Along several decades of development, many methods based on reverberation chambers (RCs) have been proposed for accurate and efficient measurement of antenna efficiency. The standard reference antenna method (RAM) is applicable for almost all frequency bands of interest and all types of antennas, and thus, it may well be the most widely applied method in the industry. In this article, an average Rician <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> -factor-based uncertainty model for the standard RAM is proposed. It takes both the number of independent samples and the nonideality of the RC into consideration. Extensive measurements are performed for different RC configurations. It is shown that the newly proposed uncertainty model provides good and stable estimations of the measurement uncertainties for both unloaded and loaded RCs. Moreover, it is experimentally verified that the differences in the radiation characteristics between the reference antenna and the antenna under test (AUT) have little effect on measurement uncertainty regardless of the loading conditions of the RC. The findings of this work allow more flexible and efficient antenna efficiency measurements, including estimation of the associated uncertainty.