computer models are done by people
Here's another science example of the fallacies inherent in computer models. Remember all those computer models of solar systems pre-Marcy & Butler? No, none zero, hot Jupiters. Models produced variations on a theme of our solar system. Maybe a coupla Jovians here an there, but way far out from the star. Yet, they were done by very astute astrophysicists. With computer models, you tend to model what you expect to get. Preconcieved notions can drive the outcome even without conscious effort by the modeler. Reminds me a lot of guys tweaking their lab results in college to get the result they knew they needed.
-philw (not that moi would do such a thing)