I am pretty sure there will be complaints about the accuracy of the forecasts for Irma. However, as a certified geezer, I can assure anyone that what we have now is far better than what was available fifty or even forty years ago (maybe even ten years ago?.

If you want to look at really mistaken forecasting, or almost no forecasting at all, look at what happened to Galveston in 1900 - the death toll is undetermined, but was somewhere around 12,000, perhaps as many as 20,000. The local forecaster had argued against construction of a seawall, since a big hurricane could not occur in the city.

The forecasts I listened to (weather underground) were pretty clear about the probability cone and the uncertainties involved.