Gob gas ventholes (GGVs) are an integral part of longwall coal mining operations, enhancing safety by controlling methane in underground workings. As in many disciplines in earth sciences, uncertainties due to the heterogeneity of geologic formations exist. These uncertainties, and the wide range of mining and venthole operation parameters, lead to performance variability in GGVs. Random variations in parameters affecting GGV performance and influencing parameters that cannot be quantified sufficiently due to lack of information limit deterministic GGV models and even introduce error in severe cases. Therefore, evaluation of GGV performance data and the uncertainty in input parameters is valuable for understanding the variability in GGV production and for designing them accordingly. This paper describes a practical approach for implementing stochastic determination of GGV production performances and for generalizing the prediction capability of deterministic models. Deterministic site-specific models were derived by using the GGV module in the recently developed MCP (Methane Control and Prediction) software suite. These models were generated using multi-parameter regression techniques and were then improved by inclusion of extra input parameters that eliminated the site dependency and improved the predictions. Statistical distributions of input parameters in these models were quantified and tested with the Kolmogorov-Smirnov goodness-of-fit technique. Next, Monte Carlo simulations were performed using these distributions and generalized results for GGV performances were generated. The results of this work indicate that this approach is a promising method of representing the variability in GGV performances and to improve the limited and site-specific character of the deterministic models. Published by Elsevier B.V.