Prediction is hard, especially about the future, as eminent thinkers from the Danish physicist Niels Bohr to the baseball manager Yogi Berra have observed.
You can spend months scouring newspapers for clues as to who will win the presidential election in the United States, for instance, and then a once-in-a-lifetime mega storm smashes into New York and upends the election race altogether.
It could, therefore, be hardly a more appropriate time for the publication of The Signal and the Noise: Why so Many Predictions Fail - but Some Don't by Nate Silver.
The darling of political soothsayers has taken an in-depth look at how predictions are made, from earthquakes to baseball scores to political punditry and - appropriately enough - hurricane forecasting. Funny how that worked out.
Key to his argument is the sifting of meaningful indicators from the chaff of bogus, but widely accepted, arguments.
For example, newspaper columns making a song and dance about Mitt Romney's latest gaffe might be a less useful indicator of an election result than whether previous polls have been correct.
But there are dangers in overreliance on data when underlying assumptions have not been tested, as customers of the ratings agencies Standard & Poor's and Moody's found during the sub-prime mortgage crunch.
As if to underscore the point, while next week's presidential poll has been inevitably labelled as "too close to call", Hurricane Sandy has thrown enough uncertainty into the race that the outcome of the race now is anyone's guess.
The book is not as concisely written as one would hope, which gives the sense that it was a little rushed out the door ahead of the US election.
Nonetheless, this book is a solid read and an excellent reality check for any would-be makers of "forward-looking statements", or for any politically aware teenage son or daughter disappointed by the outcome of Tuesday's election.
In an age when industrial nations welcome Big Data into the corporate fold and even journalists start to ditch pundits for indexes, the need for solid data analysis is becoming increasingly paramount.
It'll probably sell like hot cakes.
