Well, to be honest they aren't far off. It's about 40 miles further east right now then where predicted at the 11pm forecast last night. Less then 100 miles if you go back 24 before then to the 11pm Tuesday.

And the projected landfall is still well within the errorcone from the last several days.

This hasn't been a very easy storm to predict and they aren't doing bad.

Also, you need to know something about the predicted tracks. They will not and do not jump them long distances. They ease them out there. You could have everyone in NHC in agreement that the hurricane will land near NOLA, but if the previous track had it landing near Houston, they will not put the projected path under NOLA. Instead they'll slowly walk it towards NOLA eventually getting there in a few forecasts.
They often even tell you this flat out in their discussions. And knowing how they use the computer models to aide in their forecasts you can usually predict the direction they'll walk it to and even where it will end up if the models agree well.

They do this because these things are not easily predictable. And they don't want landfall jumping all over the map. Could you imagine NOLA at 11AM, Houston at 5pm, Mexico border at 5am and back to NOLA at 11AM? Not that any storm I can think of would have been like that, but that is worse case scenario if they don't slowly walk the landfall predictions. Of course the speed at which they walk them is directly related to time till landfall.

But either way, it isn't politics. It's not overrating to current data and throwing out everything that was previously predicted leading to large jumps back and forth in landfall projections.