Off-road and racing motorcycles require a particular setup of the suspension to improve the comfort and the safety of the rider. Further, due to ground unevenness, off-road motorcycle suspensions usually experience extreme and erratic excursions in performing their function. In this regard, the adoption of nonlinear devices, such as progressive springs and hydro pneumatic shock absorbers, can help limiting both the acceleration experienced by the sprung mass and the excursions of the suspensions. For dynamic analysis purposes, this option involves the solution of the nonlinear differential equations that govern the motion of the motorcycle, which is excited by the stochastic road ground profile. In this study a 4 degrees-of-freedom (4-DOF) nonlinear motorcycle model is considered. The model involves suspension elements with asymmetric behaviour. Further, it is assumed that the motorcycle is exposed to loading of a stochastic nature as it moves with a specified speed over a road profile defined by a particular power spectrum. It is shown that a meaningful analysis of the motorcycle response can be conducted by using the technique of statistical linearization. The validity of the proposed approach is established by comparison with results from pertinent Monte Carlo studies. In this context the applicability of auto-regressive (AR) filters for efficient implementation of the Monte Carlo simulation is pointed out. The advantages of these methods for the synthesis of excitation signals from a given power spectrum, are shown by comparison with other methods. It is shown that the statistical linearization method allows the analysis of multi-degree-of-freedom (M-DOF) systems that present strong nonlinearities, exceeding other nonlinear analysis methods in both accuracy and applicability. It is expected that the proposed approaches, can be used for a variety of parameter/ride quality studies and as preliminary design tool by the motorcycle industry.