A recent piece by Dennis Mersereau in Forbes described how President Trump's proposed budget (likely "dead on arrival" in Congress) would result in mass layoffs of National Weather Service (NWS) meteorologists. On the Forbes Facebook page, I saw cliche and misinformed comments like "Good, meteorologists are always wrong anyhow." Such statements are rooted in misperceptions based on experiences, lack of math-statistical literacy, and knee-jerk reactions. The reality is that weather forecasts are quite good when consumed with proper perspective. I pose the following question: If the weather forecast called for 70 degrees F in two days and it ended up being 68 degrees F, will people say the forecast was wrong? The answer is probably like many relationships. It's complicated. Meteorologists seem to be held to a different standard than other professions that predict the future.
In response to the question posed, some people will say it was right and others will say it is wrong. How do meteorologists compare to other professions that try to predict the future like investors, economists, sports analysts, doctors, and political pundits? If an investor could pick the best performing stocks 80 to 90 percent of the time, would you likely give her your business? What if your doctor said that there was a 90% chance that your symptoms will worsen unless you take a certain medication, are you likely to fill the prescription? There is probably someone reading this and overanalyzing the questions. However, most people probably said "yes" to both questions. The National Oceanic and Atmospheric Administration (NOAA) Scijinks website is a good place to start:
A seven-day forecast can accurately predict the weather about 80 percent of the time and a five-day forecast can accurately predict the weather approximately 90 percent of the time. However, a 10-day—or longer—forecast is only right about half the time.
These percentages are even higher within 2 to 3 days. Jason Samenow and Angela Fritz wrote in the Washington Post Capital Weather Gang:
A one-day temperature forecast is now typically accurate within about two to 2.5 degrees, according to National Weather Service data. In other words, when you see a forecast high of 82, most of the time the actual high will be between 80 and 85.
At this point I want to recall a story. My son and I were tubing down a river in north Georgia. It started to rain, and a woman started complaining that meteorologists were wrong because there was "only" a 20% chance of rain. I thought to myself, "Well, it wasn't a 0% chance of rain so why are you complaining?" As a meteorologist and atmospheric sciences professor for over 25 years, I have come to learn that people's perceptions (and often misunderstandings) shape how they view weather forecasts. She thought the forecast was wrong though if you look at the definition of percent chance of rain on a NWS website, it clearly states:
The "Probability of Precipitation" (PoP) describes the chance of precipitation occurring at any point you select in the area....Mathematically, PoP is defined as follows: PoP = C x A where "C" = the confidence that precipitation will occur somewhere in the forecast area, and where "A" = the percent of the area that will receive measureable precipitation, if it occurs at all.
I can't tell you how often someone asks me if it is going to rain at 7 pm near the white Gazebo in the park 3 days from now during Prom pictures. Unrealistic expectations about aspects of forecasting shape misinformed judgments about my profession. During Hurricane Irma (2017), I noticed people evacuating from one part of the "cone of uncertainty" in southeast Florida to another part of the cone, southwest Florida. People often misinterpret what the cone is showing them.
Nate Silver's excellent essay, "The Weatherman Is Not a Moron," was published a few years ago in the New York Times. He clearly laid out that weather forecasting is an area that has seen tremendous strides in recent decades. He writes,
Still, most people take their forecasts for granted. Like a baseball umpire, a weather forecaster rarely gets credit for getting the call right....Six years earlier, the National Weather Service also made a nearly perfect forecast of Hurricane Katrina, anticipating its exact landfall almost 60 hours in advance.
His point is that people tend to only remember the relatively few negative outcomes and not the more numerous positive outcomes. For example, the "March Madness" season of college basketball is wrapping up. A player could make 97% of his free throws during the season, but people will judge him on the 1 miss that costed them the game in the Sweet 16.
Ok, let's get back to the comparison of meteorologists to other professions that predict the future. The website The Mathematical Investor posed the question, "How accurate are market forecasters?" Based on analysis of 68 market forecasters, they concluded:
The top-ranking forecaster was 78.7% accurate by our metric. The next three had 72.5%, 71.8% and 70.5% accuracy scores. A total of 11 of the 68 had accuracy scores exceeding 60%. At the other end of our ranking, two had accuracy scores near 17%; three others had scores 25% or lower. A total of 18 had accuracy scores less than 40%.
A study out of Hamilton College analyzed the accuracy of political pundits. In their analysis of 26 political experts, they considered over 472 predictions made over a 16-month period on Sunday talk shows. The results, summarized in a press release, confirmed that
only nine of the prognosticators they studied could predict more accurately than a coin flip. Two were significantly less accurate, and the remaining 14 were not statistically any better or worse than a coin flip.
Meteorologists are able to predict, with up to 90% or more accuracy within 2 to 5 days, how a complex fluid on a rotating planet with oceans, mountains, and varying heat distributions changes. Kudos colleagues.
