• 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!

Probability Thread

cmhardw

Premium Member
Joined
Apr 5, 2006
Messages
4,115
Location
Orlando, Florida
WCA
2003HARD01
YouTube
Visit Channel
Last edited:

Carrot

Member
Joined
Feb 9, 2009
Messages
1,910
WCA
2008ANDE02
YouTube
Visit Channel
thanks I didn't know how to calcute the percentage though

jeez... percentage... derives from 'per cent' (kind of ish) which roughly means 'hundredths'. so basically multipy any number by 100 and you have the number represented as percentage. (0.25 = 25%, 0.01 = 1%, 25 = 2500%) It's THAT simple

EDIT: Ben, y u ninja?
 

kinch2002

Premium Member
Joined
Dec 22, 2008
Messages
2,504
Location
Guildford! UK!
WCA
2009SHEP01
YouTube
Visit Channel
What would the prob. be of getting 3 LL skips in about 17k solves? (this is my situation)

I believe this should just be a binomial probability, right?
(17000 choose 3) * (1/15552)^3 * (15551/15552)^16997 = 7.3% chance of 3 LL skips in 17000 solves.

Indeed, but your question is needlessly specific, resulting in an answer that exaggerates your true 'luckiness'. Less misleading would be the probability of getting at least 3 LL skips in 17k solves, which is 9.82%
 

Robert-Y

Member
Joined
Jan 21, 2009
Messages
3,289
Location
England
WCA
2009YAUR01
YouTube
Visit Channel
^People need to stop forgetting this :p

On a related note... I was wondering how unlucky you have to be to get x solves in row without a LL skip. Here are the probabilities that you will receive at least one LL skip using the standard CFOP method without any special tricks, for certain x number of solves in a row:

100 solves: 1 - (15551/15552)^100 ~= 0.64%
500 solves: 1 - (15551/15552)^500 ~= 3.16%
1000 solves: 1 - (15551/15552)^1000 ~= 6.23%
5000 solves: 1 - (15551/15552)^5000 ~= 27.5%
10000 solves: 1 - (15551/15552)^10000 ~= 47.43%
50000 solves: 1 - (15551/15552)^50000 ~= 95.98%
100000 solves: 1 - (15551/15552)^50000 ~= 99.84%

So you could say that if you've done at least 50000 solves in your lifetime, you have probably received at least one LL skip hopefully :)
But if not, it doesn't mean you're *that* unlucky since the chances that you haven't received one in 50000 solves is ~= 4% or 1/25.
 

advincubing

Member
Joined
Jun 12, 2012
Messages
93
YouTube
Visit Channel
What is the probability of the last F2L corner + edge set being already paired in the top layer, without manipulating the earlier inserts to force pairing? (Relates to question of how often partial edge control -- sledgehammer vs. RU'R' and their mirrors -- can be applied through the last insert.) The calculation is easy enough if someone can point me to a summary of F2L case probabilities....

Thanks.
 

Tempus

Member
Joined
Apr 3, 2014
Messages
72
What is the probability of the last F2L corner + edge set being already paired in the top layer, without manipulating the earlier inserts to force pairing?
Assuming I understand you correctly, there should be five corners and five edges in an indeterminate state. As for the corner piece, there's a 4/5 probability that the relevant corner is in the top layer and a 2/3 probability that the cross color on the relevant corner isn't facing up. As for the edge piece, there's now a 1/5 probability that it's against the correct face of the corner piece and a 1/2 probability that it's oriented correctly.

Multiply them all together and you get p=4/75, or about a 5.333% chance that they are matched up together in the top layer.
 

advincubing

Member
Joined
Jun 12, 2012
Messages
93
YouTube
Visit Channel
Assuming I understand you correctly, there should be five corners and five edges in an indeterminate state. As for the corner piece, there's a 4/5 probability that the relevant corner is in the top layer and a 2/3 probability that the cross color on the relevant corner isn't facing up. As for the edge piece, there's now a 1/5 probability that it's against the correct face of the corner piece and a 1/2 probability that it's oriented correctly.

Multiply them all together and you get p=4/75, or about a 5.333% chance that they are matched up together in the top layer.

Thank you for the response. That methodology makes perfect sense, but the result seems surprisingly low. I feel like I get a made pair in the top layer more like 25-40% of the time. I wonder if the difference is due to something I didn't think to address/describe in my question: Often the last F2L edge and corner are setup so that they can be paired before being inserted. This would be a case like F2L #5 where you would do (U' R U R') to pair the edge and corner in the U layer, AUF, then have the option to insert with R U' R' or a hedgeslammer. Those cases are in contrast to the ones that are solved and inserted in one fell swoop with, say, R U R'.

That complicates things quite a bit, I know. Not sure the best methodology for calculating with that nuance....
 

goodatthis

Member
Joined
Jan 11, 2014
Messages
841
Location
NY
WCA
2014CAVA01
YouTube
Visit Channel
Okay, kind of a complicated question here, but is there any difference between the probability of getting a PLL skip when you use OLL (non edges oriented) vs OCLL? Basically what I mean is, if a PLL skip is just a 1LLL case (or ZBLL) that you happened to get as your OLL, and since OLLs permute pieces as well, do non edges oriented OLLs permute in a more common pattern than OCLLs? (Or less common?)

For example, let's say that a certain OLL permutes in a U perm like fashion. U perms are common, so another OLL that permutes in an H perm (less common) like fashion should be less likely to cause an PLL skip, right? I'm sorry if I'm a noob at this kind of thing, haha.
 
Last edited:

ryanj92

Member
Joined
Dec 26, 2011
Messages
1,575
Location
Sheffield, UK
WCA
2012JONE03
YouTube
Visit Channel
Okay, kind of a complicated question here, but is there any difference between the probability of getting a PLL skip when you use OLL (non edges oriented) vs OCLL? Basically what I mean is, if a PLL skip is just a 1LLL case (or ZBLL) that you happened to get as your OLL, and since OLLs permute pieces as well, do non edges oriented OLLs permute in a more common pattern than OCLLs? (Or less common?)

For example, let's say that a certain OLL permutes in a U perm like fashion. U perms are common, so another OLL that permutes in an H perm (less common) like fashion should be less likely to cause an PLL skip, right? I'm sorry if I'm a noob at this kind of thing, haha.

No, because during OLL the permutation of your LL pieces is generally random, so the chance of the OLL and OCLL actually causing a PLL skip is pretty slim :p
Maybe the nuances you mention have some minute effect on the probabilities - however, it will definitely be minute and calculating the probabilities won't give us any profound results

Also, not everybody uses the same OLL's or even OCLL's anyway, so the probabilities will vary from person to person :p
 

qqwref

Member
Joined
Dec 18, 2007
Messages
7,834
Location
a <script> tag near you
WCA
2006GOTT01
YouTube
Visit Channel
a) What's the probability that a random 2x2x2 position can be completely identified by only seeing the U, F, R, and L faces?
b) Same as a, but the 6 faces are a random permutation of the 6 colors.
c) How about on a 3x3x3?
d) How about on a 4x4x4? :p
 
Top