I think we're not seeing something here, every model might be right and every model might be wrong.
Best to stick to the 120 hour rule for the next few weeks.
There are alot of systems in the pipe, and any run can focus on one or the other and change the outcome.
You can get fuji wars for one, system splits second, and then changes a third system down stream, 4th something moves slower or faster then anticipated.
At the betting stage now I would say the broad low splits, and a piece follows Erin, pitters and gets going at the end as it recurves and a piece moves through the carribean and consolidates in the northwest carribean or gulf, followed by the Eastern system and possibly what looks like another nice wave 4th system.
The south east likes to get hit when there are lots of systems out there.
The one two three, one ahead the south east us storm in the middle and one behind.
Forgetting surface maps one can see on upper level of the gfs the anti cyclone involved with the broad low tracks with or without a surface low through the islands of the carribean , and ends up in the Gulf sparking a hurricane there as modeld now.
The second anti cyclone associated with the second east system now track up behind and north of the islands with what might be a substantial high pressure area north of it, and it takes longer then forcast to get there.
And what about the system over africa now if that is strong enough it could help to make the second system hold south. With the models wanting to put that big high pressure north.
The looks of the 28th -3rd sept I would say there should be a storm some where near the south east us and maybe one in the Gulf with one or two out in the atlantic...just my guess