What do we want? More uncertainty! When do we want it? Um, not sure …

snowmanIn case you hadn’t noticed, we recently had a lot of snow across much of the UK. The weather forecast stated clearly that heavy snow would fall late on Friday. Then the forecast changed. The snow would begin at 9am precisely on Friday morning. In the event, the snow started to fall on Thursday and so there was already a white blanket covering much of England and Wales when I woke on Friday to hear the weather forecaster state without a hint of irony that “it’s snowing now.”

With each new forecast, a greater degree of accuracy was achieved. However, the precision never changed. There was never a shred of doubt in each confident prediction of arctic conditions and traffic chaos at some precise time.

If a forecast changes, has it become more accurate, or was it simply wrong in the first place?

Those guys at the Met Office are pretty smart. They probably knew all along that there were big uncertainties in the timing of the snowfall. But some focus group must have told them that the public can’t handle uncertainty in weather forecasts. Predictions like “60% chance of rainfall today” or “snow likely sometime this week” are too hard for the poor guy in the street to comprehend. Instead, the public demands utterly precise forecasts of “snow beginning at 1.15 on Tuesday followed by sleet until 9.30pm.”

Except that precision forecasts are usually wrong. They turn out not to be accurate. If I told you that it was going to start snowing at 3.17 and 22 seconds but it didn’t start until half past four, you wouldn’t be too impressed.

Not only is it unscientific to confuse notions of precision and accuracy, but it’s condescending to assume that the public won’t notice or just can’t understand. Sports commentators are always spouting statistics, odds and probabilities, and they are clearly understood. People understand uncertainty, so why not tell them the uncertain truth?

The daily weather forecast is one of the very few areas where scientific predictions come under direct public scrutiny. At a time when public confidence in science is more important than it’s ever been, please don’t mess it up.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.