The Global Warming Challenge

It’s been over 2 months since Armstrong sent a simplified challenge with extended deadline to Al Gore. We’ll be keeping track until Mr. Gore responds.

Paul Georgia on National Review Online concludes with a reference to the Green and Armstrong (2007) paper in his article “Speculation Elimination,” critiquing the press’s emphasis on the Bush administration’s “censoring” of science.

The claim that the Bush Administration censored science is without merit. What it seems to have done, is cut the portions of the testimony that were based in expert speculation about the future. According to the scientific literature on forecasting, expert opinion is the least reliable source for accurate predictions.

A new paper by Professors Scott Armstrong and Kesten Green, leading experts on forecasting, argues that “Comparative empirical studies have routinely concluded that judgmental forecasting by experts [rather than scientific forecasting] is the least accurate of the methods available to make forecasts.” They also show that, “Agreement among experts is weakly related to accuracy,” when it comes to forecasting.

The media storyline is backwards. Rather than censoring science, the Bush Administration responsibly removed baseless speculation from the CDC’s testimony. If the purpose of congressional hearings is “fact finding,” then such speculation is inappropriate and the Administration acted appropriately.

Paul Georgia is the executive director of the Center for Science and Public Policy. The article was adapted from a paper published by the Center.

After a series of cordial exchanges, Al Gore had said that he was too busy when the challenge was first offered. As a result, Scott Armstrong extended the deadline and simplified the process such that the only actions required of Al Gore are to 1) sign the letter and 2) deposit $10,000 in the Challenge Fund. The letter below explains the new simplified process.

November 28, 2007

Honorable Albert Gore
2100 West End Avenue
Suite 620
Nashville, TN 37203

Dear Mr. Gore,

Thank you for your previous emails and your letter concerning the Global Warming Challenge. You had mentioned that you were looking forward to reading my book, Principles of Forecasting; I believe the principles are critical to making accurate forecasts for global warming.

In earlier communications, your staff mentioned that you were too busy at the time to enter in the Global Warming Challenge. As a result, I am happy to extend the deadline beyond December 1st, 2007, to March 26th, 2008. In addition, I have found a way to make your task much easier. I propose that you place $10,000 on the Hadley Centre Forecasts.

This would merely require that you sign on the “I agree” line at the bottom of this letter. Each of us would then arrange to transfer $10,000 to a trust of your choosing. The money would be invested in mutual funds, and the proceeds would go to the winner’s charity on March 26th, 2018.

As a brief reminder, here is an overview of the original Global Warming Challenge (details can be found at http://theclimatebet.com):

Al Gore is invited to select any currently available fully disclosed climate model to produce the forecasts (without human adjustments to the model’s forecasts). Scott Armstrong’s forecasts will be based on the naive (no-change) model; that is, for each of the ten years of the challenge, he will use the most recent year’s average temperature at each station as the forecast for each of the years in the future.

Details on the 10-year bet would be handled with discussions between me the Hadley Centre. I would ask an independent board to aid in this process of finding an appropriate design and to monitor the progress of the bet. You would be kept up to date, and you would have the right to ask the board to consider changing aspects of the design.

In The Assault on Reason, you stated your interest in the use of science in addressing global warming:

“We must, for example, stop tolerating the rejection and distortion of science. We must insist on an end to the cynical use of pseudostudies known to be false for the purpose of intentionally clouding the public’s ability to discern the truth. Americans in both parties should insist on the reestablishment of respect for the rule of reason. The climate crisis, in particular, could cause us to reject and transcend ideologically based distortions of the best available scientific evidence.” p. 10

I believe we have a common goal. Although I expect to win, the purpose of the challenge is to promote interest in a scientific approach to forecasting. Climate experts have done much useful work to explain the past, but their approach to forecasting does not adhere to scientific principles.

Sincerely,

J. Scott Armstrong

For the signature of Mr. Gore:

____________________________________________
I agree to the terms of the Global Warming Challenge with the variations specified in this letter

October 26, 2007: Senator Inhofe addresses the Senate floor for two hours on the tipping point of climate alarmism (full text| video clip 1 | video clip 2). Scott Armstrong’s Global Warming Challenge to Al Gore, along with the Green and Armstrong’s global warming paper (2007) were both mentioned when citing challenges to climate model accuracy:

Internationally known forecasting pioneer Dr. Scott Armstrong of the Ivy League University of Pennsylvania’s Wharton School, challenged Gore to a $10,000 bet in June over the accuracy of climate computer models predictions. Armstrong and his colleague Professor Kesten Green of Monash University’s in Australia, found: “Claims that the Earth will get warmer have no more credence than saying that it will get colder.” According to Armstrong, the author of “Long-Range Forecasting,” the most frequently cited book on forecasting methods.: “Of 89 principles [of forecasting], the [UN] IPCC violated 72.”

