Community Research Partners is now Thoughtwell.



It will snow zero to 24 inches: A brief synopsis of meteorology statistics


In early January weather forecasters started predicting that Central Ohio would receive somewhere between zero and 24 inches of snow over the coming weekend. Naturally, the jokes started pouring out immediately. “Finally, an accurate forecast!” “If I was this bad at my job I would be fired!”

As fun as it is to mock meteorologists,[1] it’s also important to realize that forecasting is an art, not a science, and that statistics do not tell you what WILL happen – they merely tell you what MIGHT happen. I’m sure everyone remembers the epic failure of predictions the night of the 2016 Presidential Election, when most news sources assumed that Clinton would be our next President and were roundly shocked into silence when Trump instead delivered a clear victory.

But why do models and statistics so often fool us? I think it’s because statistics sound more convincing than they are. Whether there is a 10% chance of rain or a 90% chance of rain, it either will rain or will not rain. Weather forecasting, like election forecasting, is based on models that are only as good as their inputs. Even if the meteorologists can effectively predict we will get snow, knowing how much will fall to a precise degree is nearly impossible, particularly well in advance of the event.

So the next time you hear a potential doomsday forecast, fear not – it may not snow at all. Or, it might be the biggest snowfall to hit Central Ohio in years. I predict a 50% chance of either outcome.


[1] I used to want to be a meteorologist until I discovered they have to know Physics. Instead of becoming a meteorologist I just followed them around Boston when they came to report on blizzards (see picture of me and Jim Cantore).