Recently, Kolboe and Svelle pointed out that second-order Moller-Plesset perturbation theory (MP2) incorrectly predicts a barrierless reaction of the benzenium-ethene complex to the ethyl-1H-benzene cation in contrast to other considered quantum chemical methods [J. Phys. Chem. A 2008, 112, 6399]. In a subsequent Letter in this Journal, van Mourik related this behavior to the basis set superposition error [J. Phys. Chem. A 2008, 112, 11017]. Here we call show that this is not the case but that the failure is due to an intrinsic (overcorrelation) problem of MP2. Improved perturbation methods (SCS-MP2 and B2PLYP double-hybrid functionals) provide correct results.