Use and Abuse of Statistics

This is one of my pet peeves -- see my blogs on Jobs and Oil Spill Calculations.
Today the Weather Channel announced that certain areas are overdue for a hurricane!  While they were smart enough not to come and say that these areas should therefore expect a hurricane during the current season, they certainly left this impression.
Hurricanes, tornadoes and other severe weather events (as well as millions of other seemingly random events in our lives) are, by and large, independent events.  I.e. the occurrence of one event has no effect on the occurrence of another.  On the other hand, events like habitual smoking and lung cancer, are highly correlated. 
Certainly, some areas (historically) experience climatic conditions that are favorable to (say) tornadoes, while others, by virtue of their geographic position, are more prone to (say) hurricanes than others.  However, the fact that city A was hit by a tornado this year has no bearing on the likelihood of its being hit by a tornado next year.
I must also take issue with the practice of assigning probabilities to single events.  While it is OK to say that "on average, two out of three days in April are rainy", it is not proper to say "the chance of rain today is 67 percent".  Mathematically, it is not possible to calculate the probability for a single event, whether it be heads or tails or the number on a die, only for the outcome in a long run of trials.  After all, if you did assign a probability to a single event, how would you know if you were right or wrong?