Why the "classical" explanation of weak values by Ferrie and Combes does not work: a comment on Phys. Rev. Lett. 113, 120404 (2014)

In Phys. Rev. Lett. 113, 120404 (2014) [arXiv:1403.2362], Ferrie and Combes claim that weak values could be a statistical artifact caused by correlations between the disturbance and the post-selection condition imposed on the output. In this comment, we show that the disturbance caused by a weak measurement is sufficiently low to definitely rule out the model proposed by Ferrie and Combes.