I see way too much predicting how a model will behave rather than interpreting the current weather and then forecasting the potential going forward. When I took meteorology courses in college 35+ years ago we looked at what was happening in key areas of the hemisphere and drew maps for however many days in the future using a variety of methods to help determine what could happen. We obviously didn't have the computer power available now so we used science, math and historic references and made a few maps with different outcomes then tried to home in on the most likely result. Sometimes now I get the feeling that forecasting is done solely off what the model thinks it sees rather than what's actually happening.