The real strengths of models is not to predict, but to explain now that's a bold claim... if i'm not mistaken general relativity ultimate test was actually to

The real strengths of models is not to predict, but to explain now that's a bold claim... if i'm not mistaken general relativity ultimate test was actually to predict trajectories we hadn't observed yet. i can model any behavior happening in the past, with a sufficient number of parameters in my equation. that doesn't say much on the quality of the model.