Please, please stop sharing spaghetti plots of hurricane models

Enlarge / Hurricane Irma creates enough uncertainty on its own. Don't add to it by sharing spaghetti plots. (credit: NOAA)

Nate Silver, the statistician and editor of FiveThirtyEight, is unquestionably a smart guy with an exceptionally good grasp of numbers and probability. But he's not a meteorologist. So when he shared a spaghetti plot of models for Hurricane Irma on Tuesday night, I had to shake my head. Just another social mediarologist.

Still some uncertainty in the Hurricane Irma forecast but more and more projections are converging on Florida. https://t.co/Tk8G06vz0g pic.twitter.com/ADFYTHTsdG

- Nate Silver (@NateSilver538) September 6, 2017

Please don't misunderstand me. I am not some kind of meteorologist elitist. Honestly, it doesn't take that much education to have a basic understanding of weather forecasting. And I don't have any problem with people looking at all of the available forecast data to draw their own conclusions about hurricane forecasts and make decisions for their families in the face of a natural disaster.

But spaghetti plots are not good decision-making tools. Sorry, they're just not. To understand why, let's take a look at the models on Nate Silver's plot, which he shared with his 2.5 million followers at 7:34pm ET Tuesday:

Read 3 remaining paragraphs | Comments