You know what I can't accept about statistical probability? While it does indeed acurately measure past events, to accept it you also have to accept that outcom

You know what I can't accept about statistical probability? While it does indeed acurately measure past events, to accept it you also have to accept that outcomes are not random but somehow implementing causal events a certail way like in the solution to this problem will result in more likelihood of desired results. The only thing that this tells me is there is an underlying causal relationship. Otherwise, what is the alternative? That all events in reality are a complete throw of the dice and probability just makes sense of how things were pre-determined? Another issue I have is (taking the problem at hand for example), the applicants are ordered randomly right? Thay discards a key piecie of information on their ordering (first applicants want it most or they are undesirable for example). And when you randomize, that simply means you discarded the information about your new reordering. In reality, your arbitrary shuffling has a cause. If I asked you to pick a number at random, there is a reason why you arrived at that number and that reason or cause also has cause and so on, so nothing that is a product of cause can be truly random. Last point for this post, what does the pool of candidates look like? keeping it simple, 3 people apply, since you shuffled them, the desparate first applicant is middle,the mediocre last applicant stays last and the talented person that applied a bit late is first. So you get bad or mediocre. Since you get to shuffle only once, this ordering I mentioned really has two groups which are desirable top talent person and undesirable mediocre or bad hires. So if your shuffling is truly random it is similar to a coin toss where probability is 50%. My point with this is, in reality not just in hiring but in many other things the vitality curve applies where you really only want the top 10/20%. It might work if your goal is "the best of the mediocre middle" but if your goal is binary "top 10%" vs "bad hire" the implication here is your sampling of the first person is the equivalent if a coin toss but the coin toss here was really your one chance shuffling of applicants. Does this make sense? The solution makes sense to me if maybe you kept reshuffling their order after each interview.