If by analogs you mean tracks, Wikipedia is the site to go to. They have track maps for every recorded tropical system from the 1851 hurricane season onward. I use it a lot to compare tracks.
Looking at the latest NHC analysis, #3 looks to be Irence and #2 looks to turn into Jose. Assuming Jose forms in the next couple days (which is a bit aggressive), it is worth noting that the "J" storm from last year formed on September 12th.
Yeah, that's what I meant, I'll check there and see then. 6z GFS shifted east also just about where it was on the runs before the 18z run making landfall in the E. GOM in the eastern FL panhandle, moving north through GA then NE. It's also been shifting south of Hispanola avoid a direct hit, but a hit nonetheless so it would still be terrible either way because they'll be in the right front quadrant of the storm. The EURO and Canadian shifted much further east focusing on a SE threat. (EURO ensembles still suggest GOM threat, so were not done with more flip flopping yet...
)
I agree with you about who becomes "Irene" and "Jose". Invest 98L looks to be a fish storm and Invest 97L looks like it's developing convection on the northside, but not around the center yet obviously.
Also has anyone ever heard of the GFS model having a history of underdoing pressures on hurricanes? I've heard that on other forums. One member in particular said that when Katrina was still an invest, the GFS only modeled it to drop into the 960's, even all the way up into short range when it was rapidly strengthening in the GOM. He also said it did the same with Rita that year. If that ends up being the case with this were gonna be in serious trouble if it makes landfall in the US (a recurve out to sea is still a small possbility btw)
EDIT: Tracks similar to Hurricane David and Frederic (1979) seems plausible right now. I should also mention I only looked at storms that originated from the Cape Verdes and these two seem to be the only ones close to what the models are showing at the moment.