The size effect of Muttoni et al.'s critical shear crack theory (CSCT) is shown to be quite close (with differences up to 15%) and asymptotically identical to the energetic size effect law (SEL), which has been extensively verified experimentally and theoretically (and is adopted for the 2019 ACI Code, Standard 318, for both beam shear and punching). However, the CSCT derivation and calculation procedure obfuscates the mechanics of failure. It is shown to rest on six scientifically untenable hypotheses, which would have to be taught to students as an article of faith. They make CSCT untrustworthy outside the testing range; ditto for beams with T, I and box cross section, or for continuous beams. The present conclusions are supported by experimentally calibrated finite element simulations of crack path and width, of stress distributions and localizations during failure, and of strain energy release. The simulations also show the CSCT to be incompatible with the strut-and-tie model, which is (for 2019 ACI Code) modernized to include the size effect in the compression strut. Finally, further deficiencies are pointed out for the Modified Compression Field Theory (MCFT), currently embedded in the Model Code.