Most milk-applied antioxidant assays in literature are based on the isolation and quantification of individual antioxidative compounds, whereas total antioxidant capacity (TAC) gives a more holistic picture due to cooperative action of antioxidants. Recently, the cupric reducing antioxidant capacity (CUPRAC) method has been modified to measure the antioxidant capacities of thiol-containing proteins, where the classical ammonium acetate buffer – that may otherwise precipitate proteins– was replaced with concentrated urea buffer (able to expose embedded thiol groups of proteins to oxidative attack) adjusted to pH 7.0. Thus, antioxidant capacity of milk was investigated with two competing TAC assays, namely CUPRAC and ABTS (2,2′-azinobis(3-ethylbenzothiazoline-6-sulphonic acid))/persulphate, because only these assays were capable of evaluating protein contribution to the observed TAC value. As milk fat caused turbidity, experiments were carried out with skim milk or defatted milk samples. To determine TAC, modified CUPRAC method was applied to whole milk, separated and redissolved protein fractions, and the remaining liquid phase after necessary operations. Both TAC methods were investigated for their dilution sensitivity and antioxidant power assessment of separate milk fractions such as casein and whey. Proteins like β-lactoglobulin and casein (but not simple thiols) exhibited enhanced CUPRAC reactivity with surfactant (SDS) addition. Addition of milk protein fractions to whole skim milk produced significant ‘negative-biased’ deviations (up to −26% relative standard error) from TAC absorbance additivity in the application of the ABTS method, as opposed to that of the CUPRAC method less affected by chemical deviations from Beer's law thereby producing much smaller deviations from additivity (i.e. the property of additivity is valid when the measured TAC of a mixture is equal to the sum of individual antioxidant capacities of its constituents).