Climate Forecasting Tournament by Good Judgment
Phil Tetlock's work suggests that forecasting tournaments may moderate polarizing policy debates by focusing on accountability.
Forecasting tournaments have the potential to moderate polarizing policy debates because accountability fundamentally alters the parameters of the discussion.
Phil Tetlock and the rest of the Good Judgment team are actively involved in running forecasting tournaments focused on geopolitical events and policies. [Seehttps://www.gjopen.com/challenges/1-polarization-v-prediction-iran-2016
A long term forecasting tournament on climate science and policy, that rewards forecasters for accurately predicting future outcomes, may create a more fruitful platform for discussion on complex issues. Forecasters who are too extreme in their forecasts score poorly in a forecasting tournament, while those who assign more accurate probabilities are rewarded. Through this mechanism, all participants have an incentive to forecast accurately, or they will look silly by receiving a poor score.
The infrastructure to run a forecasting tournament is already established. See the Good Judgment Open as an example.
What actions do you propose?
MIT could sponsor a long term (more than a decade) forecasting tournament on climate science and policy issues. Everyone would be invited to participate. The basic model is already established - see Good Judgment Open.
In an ideal world, leading public figures from both sides of the debate would participate and forecast on the same questions. The Koch brothers vs. Bill McKibben would be an ideal contest.
Even if the only participants are members of MIT, the forecasting tournament would provide useful information on various topics. How much will the sea level rise over the next decade? How much will the average surface temperature change in the next decade?
The physical actions for running a forecasting tournament are straightforward and established. All that is required is the will.
Who will take these actions?
MIT would need to fund the forecasting tournament. Someone will need to keep the forecasting site up and running, figure out the right questions to ask, market the tournament, etc.
Where will these actions be taken?
MIT can physically host everything in a small office on campus.
How will these actions have a high impact in addressing climate change?
What are other key benefits?
A forecasting tournament would allow for a potentially less polarizing debate on important climate issues.
What are the proposal’s costs?
The costs could run quite low if interested parties at MIT were willing to volunteer their time and energy.
But, two half time employees or a couple graduate students might be enough to run the tournament. The costs including space, computers systems, salaries, marketing expenses, etc could run at $250,000/year.
In the short term, the forecasting tournament could partner with Good Judgment Inc, who already offers services that allow for forecasting tournaments. [Full disclosure: I am not paid by Good Judgment Inc, but I know the people involved and help them on a voluntary basis.]
In the medium/long term, MIT could choose to develop the software and systems to run their own tournament completely in house - although this may not be necessary.
I'm not sure.
Good Judgment Open is an example of what is currently happening and available:
"Superforecasting" by Phil Tetlock is a recently published NYT bestseller that summarizes his research finding for forecasting tournaments.