I disagree with this to an extent. A huge part of meteorology and forecasting is having a tremendous understanding of climatology, patterns, and how pattern evolutions can influence short-term sensible weather.
The best forecasters out there can typically have a solid understanding on how storm potential may evolve several days out or at least some expectation and these forecasters, if working within the private sector, can make some damn good money with this skill. There are people who dedicate their life and career to this sort of thing. The best forecasters can also assess forecast model data and each model and develop a sound idea of which forecast model is probably going to be the best at handling the situation. While not directly used in the field unless you're into heavy research or modeling, even having an understanding of mathematics (physics/calculus) and the governing equations can provide you clues and guidance in this regard.
There are jobs within the private sector which almost require providing some "guarantee" or "most likely" scenario for such storms even as far as 5-6-7 days out. Many companies are understanding the need of weather and how it impacts their business and with this the private sector is growing substantially. Many places are realizing how planning around/ahead weather can save them enormous amounts of money. So let's say in the case of Ernesto, and you're either in the Northeast or someone who does business within the Northeast, and you hear of a possible hit 5-6-7 days out...you're doing to want/have to start planning on that potential this far out and you are going to want to know what this probability is so you can make your preparations.
When it comes to tropical weather and hurricanes, we can easily know the likelihood of a potential system hitting our region as far as 7-10 days out. There is a very specific upper-level pattern that is needed to happen, this is why it is very difficult to get landfalling canes here.