Choosing the best depth conversion model for an oil or gas reservoir can be difficult. One of the best ways to choose between two competing models is to use cross-validation of wells. This means that both models are estimated from all observation except one, and then the prediction error for the observation left out is measured. By doing this successively for all observations, a root-mean-square-error (RMSE) can be estimated for both depth-conversion models. The model with the lowest RMSE is the better. The RMSE does not suffer from over-fitting.