.

#### squidgecubing

##### Member
This is just my opinion.

So, as many of you know the 3x3 WR was beaten. It went from 4.22 to 3.47. This is a huge leap in a world record, and could change cubing forever.

First off, the WR was beaten by a Cuber who almost no one knew. Everyone thought Max/Feliks/Jay would get it, but did not. Is this important? Yes. Very much so. Why? Because it shows that people who, even though they are unknown can still achieve major accomplishments. I am inspired by this and makes me want to try harder to get a world record someday.

Secondly, 3.47 is an insane time, and may possibly be the peak of the 3x3 single WR. This is a disappointment to some, but also a challenge to others. Or, it may even cause some to stop 3x3 and practice different cubes (skewb, 4x4, Megaminx, etc.) This May be a good thing as cubing is mostly centered around 3x3, and it is good to get people to practice other events and have fun with not only 3x3, but also with other events. (Or if you think one day you can beat the 3x3 WR, then this is an amazing challenge for you. - Good Luck)

And for my third point I will be discussing how this affects world records. This has already been discussed by many, and only now has there been evidence of this happening, is that the 3x3 WR has gotten to a point where it is almost unbeatable. Is the WCA going to add a yearly record or something? If not, then the 3.47 WR could stay forever.

These were just my thoughts on what might happen, and could be completely different. I have no clue, but I am excited to see the future of cubing.
Maybe the 3x3 average will become more recognized and sought after instead?

#### Cefe origol

##### Member
I agree that this WR is far from unbeatable, but some of this math is pretty far off.

Let's eliminate the probability of nerves since that's so hard to estimate, just so we can get an idea of how many solves have WR potential. Given your other probabilities, we end up with a 0.000946 chance of a potential WR solve. I'm very skeptical of your 29.6% estimate for 2 three-move pairs in a row (and @Sajwo brings up some good points above), so let's be conservative and estimate a 0.0001 chance of a potential WR solve.

You can't add probabilities like you did in your post; the correct formula for the chances of a WR-potential solve is:
$1-(1-P)^n$ where P is the probability of a given single solve having WR potential, and n is the amount of solves done. Let's say we want a 50/50 chance of a WR-potential solve. This would require: $(0.9999)^n=0.5$

Solving for n we get approximately 6931 solves.

This means, assuming each solve has a 0.0001 chance of having WR potential if done by a world class cuber (which is a very hard number to accurately estimate), 6931 official solves would need to be done by world class 3x3 solvers in order for there to be a 50% chance of one of these solves showing up. Additionally like you mentioned, they would have to control their nerves, which is definitely quite difficult to do.

I'd be surprised if this record is broken before 2020, but it's totally possible for it to be broken. These calculations make too many assumptions no matter what, since we can't account for improvements in hardware and solving methods. There's always a chance of someone coming up with a method that gets 30-move solves on average, and in that case, a 10TPS average solve would beat the record.
1- I actually used that same equation, but I needed to change the probability a lot of times. Thats the reason why I added nerves, else after 4 comps my calculator said 100 percent. So I probably wrote something wrong or you don't know how to sum.
2-the 29.6 is actually wrong, thanks. Here is how I found it out. For the first pair their are 17 three-move cases:
1. R U R'
2. U R U R'
3. U2 R U R'
4. U' R U R'
5. R U' R'
6. U R U' R'
7. U2 R U' R'
8. U' R U' R'
9. F' U F
10. U F' U F
11. U2 F' U F
12. U' F' U F
13. F' U' F
14. U F' U' F
15. U2 F' U' F
16. U' F' U' F
17. (SOLVED)
Many wont consider all of these 3 moves, but I still do and counted the extra moves for the 3.6 time.
for both cases it is around 578(17*17*2) and the total cases are 32400. Giving us a total of 1.7 percent. 0.051 percent in total. 1358 times for a 50/50 chance.

