A website I follow,
www.wunderground.com has a good blog if anyone is interested. The website, unfortunately, loads very slow so I would warn against trying to click on the link from a 3G, DSL or anything other than highspeed. The reports for Danielle state
"The formation of Danielle is remarkable in this it was successfully forecast by the GFS model nearly two weeks in advance. The ECMWF and NOGAPS models also did a good job of predicting Danielle's formation a week in advance. The models are getting better and better each year at forecasting genesis of tropical cyclones, though a successful 1-week forecast of genesis is still a rarity. For example, none of the models foresaw the development of 96L until just 3 - 4 days ago"
This was a pretty accurate prediction and they also predicted TD7 a few days ago. If you look at the 5 or 6 computer models, it will give you the forecasts and the statistics for who has been the move accurate in the past. When you watch the weather channel or the NHC site, they only give you the average of the stats, but in actuality, there are more accurate forecasts (like U of W, etc.)
Just some dumb information to read. The site is pretty good. But like I said, it takes a while to load and the guy who runs it is a global warming advocate so you have to weed through the opinions.
As usual, form your own opinion. As of this writing, it looks likes TD7 might head south of Bermuda and hit the coast. Of course, it's many days away.