{{
`${userDetail.user.displayName}${userDetail.user.cachedBungieGlobalDisplayNameCode ? `#${standardizeBungieGlobalCode(userDetail.user.cachedBungieGlobalDisplayNameCode)}` : ""}`}}
{{getUserClans()}}
Edited by steppinrazor:
12/1/2014 3:27:42 PM
2
So, voidfangs happening 4 times in a row is less likely than every other combination added together. Sure. But, whatever the next four weeks bring, at the end of that 4 weeks, it will have been 1/6 * 1/6 * 1/6 * 1/6 to have happened.
Let's pretend that for the 4 Fridays in December, he sells PF, Nemesis, Sunbreakers, Veil. 1/6 * 1/6 * 1/6 * 1/6. Or he sells Voidfangs, Sunbreakers, Veil, PF. 1/6 * 1/6 * 1/6 * 1/6. Veil, Veil, Sunbreakers, PF. No matter what, after 4 Fridays, there is going to be a specific order. So at the end of December, I can say, 'look how unlikely it is for him to have sold X this month' and it would seem ridiculously unlikely.
You cannot use the results of 4 trials to make any accurate assessment. You need huge samples to approach the true probability. This is the Law of Large Numbers.
English
-
I'm not saying that next week has been affected , I'm saying that 4 of the same items to occur in 4 weeks is the most unlikely thing that could occur. The chance of getting 3 in 4 weeks is significantly higher. Say there is 3PF and 1 void fang in the 4 weeks, the chance of that is not 1/6*1/6*1/6*1/6 It could occur as pf, pf, pf, void Or pf, pf, void, pf Or pf, void, pf, pf Or void, pf, pf, pf Therefore the chance is 1/6*1/6*1/6*1/6+1/6*1/6*1/6*1/6+1/6*1/6*1/6*1/6+1/6*1/6*1/6*1/6 Therefore even the next lowest chance of a combination in 4 weeks to occur is 4 times more likely
-
It will only have the same result if you are talking about the items occurring in that exact order, because you can have those items occur in multiple orders which add together and increase the probability
-
It has to occur in some exact order over 4 weeks. He is going to sell something, then something else, then something else, then something else. There will be a specific result, there is no way around that. It is completely true that selling voidfangs 4 times in a row is less likely than every other combination added together. It just doesn't matter, because of what I wrote above - at the end of 4 weeks, there will be a specific order. That is why you shouldn't take 4 results of a trial and make a judgement about it.
-
This is similar to tossing one hundred coins where 75% are heads and me saying it's odd and unlikely You then saying, yeah, but it's just as likely to get 60% heads. The exact sequence doesn't matter, we are looking at the total occurrences inside this 4 weeks. Which is very unlikely to be 4 of the same item
-
[quote]The exact sequence doesn't matter[/quote] This is the whole problem. After the fact, there [b]is[/b] an exact sequence. You can't point to that sequence and say, 'look, that was really unlikely' as if it's some kind of proof it's not RNG. You're mistaken in your coin analogy. What I am saying is that looking at one hundred coin toss [i]results[/i] does not produce a sample upon which one can judge the randomness of coin tosses. Any time you look at results over a specific number of trials, you will have a standard deviation. This is the deviation from the mean, or the frequency at which it should occur. I'll just be repeating myself if I continue, but I hope you can see that making a conclusion based on the results of 4 trials is pointless.
-
Plus every sequence has the same possibility of occurring, however looking at the overall outcome within the sequences timeframe, you have to agree that getting every outcome the same is far more unlikely
-
We are making the conclusion because they occurred in a row, we don't care about the previous weeks
-
But we are not commenting on the order each voidfang came out, cos it's all the same. We only need to refer to the content inside those 4 weeks
