[Unofficial] Weird 3x3 PLL skip

[Feanaro]

As you will see in the video, my friend complains about never getting a skip at all, and then gets one on that solve. I thought that was kind of weird

 

[IamWEB]

A)This was real
B) The cube fairy did it
c) Planned, bad acting

Notice how the answer getting more realistic the further into the alphabet you go.

J?K

 

[Feanaro]

This was totally real, why would I fake this? I only realized he said the whole thing about never getting skips when I reviewed the video. Besides, we were originally just racing because my OH times are comparable to his 2 handed times.

 

[Berry]

Haha nice I do the same thing with my little sister but her times are getting good really fast.

 

[imaghost]

Don't you have a better chance getting PLL skip with 2 look OLL? It makes sense, you do the OLL more times, making it a higher chance of getting a PLL skip other than doing multiple OLL's. I have gotten them like 2 out of 10 times. They don't come that often, but I get maybe 5 a day, then again I solve the cube more than 100 times a day.

 

[AvGalen]

Don't you have a better chance getting PLL skip with 2 look OLL? It makes sense, you do the OLL more times, making it a higher chance of getting a PLL skip other than doing multiple OLL's. I have gotten them like 2 out of 10 times. They don't come that often, but I get maybe 5 a day, then again I solve the cube more than 100 times a day.

No, 2 look OLL doesn't give you a better chance of PLL skip.
Situation 1: use both looks; After your first OLL you might have a PLL skip, but then you would still need to orient corners
Situation 2: edge OLL skip; After you skipped your first OLL you are going to do a normal PLL with 1/72 chances of skipping it (unless you use COLL or another variation)
Situation 3: corner OLL skip; After you do your first OLL you are going to skip the second OLL and then do a normal PLL with 1/72 chances of skipping it

If you know 10 algs for orienting edges and 10 algs for orienting corners you DO have a > 1/72 chance of skipping PLL, but only if you think about which combination of these algs you should use. This can be useful for FMC, but not really for speed

PLL skips can come really randomly. I have done 900 solves and only gotten 4. Then I had 4 in the next 8 solves!

 

[Johannes91]

Don't you have a better chance getting PLL skip with 2 look OLL?

No.

It makes sense, you do the OLL more times, making it a higher chance of getting a PLL skip other than doing multiple OLL's.

What?

I have gotten them like 2 out of 10 times. They don't come that often, but I get maybe 5 a day, then again I solve the cube more than 100 times a day.

Either you've been quite lucky, you know more than one alg per OLL case and look ahead to get a skip, or you just remember wrong. Count the skips next time you do a session to see if it really is 1/20.

If you don't do anything special to get a PLL skip, the way you solve F2L and OLL is irrelevant and the probability is 1/72.

 

[ccchips296]

A)This was real
B) The cube fairy did it
c) Planned, bad acting

Notice how the answer getting more realistic the further into the alphabet you go.

J?K

......do u have ANY proof of that?

 

[imaghost]

Don't you have a better chance getting PLL skip with 2 look OLL?

No.

It makes sense, you do the OLL more times, making it a higher chance of getting a PLL skip other than doing multiple OLL's.

What?

I have gotten them like 2 out of 10 times. They don't come that often, but I get maybe 5 a day, then again I solve the cube more than 100 times a day.

Either you've been quite lucky, you know more than one alg per OLL case and look ahead to get a skip, or you just remember wrong. Count the skips next time you do a session to see if it really is 1/20.

If you don't do anything special to get a PLL skip, the way you solve F2L and OLL is irrelevant and the probability is 1/72.

Well what I am saying, is that you do the same OLL in 2 look, making it a higher chance in getting a PLL skip. I only know 11 OLL's. I do 6 more than the other 5, since it is 2 look. The more times you do the same OLL, you have a better chance of getting a PLL skip.

 

[Ethan Rosen]

No that isn't even remotely true

 

[imaghost]

No that isn't even remotely true

How? I don't like comparing cubes to marbles, not that I ever do, but why not. If I have a bag of 10,000 green marbles, and one red marble. If I pull out one marble at a time, the chances of pulling out the red one start increasing. Would it not work for the cube?

 

[Ellis]

I having trouble even understanding what you're trying to argue, that marble analogy doesn't really help.

The way I see it is, yea, if you use two look OLL there is a slightly more of a chance that you get a PLL skip, but only in the sense that you could skip CO and PLL at the same time.

Do EO, then there's a 1/1944 chance that you'll skip CO+PLL, but if not, then there's still the 1/72 chance that you'll get a PLL skip after CO. So slightly more of a chance to skip PLL with 2-look OLL (if my logic isn't failing, which it could very well be), but nothing significant enough to even mention. Btw, someone please point out the hole in my logic, because I know there's something I'm not taking into consideration.

