Evaluating the reliability of slopes against sliding failure is complicated by the fact that most slope soils are spatially variable. This means that instead of nice circular failure surfaces, slope failures tend to be more complex, following the weakest path or zones through the material. The nite element method is well suited to slope stability calculations since it allows the failure surface to seek out the weakest path through the soil. This paper weds the nite element method with random eld simulations to perform a `random nite element method' (RFEM) analysis of a slope. The simulation results are used to validate a simplied analytical reliability model for slope stability which is based on harmonic averaging. The harmonic average is low­ trength dominated, so that it captures the weakest path characteristics of slope failure and provides good estimates of the failure probability.