Statistical models are routinely used in Before-After monitoring studies to detect, quantify, and forecast environmental change caused by natural or anthropogenic disturbances. For monitoring programs that have not established standard statistical procedures, an evaluation of models’ abilities to measure change over a range of scenarios is vital to develop best practices for analyzing monitoring data. A comprehensive evaluation was developed and applied using Marine Renewable Energy (MRE) tidal turbine site as a case study of a developing industry with no standard monitoring methods. Before-After monitoring datasets that contained change were simulated using normal and non-normal empirical baseline data collected at a MRE site. Thirteen regression models from three classes were evaluated: 6 generalized regression models, 4 time series models, and 3 nonparametric models. Intervention Analysis was used to fit models to five change scenarios that included three amplitudes of change and a lag in the onset of change. A power analysis was used to evaluate model ability to detect change. Residual error diagnostics were used to quantify model fit and forecast accuracy. Parametric models that did not include lagged dependent variables were the most capable at detecting change. A comparison of the fit and forecast metric results indicated that deterministic time series models in conjunction with semi-parametric generalized regression models provided a robust and informative quantification of change. Nonparametric models most accurately forecasted change. The results provide insight on model behavior, which is used to recommend specific models to measure change in the case study data. These recommendations form best practices for analyzing monitoring data, which enables comparisons among monitoring sites and reduces uncertainty when quantifying environmental effects.