Do you have a pair of crystal balls? IARPA would like to know. The acronym that stands for U.S. Intelligence Advanced Research Projects Activity is out for a random walk. The operative premise is that “a blindfolded monkey throwing darts at a newspaper’s financial pages could select a portfolio that would do just as well as one carefully selected by experts.”* If our expansive and well-equipped intelligence apparatus can be blindsided by the collapse of Evil Empires on life-support, what’s to say that the likes of Joe the Plumber could do any worse placing bets on the fate of North Korea’s quackocracy?
I was driving along several weeks back with my globally aware fourteen-year old son, Jed. We were considering the Twitter revolutions in Tunisia and Egypt. I posed a challenge: which neighborhood tyrant would be the next to go down? There was already some stirring in Yemen and Bahrain. What other candidates for collapse might there be? Oman? Iran? Jed went with Algeria, not a bad choice given the military suppression of an Islamic party election victory twenty years ago. I chose the monkey in the middle of Egypt and Tunisia – Libya’s Qaddafi. Can’t say precisely what it was in my caldron of knowledge of all things Qaddafi and Libyan. Maybe it had something to do with Keeping Up with the Ghonims, Google’s young social media man next door in Egypt. Qaddafi, being Qaddafi, hasn’t gone quietly into the night to join his Egyptian counterpart, Mubarek, so I haven’t collected my bet with Jed.
Qaddafi is every inch the devil-incarnate that we know. But who are these rebels that we don’t know? When asked to characterize the opposition, Secretary of Defense Robert Gates said, “It’s pretty much a pick-up ballgame…with no command and control.” The future government, Gates suggested would probably be worked out among Libya’s numerous powerful tribes. It is sobering, given Gates’ compound authority as former Acting Director Central Intelligence and SecDef, that Jed could have expressed equal uncertainty.
If Libya is a pick-up game played with tanks and RPGs, then it plays to a core skill of our President. It is Obama’s half-court game and die-hard fan’s eye that got him on ESPNs panel of forecasters for the NCAA men’s basketball tournament. This year his bracket choices went 29-3 through the second round placing him in the 99th percentile on ESPN.com after three rounds. His Final Four choices, number one seeds all, didn’t escape the Elite Eight, dropping Obama to a, nonetheless, respectable 83rd percentile when the net was cut down by the champions. Snoop Dogg came panting in at the 43rd while Dick Vitale, the voluble former coach turned b-ball bloviator dogged it out at the 21st along with ESPN pundit Scott Van Pelt.
These results line up with the findings of “Expert Political Judgment: How Good Is It?” by Penn Professor Philip Tetlock. The takeaway – acclaim of so-called experts is inversely proportional to the accuracy of claims they make about the future. It is far better to go with the “wisdom of crowds,” perils of groupthink and the lowest common denominator of consensus, notwithstanding. By Tetlock’s reckoning, the average of all five million March Madness predictions registered at ESPN will beat 80-90% of the individual predictions. Faced with a world checkered with unknown unknowns, IARPA’s Aggregative Contingent Estimation (ACE) Program has retained the professor to lead one of five teams in capturing this oracular phenomenon algorithmically.
ACE aims “to dramatically enhance the accuracy, precision, and timeliness of forecasts for a broad range of event types, through the development of advanced techniques that elicit, weight, and combine the judgments of many intelligence analysts.” Tetlock’s Good Judgment Project (GJP) team, has been soliciting recruits with no specialized background to conjecture on 100 impending possibilities on the world stage. “Will former Pakistani President Pervez Musharraf return to high office?” “Will Hamas recognize the state of Israel by the end of 2012?” Researchers from the GJP will evaluate combinations of individual forecasts to optimize the yield of “collective wisdom.”
A decade ago IARPA’s cousin DARPA over in Defense sponsored “Futures Markets Applied to Prediction.” FututreMAP was to harness collective intelligence through market-based trading mechanisms for predicting geopolitical instability and threats to national security. After all, orange juice futures, as we saw in Trading Places, are better predictors of weather than the National Weather Service’s forecasts. Detractors accused the Pentagon of wasting millions of taxpayer dollars on “terrorism betting parlors,” and “fantasy league terror games.” Defenders pointed out that traders in the Iowa Electronic Markets have been betting since 1988 with remarkable accuracy on the likely winner of the US presidential elections. As circumstances unfolded, it was left to Tradesports.com, a Dublin-based online trading exchange to take bets on the “survivability of Saddam.” But now Google has legitimized this approach by using prediction markets to “forecast product launch dates, new office openings, and many other things of strategic importance.”
For the C-Spanners out there who have longed to have their very own version of ESPN bracketology, the call is out for wonks looking to improve their “forecasting ability as part of cutting-edge scientific research.” Pay is a mere $150, but GJP will be providing an invaluable reality check. So go to http://surveys.crowdcast.com/s3/ACERegistration and get in touch with your inner Nostradamus.
*Random Walk Down Wall Street, Burton Malkiel