 

[hcbartek]

Don't you have a better chance getting PLL skip with 2 look OLL?

No.

It makes sense, you do the OLL more times, making it a higher chance of getting a PLL skip other than doing multiple OLL's.

What?

I have gotten them like 2 out of 10 times. They don't come that often, but I get maybe 5 a day, then again I solve the cube more than 100 times a day.

Either you've been quite lucky, you know more than one alg per OLL case and look ahead to get a skip, or you just remember wrong. Count the skips next time you do a session to see if it really is 1/20.

If you don't do anything special to get a PLL skip, the way you solve F2L and OLL is irrelevant and the probability is 1/72.

Yeaah, my English is bad, but 2 out of 10 times doesnt mean that is 4/20? 1 is out of 5 times.

 

[Johannes91]

The way I see it is, yea, if you use two look OLL there is a slightly more of a chance that you get a PLL skip, but only in the sense that you could skip CO and PLL at the same time.

Oh yeah, I didn't realize that. Thanks for explaining clearly what imaghost was probably trying to say. Not at all significant in practice, but an interesting point.

Edit: Heh nevermind.

 

[Stefan]

Do EO, then there's a 1/1944 chance that you'll skip CO+PLL, but if not, then there's still the 1/72 chance that you'll get a PLL skip after CO. So slightly more of a chance to skip PLL with 2-look OLL (if my logic isn't failing, which it could very well be), but nothing significant enough to even mention. Btw, someone please point out the hole in my logic, because I know there's something I'm not taking into consideration.

Probablity(PLL skip)
= Probablity(PLL skip with CO skip) + Probablity(PLL skip without CO skip)
= (1/27 * 1/72) + (26/27 * 1/72)
= 1/72

 

[mazei]

Trust me imaghost, once Stefan puts up numbers like that you are wrong.

 

[Stefan]

Trust me imaghost, once Stefan puts up numbers like that you are wrong.

Like I've never been wrong.

And another way to say it: Consider CFOP. PLL skip means that the LL pieces are correctly permuted at the point when CFO are done. If you use 2-look OLL, that point can be reached after the second OLL or after the first or even after zero if you have OLL skip. Or even right after the cross if your granny thought scrambling with five moves suffices. But it doesn't matter how early or late you get to that point. Yes you have a small probability to skip the entire rest of the solve early (like skipping FOP after C), but you don't gain PLL skip probability from that. If the probability to skip the rest of the solve (including PLL) after the cross is X, and you consider this, then you must also consider that you only continue solving (and later have 1/72) with probability 1-X. This 1-X btw was the 26/27 that was missing in Ellis' argument.

So if you want to do this really properly and consider all early rest-of-solve-including-PLL skips, the formula gets way messier but still results in 1/72 for PLL skip. The formula I gave above doesn't consider these early skips but only starting after F2L and EO.

The only way to really improve your PLL skip probability is to work for it, i.e., intentionally affect the permutation of the LL pieces before you get to PLL.

For imaghost I see these possible explanations (in order of decreasing probability):
1) He's wrong about the 5 PLL skips per day.
2) His "more than 100 solves a day" are really about 360 (5*72).
3) He does affect PLL before getting there.
4) He has been lucky.
5) He uses a bad scrambler and a "corresponding" solving method.

For case 5, consider a bad scrambler that only scrambles in <U,D,L,R,F2,B2>. If his method is to only solve using these moves, then of course he has a higher than normal OLL skip probability because edges will always be oriented. Similarly, a bad scrambler and "corresponding" solving method can lead to higher PLL skip probability.

P.S. No offense to grannies, I have no reason to believe they're any more stupid than the average person.

 

[Ellis]

Probablity(PLL skip)
= Probablity(PLL skip with CO skip) + Probablity(PLL skip without CO skip)
= (1/27 * 1/72) + (26/27 * 1/72)
= 1/72

If the probability to skip the rest of the solve (including PLL) after the cross is X, and you consider this, then you must also consider that you only continue solving (and later have 1/72) with probability 1-X. This 1-X btw was the 26/27 that was missing in Ellis' argument.

Thank you. Even with the numbers right in my face, it took me a few minutes to accept it. I would never have thought of it like that.

 

[imaghost]

I am not saying anybody is wrong, just logically, if you do something that has a random outcome, the more times you do the same thing, you have a higher chance of getting what you want. Yesterday, I had like 4-5 PLL skips, 2 being within like 5 solves. I probably solved it over 200 times yesterday. I timed about 100, and know I did a bunch more solves.

 

[Daniel Wu]

A)This was real
B) The cube fairy did it
c) Planned, bad acting

Notice how the answer getting more realistic the further into the alphabet you go.

J?K

A and B. It was real and the cube fairy did it.