It is well known that inappropriate dimensional tolerances can lead to diminished performance of an engine. This paper presents a method to determine the probabilistic dynamic errors of a single cylinder engine. The uncertainties considered in the analysis are tolerance on the link length, radial clearance and random pin center location. Such uncertainties create mechanical errors in the position, velocity and acceleration of the piston and crank and thus, influence the engine's performance. This research statistically determines the effect of tolerances on the global dynamic behavior of an engine. This study uses the effective link length model and adopts the Monte Carlo simulation method to determine mechanical errors in terms of standard deviations. The method developed and presented by this paper can be used to provide the design engineer with the necessary tools to predict the dynamic behavior of a piston-cylinder engine whose parts have dimensional uncertainty as in a manufacturing environment. This work is critical since standards for tolerance allocation for the functional performance of dynamics mechanical systems do not exist.