50-50 chance with almanac forecasts
I’ve been asked a number of times of my opinion of the Old Farmer’s Almanac and its accuracy. With a new edition on the horizon, many have wanted to know if this publication is as “on the money” as some claim. The Old Farmer’s Almanac still uses mathematical formulas that were developed in the 1800s to forecast the weather. The publication was first developed in 1772 and continues to provide weather outlooks up to several years in advance.
With an increasing need for more-accurate long-range weather forecasts, many people tend to rely on rather questionable information. In my opinion, the Almanac’s forecasting methods are neither conventional nor well-documented.
The annual publication carries this statement: “Our forecasts are determined by both the use of a secret weather forecasting formula derived by the founder, Robert B. Thomas, in 1792 and by the most modern scientific calculations based on solar activity, sunspots, etc.”
According to a five-year study conducted in the early 1980s by two research meteorologists, John Walsh and David Allen of the University of Illinois at Urbana, the Almanac’s accuracy differed little from a flip-of-a-coin, approximately 51 percent accuracy for temperature and 52 percent correct for precipitation.
Walsh and Allen used 32 cities from the 16 various forecast regions of the country used by the Old Farmer’s Almanac. Spokane and Miles City, Mont., made the list.
The study showed that the National Weather Service averaged about a 60 percent accuracy in “seasonal” outlooks published in advance. The private long-range prognosticators, including our long-range weather service, received the highest accuracy ratings at nearly 70 percent.
However, I had many relatives who were farmers and swore to the Almanac’s accuracy. One could only imagine the conversations that went on, but it was all in good fun. I also have to give the Almanac credit for predicting our current colder global temperature cycle.
The big heat wave in early August was likely the hottest period of 2009. We still have some very warm to hot days, but not as intense. I’m still expecting things to turn much wetter and cooler after Labor Day, probably around September 13-19.