Al Gore is lucky he isn’t a betting man.
In 2007, Professor Scott Armstrong at the University of Pennsylvania's Wharton business school challenged Al Gore to a $10,000 bet about temperatures over the next decade. Fox News reported on the challenge at the time.
The bet proposal was to compare the U.N.’s standard global warming model against Armstrong’s prediction of no increase at all. The money would have gone to charity.
Gore declined the bet. According to Armstrong, a Gore spokesman said that, “Mr. Gore simply does not wish to participate in a financial wager.”
Now, 10 years after the offer, Armstrong is declaring victory, albeit a moral one.
From the would-be bet period of 2008 through the end of 2017, Armstrong’s prediction of zero temperature change was more accurate in more months than the standard U.N. model, which predicts an increase in temperatures.
But the bet result comes with the caveat that, in the last two years, warming has been high. In those years, the U.N. model’s prediction was most accurate. But overall across the whole decade, Armstrong's "no change" model edged out the U.N. model that Gore relies on.
The graph below from Armstrong’s TheClimateBet.com shows the bet measurements. Black is temperature, red is the UN’s forecasted increase, and green is Armstrong’s no-change forecast.
Armstrong says the bet undercuts extreme predictions Gore made. In his 2007 book "Assault on Reason," Gore warned of “‘tipping points’ that could - within as little as ten years - make it impossible for us to avoid irretrievable damage of the planet's habitability for human civilization.”
Ten years out, Armstrong said that has not happened.
But the fact that the U.N. model’s prediction is closer regarding today’s temperature has prompted some to question Armstrong’s methodology.
“Anyone objectively looking at the data, even at the graph produced above, can see that climate models were much, much, much better at predicting global warming over the past decade,” John P. Abraham, professor of thermal sciences at the University of St. Thomas told FoxNews.com.
The bet offered to Gore indicated that the models would be judged throughout the period, and not on just the final year.
Armstrong says the last two years are just an anomaly and that his no-change model performed better overall.
“Temperature goes up, it goes down. If you happen to end on an upnote... that’s not the scientific thing to look at,” Armstrong said.
Abraham also critiques Armstrong’s expertise, saying that he “has no experience in climate science, [and] has to rely on fancy statistics to claim victory.”
Armstrong is a marketing professor, but says that he focuses on forecasting methods. He often predicts things like automobile sales.
“I've been doing forecasting research for 40 years, and heard about the global warming movement... it took about a week to find that this is a propaganda campaign. It's just a mass hysteria”
Over recent decades, however, the earth has been warming. The data source Armstrong uses for his bet – official satellite data that’s logged by climate professors at the University of Alabama – show that the Earth has warmed by about one degree Fahrenheit since 1979, when data collection began.
A spokesman for the Union for Concerned Scientists said that people should instead refer to government data based on weather stations, which show slightly more warming.
Armstrong says he uses the satellite data because the weather station data are “contaminated by poor maintenance and location of weather stations... and unexplained adjustments.”
A spokesman for Gore did not return a request for comment Thursday.
Armstrong adds that, if Gore is unhappy with the results and thinks they were due simply to luck, he is happy to extend the bet for another decade.
Who would win a bet from 2007 – 2027? Armstrong and Australian researcher Kesten Green say they’ll track that every month at TheClimateBet.com.
Maxim Lott can be reached on Twitter at @MaximLott