A comment on the paper "How can a Result of a Single Coin Toss Turn Out to be 100 Heads" by C. Ferrie and J. Combes

The authors of a recent paper [Phys. Rev. Lett. 113, 120404 (2014)] suggest that "weak values are not inherently quantum but rather a purely statistical feature of pre- and postselection with disturbance". We argue that this claim is erroneous, since such values require averaging with distributions which change sign. This type of averaging arises naturally in quantum mechanics, but may not occur in classical statistics.