Throughout his career, Stephen Hawking made several famous public wagers on various topics in astrophysics and cosmology. The stakes were never truly life-altering. Hawking, on the other hand, was notorious for his willingness to stake some scientific speculation. That's a rare occurrence, which is strange given how much time we spend speculating about the future. Consider this.

Every political discussion, every disagreement over where to eat for dinner, every debate over when to use the Oxford comma boils down to a disagreement about what will happen in the future. That person's election will have negative consequences. This restaurant's food is delicious. People will be more confused if the Oxford comma is not used, because there is so much prediction in everyday discourse.

You'd think we'd have a lot of respect for people who have exceptionally accurate mental models and are consistently correct about what's going to happen. However, Hawking's public declarations of his guesses are the exception rather than the rule. Typically, we are content to state or overstate our opinions, frequently using language that vastly exaggerates our confidence in them for rhetorical purposes. How often have you heard someone state that they are "99 percent certain" of something? Usually, it's just a bit of exaggeration. They're fairly certain, but not certain.

If their confidence estimate was accurate, there's a wager they should be willing to accept. They win a dollar if their prediction is correct. If it is not true, they must pay a few dollars like $87. How many people do you believe would be willing to place that wager?

Prediction Markets

Prediction markets are essentially exchanges that operate on the same principle, similar to stock markets. Alternatively, rather than speculating on a company's stock or placing bets on card hands, prediction markets allow people to put money down on assertions about the future or contracts. Contracts differ from stocks in that their value is determined by a probability distribution rather than a monetary value. And when you invest, you buy in at a certain percentage, such as "computer program passes turning tests by 2025 at 65 percent," which means you're about 65 percent certain it will happen. Alternatively, you could save suckers' money on vaccines that cause autism by putting as much catch as you can spare all the way down to 1%.

The contract's value fluctuates in a manner similar to the value of a stock price. When a large number of people bet on something, the percentage value rises. If they bet against it happening, the value falls. This results in a similar market value process.

Equilibrium or speculators who believe a contract's current percentage value is significantly higher or lower than it should be will invest accordingly, shifting the curve in that direction. When the contract expires, everyone's final bids are locked in, and whether the event occurs or not, it pays out based on their level of investment. If you were one of the only people to bet on an outcome that everyone else thought was impossible, you can make money whether it occurs or not.

The end result of this entire process is that the distribution of the contract price is a real-time index of the investors' certainty that it's correct, not their casual conversational speculation, but what they truly believe in their hearts. After all, it's all about the numbers. There's money to be made if they guess correctly, and money to be lost if they don't.

That is, a contract's probability distribution represents an incredibly accurate crowdsourced guess about its likelihood. Frequently a much more consistently correct guess than even experts with access to massive amounts of data can manage on their own. That is such valuable information that many large corporations, including Google, Microsoft, and Qualcomm, use internal prediction markets to inform business decisions. Of course, no market is perfect, and there are risks in blindly trusting the wisdom of the crowd to forecast everything.

Prediction markets draw a specific type of person, and this selection bias can manifest as skewed results. When defining a contract, there is also the issue of specificity. It is relatively simple to buy and sell a company's stock at a specific price. However, determining whether a specific event has occurred can be a time-consuming process. And, as with stock markets, there are numerous devious ways to profit from prediction markets. Or, even worse, to have an impact on the real world. After all, if you have 20 large writings on a political leader being assassinated within the next year.

However, research into these prediction markets has resulted in an intriguing advancement in the psychology of accurate forecasting. In 2011, a group of researchers funded by a US Intelligence initiative launched the Good Judgment Project, a study in which thousands of people were asked to make their best guesses on a series of geopolitical and economic events, with the results providing amazingly accurate projections of the future.

The general gist, however, is outlined in the book Super Forecasting, which was written by two of the project leads. Most of the mental apparatus shared by super forecasters appears to be debiasing psychological tools and balances that allow them to compensate for the usual suspects that make human estimates more a matter of wishful thinking than rigorous analysis.

Along with the studies, findings can include the concept of refining the mojo that powers prediction markets with individual track records using statistical techniques to weight certain guesses more or less, depending on the past performance of the guess art in question.

The market for recreational online prediction Metaculus employs such an algorithm to calibrate a prediction engine that operates in parallel with the public market and has astounding accuracy, nailing its guesses more than 90% of the time, better than even super forecasters. But, hey, perhaps you believe you can do better. Metaculus is one of several online prediction market games that you can play with virtual currency to see how you compare to other forecasters without gambling. There are also distributed cryptocurrency markets for those who believe they are capable of making real money.

If you're an opinionated person who frequently makes predictions about what's going to happen next, you might want to take a page from Hawking's book and stake your claim. Hopefully, you will fare better.

Lead image credits to bitedge.com

References:

Barbara, M., Lyle, U., Jonathan, B. (2014, March 21) Psychological Strategies for Winning a Geopolitical Forecasting Tournament [Journal]. Retrieved from https://journals.sagepub.com/doi/abs/10.1177/0956797614524255

Kenneth, A., Robert, F., Michael, G. et al (2008, May 16) The Promise of Prediction Markets [Magazine]. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.320.1811&rep=rep1&type=pdf

David, R., (2009) Forecasting Elections: Comparing Prediction Markets, Polls, and Their Biases [Research]. Retrieved from https://researchdmr.com/RothschildPOQ2009

Chris, H., Benjamin, L., Nick, V., (2015, May 18) If not polls, then betting markets? [Blog]. Retrieved from https://blogs.lse.ac.uk/politicsandpolicy/if-not-polls-then-betting-markets/

This article is 100% free from plagiarism!