In cold fusion experiments conducted at the Naval Research Laboratory in China Lake, M. H. Miles and co-workers claim to have produced excess heat correlated with helium-4 production, X-rays, and Geiger-counter excitation. However, scrutiny of the claims shows that unreliable calorimetric and nuclear-product detection methods were used. Moreover, inconsistencies and errors are found in the data and data analysis. The juxtaposition of several poor techniques and inconsistent data does not make a compelling case for cold fusion. We conclude that the evidence for cold fusion from these efforts is far from compelling.