what would it take to get historical data? I think that is probably the most interesting aspect of the map. forecasts tend to be only as good as the model's best guess.
if historical data were implemented, it would also be really cool to access some of that data by code. essentially, what I'm saying is perhaps eventually the model could be cloudsourced the researchers interested in that. I'm not totally sure how it could work, but it would be cool to research that and allow a community to grow around the forecasting models.