The Winter-Kennedy (WK) method is a popular choice to estimate the relative flow rates, and thus the expected improvement in the efficiency of a low head turbine after its refurbishment. Runner refurbishment is a common way to improve the plant's efficiency. However, a previous experiment on a model turbine reported deviations between the WK coefficients obtained from two different runners - suggesting a deviation between the estimated and actual improvement in the efficiency. Without formal proof, the deviation was attributed to flow changes in the spiral casing. This paper presents a numerical investigation of the effects of a runner change on the WK method. For this purpose, unsteady Reynolds-averaged Navier-Stokes equations (URANS) simulations of a turbine model with two different runners were conducted. The runner's impact on the average flow conditions upstream and its subsequent effect on the WK coefficients were studied. The study shows the dependence of the WK coefficients to the runner - with a maximum deviation on the coefficient up to 0.7%. The larger deviations were observed in regions prone to strong secondary flow. Following a radial and circumferential sensitivity study, a suitable location to minimize the effects of runner change on the WK method is reported. (C) 2020 Elsevier Ltd. All rights reserved.