Do you ever feel the urge to pick up one of those Farmer’s Almanacs just to see “what’s going to happen this winter”? Apparently a lot of people do! But how accurate is it? Well, it’s not too hard to find opinions, for example: “…IT NAILED THE SNOWSTORM ON JANUARY 5TH LAST YEAR!”. But there are very few studies looking at its accuracy. They claim 80% accuracy…wait, I just choked up laughing…
Jan Null at Golden Gate Weather Services here has checked the almanac’s accuracy several times over the past 15 years, rarely does it do well.
I did do a brief verification of those winter forecasts back around 2,000, but that was lots in a computer change at some point. So I figured it was time to do it again…
Brian Macmillan and I have been working on a presentation based on 4 winters of Old Farmer’s Almanac forecasts. By the way, the almanac looks like this:
It’s not the other one that I referred to back in an August posting that looks like this:
We looked at the past 4 winters, but didn’t analyze any sort of “snow/cold” forecasts. Just how the temp and precip forecasts compared to reality.
Here are the results…got it all?
The precipitation forecast was particularly abysmal last winter. They expected a dry start and a wetter end. Instead the opposite occurred! OFA (Old Farmer’s Almanac) was correct on precipitation anomaly (month-wise) only 50% of the time during the 16 months we analyzed. As you can see the temperature forecasts below were even worse…OFA is wrong far more often than right.
Chief Meteorologist Mark Nelsen