We prove a query complexity lower bound for QMA protocols that solve approximate counting: estimating the size of a set given a membership oracle. This gives rise to an oracle A such that SB P A Q M A A , resolving an open problem of Aaronson [2]. Our proof uses the polynomial method to derive a lower bound for the SBQ P query complexity of the AN D of two approximate counting instances. We use Laurent polynomials as a tool in our proof, showing that the "Laurent polynomial method" can be useful even for problems involving ordinary polynomials.