HF - NHC uses 1 minute averaged winds for their maximum wind reports. The JTWC uses 10 minute averaged winds for theirs out in the WPacific; that makes some of those storms even more impressive. Most other agencies use 2, 5, or 10 minute averages.

I think some of the reason why the NHC didn't change the track maps, beyond the education issue noted in another thread, is that none of their proposed alternatives were all that appealing. They simply don't provide the information that is necessary if a change is going to be made to the graphic. GIS technology is getting advanced enough -- think Skeetobite's work -- to the point that the NHC should strongly consider using it in their applications. The NWS is testing it out with radar imagery, with impressive results. Maybe in future years we'll see a movement towards that, but the proposed alternatives weren't very good (at least IMO) this time around.

Regarding damage & wind speeds with Charley -- there is a lot of uncertainty when it comes to estimating intensity with regards to damage observations. Mesoscale (small scale) features are going to result in wide variations in damage over small areas, especially in a storm the size of Charley. Tornado intensities are only measured by the amount of damage they cause; if a strong tornado touched down and didn't hit anything, it may only be rated an F0 or F1 when, in fact, it was much stronger. Similarly, something of F2 intensity that levels everything in its path may be rated an F3 by one observer and an F4 by another.

While the damage estimators are trained in the field, it is really a very subjective process; two people can come up with vastly different estimates and still not be right. Trying to use these estimates when looking at hurricane damage is going to result in the same uncertainty. Thus, I feel it's better to take the "truth" observations from what instruments were available as the closest thing to being accurate, all while keeping the caveat in there that the true intensity of the storm may not have been accurately sampled by the available tools.

With a storm of Charley's size, the mathematics show a very rapid drop off in intensity as you move away from the center of a storm; just barely missing the strongest winds may result in you receiving winds 10-15mph less than the maximum observed. Regarding gusts within a storm -- while the nature of a tropical cyclone tends to supress convective downbursts, once the storm has moved on land, this tendency erodes (due to friction and a lack of moist inflow), allowing for the strongest winds associated with the storm (found about 1km up at the top of the boundary layer) to be transferred to the surface as gusts within downdrafts. Slightly stronger gusts may be found a bit further inland from the coast as this tendency erodes; a 170mph gust is not out of the question for a storm of Charley's intensity a few miles inland.

I don't think anyone's really off base here, just that there's a lot of uncertainty in taking video observations to assess wind speeds with a storm.