February 08-09, 2013 Blizzard Storm Total Snow Accumulation/ by Brandon Vincent (National Weather Service Raleigh, NC) [Public domain], via Wikimedia CommonsI’ve blogged a time or two about New England’s famous “farmer’s almanacs” (there are two: the Farmers’ Almanac and the Old Farmer’s Almanac) and their nonsense “predictions” of what the winter will be. The Farmers’ Almanac has just released its prediction, and as the AP reports via CBS News, it’s rather dire (locally-cached article):

The Farmers’ Almanac is using words like “piercing cold,” “bitterly cold” and “biting cold” to describe the upcoming winter. And if its predictions are right, the first outdoor Super Bowl in years will be a messy “Storm Bowl.”

The 197-year-old publication that hits newsstands Monday predicts a winter storm will hit the Northeast around the time the Super Bowl is played at MetLife Stadium in the Meadowlands in New Jersey. It also predicts a colder-than-normal winter for two-thirds of the country and heavy snowfall in the Midwest, Great Lakes and New England.

“We’re using a very strong four-letter word to describe this winter, which is C-O-L-D. It’s going to be very cold,” said Sandi Duncan, managing editor.

The AP quite thoughtfully explains how we’re to be sure this “prediction” is accurate:

Based on planetary positions, sunspots and lunar cycles, the almanac’s secret formula is largely unchanged since founder David Young published the first almanac in 1818.

Modern scientists don’t put much stock in sunspots or tidal action, but the almanac says its forecasts used by readers to plan weddings and plant gardens are correct about 80 percent of the time.

Gee, I just love how the AP tells us that the Farmers’ Almanac must be accurate because its publisher says it’s accurate — and that, in turn, they back up by asserting that people use it to plan weddings. I am just so fucking glad they could clear that up for me!

I also love how they totally dismiss modern meteorology, as though no one has learned any more about the weather since 1818. I’m well aware of meteorology’s deficiencies … I was born and currently live in rural Connecticut, after all! … but to assume a putative prediction method dreamed up back in 1818 can’t possibly have been improved upon over the last two centuries? Come on. What a fucking joke!

The AP devoted just one clause of one sentence to mentioning that the two almanacs have skeptics, but nearly a couple of years ago, Discovery News ran a story addressing the almanacs’ accuracy, and they revealed there’s actually science behind that skepticism (cached):

Many atmospheric scientists and meteorologists scoff at the ability of olde-tyme formulas used by the digests to prognosticate the weather.…

For nigh on to two centuries, Americans have taken a gander at farmer’s almanacs for auguries about the weather. Millions of readers think they are the bee’s knees but atmospheric scientists scoff at the ability of olde-tyme formulas to prognosticate the weather.

“Based on my own analysis, and that of others, the monthly mean forecasts published by the ‘Old Farmer’s Almanac’ (OFA) lack value,” Nick Bond, of the National Oceanic and Atmospheric Administration and Washington’s State Climatologist, told Discovery News.

The “Farmers’ Almanac” and the “Old Farmer’s Almanac” are in competition with each other, but also face stiff competition from meteorologists with millions of dollars worth of satellites, radar dishes and other new-fangled contraptions.

The “Farmers’ Almanac” has weathered these scientific advances with stalwart faith in the founder’s formula.…

In 2003, Bond compared “Old Farmer’s Almanac” forecasts to actual weather events in the Pacific Northwest, the results are summarized in the Washington’s State Climatologist’s newsletter.

“The forecasts are sometimes correct. In terms of getting the sense of the weather anomalies right, for example whether it will be colder or warmer than normal, the OFA is correct about 50 percent of the time,” said Bond.

“Of course this is no better than flipping a coin,” he added.

Jan Null of Golden Gate Weather Services has compared “Old Farmer’s Almanac” forecasts to actual weather conditions across the United States for much of the 2000’s. His results corroborate those of Bond.

Back in 1981 another study, published in Weatherwise, looked at 60 monthly temperature and precipitation forecasts for 32 weather stations across the U.S. and compared them to “Old Farmer’s Almanac” forecasts. Once again, the accuracy of the “Old Farmer’s Almanac” was found to be no better than flipping a coin.

The almanacs’ claimed successes, it turns out, aren’t even true “successes”:

For example, the “Farmers’ Almanac” website notes they were “on the money,” when they forecast a hurricane threat for the Southeastern U.S. at the end of August, manifested in the form of Hurricane Irene.

“A forecast of a hurricane hitting the southeastern United States in August is probably a pretty good bet in any year,” said [U. of Missouri scientist Neil] Fox. “You tend to hear about these ‘remarkable’ predictions, but not, of course all the times they get it wrong! I certainly would not make my plans based on this.”

The bottom line is that the mass media have no viable excuse for reporting any of this bilge as though it were really “news.” They have even less excuse for reporting the almanacs’ claims about their own accuracy as though they were fact, when clearly they are merely raw assertions with no demonstrable basis. There’s no place for this kind of hypercredulity, at the AP or in any other newsroom. No one is served this insipid trash.

Photo credit: Wikimedia Commons.

Tags: , , , , , , , , , , , , , , ,

  • Shannon

    Lol..Don't know if I trust the almanac completely but they seem to have gotten it fit this time on the bitterly cold temperature this year

    • Just because they were right once, doesn't mean there's anything to their predictions. As the saying goes, even a broken clock is right, twice a day.

      What makes their predictions valid, is if they consistently predict the weather correctly, repeatedly, and over the long term are correct much more often than not. Another thing that would validate their predictions, is if other predictors could apply whatever logic or method they use to generate predictions independently of them, and arrive at their own, validated, predictions.

      The fact is that, over the long haul, the Almanacs aren't more correct more often than they're wrong. And because they keep their predictive method secret, there's no way for others to parallel it.

  • You gotta' believe the Almanac, man! They're probably going to be right on the money regarding that SuperBowl prediction. And, of course, if you read the Almanac on the Internet it has to be doubly true… because you're reading it on the Internet. 😉

  • Steve Traffic

    Funny, The East Coast was bombarded by freezing temps and tons of snow. Sounds about right to me.

    • No storm on the Super Bowl itself, which is what the Almanac predicted. So, failed prediction. Not here, and not with me. If they'd preferred to have made a more general prediction, such as "it will be cold and snowy in the northeast around the time of the Super Bowl," then they might have prevailed … but that particular prediction wouldn't have meant much. It will be cold and snowy in the northeast during the first week of February? One doesn't need some nutty pseudo-algorithm to make that "prediction." One needs only to be familiar with winters, in order to do so.