Don't discount the GFDL solution just yet. Conditions are there for rapid intensification once this thing gets going, despite the direction of motion. The SSTs are very warm -- 30+ C -- in the region, shear is low, and the subtropical jet to the north will only assist in enhancing outflow while in the east Pacific. As the storm organizes and becomes better defined, we could see one round of rapid intensification before landfall, perhaps even to major hurricane status (as the GFDL calls for). I wouldn't go 125kt yet, but 100kt is possible. The people along the entire Central American coast -- Pacific and Caribbean sides -- need to watch this one. Two-three days is the time frame we're looking at, particularly the latter end of that.
As for evolution after landfall: what path the storm takes across land will determine how much survives into the Caribbean. Obviously, the narrower the path, the stronger the remnant circulation. A deep trough should be located off-shore of the U.S. coast as it re-emerges into the Caribbean, but a cut-off low over the Gulf (not the one there now, but down the road 4 days) as a ridge builds across the central Plains may interact with the storm, resulting in a deviation to the track from what we might expect. Best bet now is to go with the "flow" and call for a continued NE progression, similar to that shown in the GFDL. Shear is going to be pretty high no matter where it goes here, though, and the evolution may not end up entirely tropical as a result.
We may well get our first Atlantic May storm in nearly 25 years in the next week. I'm not going to say that it will happen for certain, but the ingredients are there to potentially see something. I'm more confident in something happening here than I was before, and I'll leave it at that for now. Watch this storm in the EPac though -- it's certainly going to affect the coast of Mexico/Cent. America pretty hard in a few days' time.
Current Tropical Model Output Plots
(or view them on the main page for any active Atlantic storms!)