Recall that the period of a polynomial p(x) is the smallest n such that p divides x^n-1.The average is known to be of the order of q^n . I am interested in order of magnitude of the tail of the distribution...Say how many polys have period at most Kn for some fixed constant K.