Bonner Cohen’s article on TCS Daily, “Gore Dodges Repeated Calls to Debate Global Warming” covers the increasing numbers of skeptics who have challenged Bush, including Armstrong. Does understanding both viewpoints change public opinion? Below is an excerpt summarizing a recent debate between skeptics and alarmists, and audience reaction:

“Gore’s reluctance to go toe-to-toe with global warming skeptics may have something to do with the - from the standpoint of climate change alarmists - unfortunate outcome of a global warming debate in New York last March. In the debate, a team of global warming skeptics composed of MIT scientist Richard Lindzen, University of London emeritus professor of biogeology Philip Stott, and physician-turned novelist/filmmaker Michael Crichton handily defeated a team of climate alarmists headed by NASA scientist Gavin Schmidt. Before the start of the nearly two-hour debate, the audience of several thousand polled 57.3 percent to 29.9 percent in favor of the proposition that global warming is a “crisis.” At the end of the debate, the numbers had changed dramatically, with 46.2 percent favoring the skeptical point of view and 42.2 percent siding with the alarmists.”

Dire consequences have been predicted to arise from the warming of the Earth in coming decades of the 21st Century. Enormous rises in sea level represent one of the more dramatic forecasts. A recent article provided sea-level forecasts based on experts’ judgments of what will happen. These judgments are in turn based on experts’ predictions of global warming. The article made no reference to scientific forecasts. As shown in Green and Armstrong (2007) experts’ forecasts have no validity in situations characterized by high complexity, high uncertainty, and poor feedback. Numerous other scientists also criticized this approach.

To date we are unaware of any forecasts of sea levels that adhere to proper (scientific) forecasting methodology and our quick search on Google Scholar came up short. If such forecasts are available, please provide citations and support as to their validity. As a first step, it would be useful to summarize studies that extrapolate long-term trends; this summary could provide a benchmark for comparison with other studies.

We will provide free access to them at publicpolicyforecasting.com and request commentary at theclimatebet.com. Media outlets should be clear when they are reporting on scientific work and when they are reporting on the opinions held by some scientists. Without scientific support for their forecasting methods, the concerns of scientists should not be used as a basis for public policy.

Kesten Green and Scott Armstrong

In the Australian Financial Review (September 8, 2007), Mark Lawson’s article, “Global warming sceptics fuel hot debate” features Dr. Green, and highlights the Green and Armstrong paper. The following is an excerpt:

Global warming sceptics fuel hot debate
(Click here for full text)

Mark Lawson

“Despite being scorned, derided and accused of links with oil companies, the climate change sceptics are still out there and, although the greenhouse lobby will never admit it, occasionally scoring major points. They may also be more numerous than the greenhouse lobby or politicians believe…

A much more serious, if not devastating, attack on greenhouse claims concerning likely future temperature increases was the recent release of a paper entitled Global Warming: Forecasts by Scientists versus Scientific Forecasts.

Written by J. Scott Armstrong, a professor of marketing at The Wharton School, University of Pennsylvania, and Kesten Green, a visiting fellow at the business and economics forecasting unit at Monash University in Melbourne, the paper assessed, as forecasts, the temperature projections made by the Intergovernmental Panel on Climate Change earlier this year. It found little to approve.”

Click for Armstrong’s PowerPoint presentation, titled “Global Warming: Forecasts by Scientists Versus Scientific Forecasts” from September 13th, 2007.

A Congressional Briefing about forecasts of global warming given by Scott Armstrong on Thursday, Sept 13 is now available on YouTube (Part 2 and Part 3). The briefing was based on the Green & Armstrong paper, “Global Warming: Scientific Forecasts or Forecasts by Scientists?” The global warming paper is the first of what they hope will be many forecasting audits of global warming studies to be presented on the new Special Interest Group page at http://publicpolicyforecasting.com.

Although climate scientists have extensive climate knowledge and are aided by computer models, their long-term forecasts are essentially their judgments of what will happen in the decades to come. Unstructured judgments are used to determine the structure and much of the content of climate models. Unstructured judgments are also used to accept or reject model outputs.

Green and Armstrong (2007) found that climate modelers do not use evidence-based principles to forecast. In fact, they violate many principles of forecasting.

One solution to this problem is to combine what is known about forecasting methods with what climate scientists know. Knowledge about forecasting is provided at http://forecastingprinciples.com. In addition, the site lists many experts in forecasting methods who are available as consultants.