TV weather forecasts
Freakonomics has an interesting study of TV weather forecasts in the Kansas City metro area. On average, the forecasters were 85% accurate in forecasting tomorrow’s precipitation, declining to 73% for 7-day forecasts. The problem with this that it only rains in Kansas City 14% of the time, so if a forecaster simply always predicted that it would not rain, he’d be accurate 86% of the time. And that’s not just for tomorrow’s weather! He could always predict that it would not rain 7 days from now, or 30 days from now, and he’d still be right 86% of the time.
The article goes on to discuss hiring practices at the TV stations:
When directly asked if accuracy mattered in forecasting, every station manager and meteorologist said it did. But when asked what steps they had taken to measure and ensure accuracy, they were without answers.
No meteorologist or television station kept records of what they predicted, nor compared their predictions to actual results over a long term. No meteorologist posts their accuracy statistics on their résumé. No station managers use accuracy statistics in the hiring or evaluation of their meteorologists.