Unreliable Oracles

Political prediction markets took it on the chin in the midterms. Meanwhile, regulators are growing increasingly skeptical of such markets. Forget all that – their future has never been brighter.


First, the bad news. How badly did the markets whiff the 2022 elections?


The popular platform PredictIt, which has since 2014 operated under a no-action letter from the Commodity Futures Trading Commission (CFTC), had Republicans at 74% to retake the Senate on the eve of the election, well north of staid election forecasters like 538 (59%) and The Economist (57%). Crypto prediction market Polymarket fared similarly, with the GOP’s Senate chances peaking at 76% before reality set in. PredictIt traders saw the most likely outcome being a red sweep of the four toss-up states (Pennsylvania, Nevada, Arizona, and Georgia). Pending a likely Warnock victory in the Georgia runoff, all four will have gone the other way.


While the markets were not alone in seeing the “red mirage,” they were far more assured of its reality and magnitude than the pollsters and media forecasters. Whether a product of the lopsided polling errors that have plagued recent cycles or the much-maligned “MAGA money” said to skew market prices, the crowd with the most tangible skin in the game got shellacked.


Ignoring for the moment the fact that the markets have acquitted themselves remarkably well in most prior cycles, performing as well or better than the clearest-eyed election modelers, doesn’t this midterm faceplant vindicate the skeptics who view prediction markets as little more than playthings of amateur degenerates?


Yes… so long as we fundamentally fail to understand what markets are and their purpose.


Prediction markets, like any markets, are real-time aggregators of information. Not just of raw data but of overlaid analysis, tinged – for better or worse – with intuition and emotion. The price signals stemming from that aggregation can be distorted by those tinges, particularly when the underlying subject matter is so inherently emotionally wrought. Still, more often than not, incorporating diverse, democratized voices can be expected to improve the consensus signal, not degrade it. Within well-structured platforms, this is nearly tautological, as markets are not mere voting mechanisms – by design, they will systematically punish ideologically-driven decision-making and reward dispassionate insight. No amount of “dumb money” can be expected to impair a market’s signal for long, as any apparent mispricing can be quickly exploited by offsetting “smart money.”


Therein lies the part of the rub, of course. Few prediction markets meet the “well-structured” test. On PredictIt, traders’ positions are artificially limited by regulators to $850 (at least until February, when they are set to shut down permanently following the CFTC revoking their no-action relief), while markets with uncapped positions like Polymarket and overseas betting exchanges cannot legally cater to traders in the U.S.


But let’s allow the reasonable stipulation that even absent these constraints, the markets would’ve gotten the midterms wrong. Now, can we finally castigate them as unreliable and thus overdue for regulatory shuttering?


Yes… so long as we proceed to shut down equity, fixed income, foreign exchange, and commodity markets next.


These most liquid, most optimally microstructured markets that comprise the realm of respectable public financial markets are, of course, at their core also prediction markets, seeking to forecast and discount future cash flows based on diverse, democratized theses on global macroeconomic trends, corporate governance, the whims of policymakers, and consumer behavior.


The institutional participants who dominate these markets are well-educated and well-compensated to get these theses right. Even so – and I do hope you’re sitting down for this – they often fail in spectacular fashion. But when a company beats or falls short of earnings expectations and its stock gaps up or down, when an economic data point roils global markets, and even when obvious-in-retrospect black swans bring the financial system to its knees, few among us call for an end to capital markets.


That’s because the value of markets is not reliant on their infallibly seeing the future. Equity and fixed-income markets are mechanisms for reallocating capital to more efficiently coordinate productive enterprises. Commodity futures markets are mechanisms for efficient reallocation of risk.


Likewise, prediction markets or, if you like, “event futures” are mechanisms to aggregate dizzyingly disparate, granular, real-time data points into probabilistic estimates of future occurrences. If they’re sometimes wrong – even catastrophically, embarrassingly wrong – their worth is hardly disproven. They have distilled and communicated an aggregate expectation, the wrongness of which might tell us something useful about our perceptions and our politics.


What’s more, even beyond the value of their usually decent price signals, event futures can (and, in my opinion, shortly will) provide a legitimate and likely gargantuan hedging function to market participants with economic exposure to the events underlying such markets. From election outcomes to energy policy, tax rates to tariffs, the policy decisions made in voting booths, legislative chambers, and high courts can make or break entire industries, to say nothing of their impact on small business owners and households.


Markets that enable participants to hedge and observers to better handicap both the likelihood and, crucially, the perceived likelihood of such critical outcomes should be celebrated and elevated, no matter how big an egg they laid in November.


Flip Pidot

Flip Pidot, CFA, is the founder and CEO of American Civics Exchange (amciv.com), a dealer of OTC swaps on political outcomes, the CIO of event futures fund manager Sharp Square Capital, and a former market curator for PredictIt. He lives on Long Island with his wife and three sons.

We welcome for consideration all submissions that adhere to three rules: nothing defamatory, no snark, and no talking points. It’s perfectly acceptable if your view leans Left or Right, just not predictably so. Come write for us.

Share With Your Connections
Share With Your Connections
More Exclusive Content
The Latest News from Smerconish.com in Your Inbox
This field is for validation purposes and should be left unchanged.

We will NEVER SELL YOUR DATA. By submitting this form, you are consenting to receive marketing emails from: Smerconish.com. You can revoke your consent to receive emails at any time by using the SafeUnsubscribe® link, found at the bottom of every email. Emails are serviced by Aweber

Write for Smerconish.com