The card tells you how to play based on your hand and the dealers card. If I'm dealt a 14 and the dealer has a low card, I know I should stay because the majority of the time, the dealer will bust. Using math and probability, you have a better chance at beating the house.
You can use the same principles while forecasting for tropical storms! Climatologically, looking at past tropical storms (1886-2010) the probability of Irene passing over CT based on her current position is 9%.
Don't believe me? The map below shows the probability of a tropical storm passing over Connecticut based on its position. Notice if Irene passes close to the Outer Banks of North Carolina (as forecast), the probability will increase to 10-15%.
Even though the chance of a Connecticut landfall is low, only 9%, it's still a possibility! The computer models are certainly leaning towards some serious impacts here in the state. This is just like the game of blackjack. Sometimes you play the hand exactly right and the house still wins. This storm could beat the odds. So what bet will you place?