#### Tabe

##### Member
Wow, all of you have excellent points here. I like what was said about the averages showing the skill of a Cuber. That’s what truly means everything. My point though (and I have not mentioned this previously) is that if a non Cuber was interested in starting cubing, then learned beginners decided to check out the world record. It may make them lose their motivation to cube.
Has anybody ever decided not to try track because Usain Bolt broke the 100m world record yet again? Of course not.

#### Matt—

##### Member
Has anybody ever decided not to try track because Usain Bolt broke the 100m world record yet again? Of course not.
I’m not saying that, all I’m saying is that new Cubers who already think that solving a cube is the most complicated thing in the world can’t even comprehend trying to get to 3.4, and with no help from friends may lose motivation quickly. (I did for a while, and it wasn’t until I decided to look up an advanced tutorial to start cubing again.)

#### bobthegiraffemonkey

##### Member
I'd recommend not taking lucky singles too seriously, average is generally seen as a better indication of skill (though getting sub 4 on a good scramble still takes a lot of skill).

I'd also recommend making sure the main reason you cube is to have fun. It's a hobby. Not that going for WR is a bad thing, but if it's all you care about you're probably going to get bored easily, especially if you're just starting out.

#### Kit Clement

Friendly reminder that 10 years ago, we all thought 7.08 was essentially unbeatable.

#### Dylan1919

##### Member
I am supportive of this new contender however I feel bad for Feliks, Matts, Max and Jay because if they got that scramble they probably would of had a good chance as well.

Friendly reminder that 10 years ago, we all thought 7.08 was essentially unbeatable.
Yes, but now its down to 3 seconds and it is highly unlikely to be beaten by a margin.

#### mark49152

##### Super Moderator
Staff member
Friendly reminder that 10 years ago, we all thought 7.08 was essentially unbeatable.
This. When I started cubing the WR was 5.66 and I thought it was unbeatable. Today it would rank him 59th.

#### Matt—

##### Member
This. When I started cubing the WR was 5.66 and I thought it was unbeatable. Today it would rank him 59th.
Maybe one day people will get so skilled at 3x3 that we will be averaging 4 seconds. One day...

This also may never happen, I’m just saying that it could be possible.

#### AbsoRuud

##### Member
When I started cubing the WR was 4.22. I am never going to beat that. I'll probably never be sub 10. But that's beside the point. Cubing is fun. That's the whole point. If you get so good to the point where you can beat WRs, all the better. But it has never been my goal to beat the WR and it never will be. I'm happy when I am faster than me.

#### Sajwo

##### Member
Maybe one day people will get so skilled at 3x3 that we will be averaging 4 seconds. One day...

This also may never happen, I’m just saying that it could be possible.
Max has 4.99 ao12. It's certainly possible to get sub5 ao50 or ao100

#### Julio974

##### Member
Well, I put all WR Singles on a graph (except Minh Thai's 22.95), and made an exponential reduction (which approximates the trend).
With the current trend, this record will probably be broken around the beginning of 2020. I remind that it just uses probabilities with a very bad model!
Here's the graph for those interested:

#### Mike Hughey

##### Super Moderator
Staff member
Nice graph! I think it's fascinating that the graph seems to indicate that the 7.08 was actually more unexpected than the 3.47 is.

#### Gomorrite

##### Member
The 7.08 was 63% of the average WR at the time by Yu Nakajima, while this is one is 60% of Feliks Zemdeg's 5.80.

#### Kit Clement

Yes, but now its down to 3 seconds and it is highly unlikely to be beaten by a margin.
How do you calculate "highly unlikely" here? You could have said that any WR will be unlikely to be beaten because it's now so much lower and there's fewer times that could beat it.

imo, i bet this wr will stand for at least 5 years
I'd bet a large sum of money that this is broken within 5 years in a heartbeat.

Last edited:

#### Matt—

##### Member
How do you calculate "highly unlikely" here? You could have said this at just about any WR that it will be unlikely to be beaten because it's now so much lower and there's fewer times that it could beat it.

I'd bet a large sum of money that this is broken within 5 years in a heartbeat.
Yep, anything is possible. Who knows, a new, move efficient method may be developed. Or, a ZZ user or a Roux user may get a record and that could completely change the way people see other methods.