Big data's big misses: 2016 was a bad year for predictions

But the bigger 2016 data lesson, where journalism is concerned, may be to stop worrying so much about predicting the future. Journalism is supposed to report on what is actually happening not offer odds on what may happen.

Data are still important when we use them correctly. Polls help us better grasp where the public stands on issues and personalities. Economic numbers and consumer surveys help explain what may be the core story for American politics in 2017: The divides coursing through the United States.

Using numbers that way could make 2017 a very good year for data, especially in political coverage.

What about the rest of data’s bad 2016? Is there a larger lesson in the numbers? Not necessarily. The stories behind 2016’s data-defiers are all different. The Cavs had Lebron. The Cubs had pitching. Leicester City had … whatever Leicester City had.

Then again, it could be that 2016 wasn’t an outlier. It was just the beginning of a new data trend, one where down is up losers become winners and long shots have the advantage.