Very insightful comment. I have long been interested in what the actual approach and factors are to generating better forecasts and you explained the important parts. Thanks!
[1] "The quiet revolution of numerical weather prediction", http://www.nature.com/nature/journal/v525/n7567/full/nature1...