Models are in a bit of chaos with the storm effectively moving faster and further south. Generally the idea for the longer term is if it gets picked up by the weakness or not. It's marginal now, which is why models have been going left and right during the runs. At the moment it's more to the right, but with Fiona already further west and slightly south than the runs, it could shift back the other way again. And may flip flop for a while until we know what happens with Hispaniola.

Currently the 6z GFS turns right in the Dominican republic and barely misses the Turks and caicos to the east then cuts right and gets close to bermuda. 18z yesterday had it scraping the Outer Banks. 0z had it moving faster. Telling is the 6z position is already slightly off for where it is now. So the chances of it flip flopping are good.

0z Euro gets into the Central bahamas and then north, but then cuts back left, moving tremendously slowly at the time, showcasing the uncertainty. 12z yesterday had a West Palm beach Florida landfall. So this too is flip flopping

0z Canadian goes through the southeastern Bahamas, then cuts up close to the Outer Banks before turning out to sea. 12z yesterday had a Florida then New Orelans Landfall. So this too is flip flopping. The NHC track still looks pretty solid, however.

The TVCN consensus has been trending west, with some minor shifting right and left during that time.

Ensembles are spread, with the general trend of stronger runs tend to turn north and weaker are more west. The 6z EPS run this morning shows that well.

There's too much chaos in model trends between runs beyond the Caribbean right now to take anything as solid. So in short Bahamas, the US East coast and Bermuda need to keep watch on it until things clear up. With Fiona moving this quickly and weak, it's in a marginal setup to recurve or not right now.