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.