The Hype, the Horror, and the Letdown of Prediction Market Augur

Perhaps no cryptocurrency project exemplifies the cycle of delirium, drama, and ultimate disappointment that characterized the ICO boom more than Augur. In 2015, the platform became one of the first startups to raise capital by creating a proprietary cryptocurrency and selling it on the Ethereum blockchain. Augur’s goal was to create a decentralized prediction market that would allow people to bet on anything they’d like, relying on markets to crowdsource odds and determine outcomes.

Remember the first episode of The Sopranos where Tony hits the guy with his car because he didn’t pay up on a bet? If they’d been using Augur, the money would have automatically gone to Tony courtesy of Ethereum’s smart contracts. This was the dream of Augur: facilitating trust-less markets on the blockchain and saving poor schmucks from having their legs broken by pissed-off bookies.

The platform aims to be a little like a combination of sportsbook, trading floor, a speculation market, and your buddy’s basement when you’re super stoned and down to make a bet about literally anything for a laugh. You can use Augur to bet on whether the United States will experience a major earthquake in the month of October, or if NASA’s InSight will make it to Mars in one piece. At least one market, addressing the question of whether the weather would be “good” on the day of the Bastille Day parade in Paris, is essentially unresolvable but raises the interesting philosophical question of whether good weather in Paris is objectively quantifiable. One site, Pdot Index, proposes to track action on Augur markets as a way to generate approval ratings for celebrities—it’s bullish on Ariana Grande, not so much on Donald Trump—while another, Predictions.Global, assembles every Augur market in one place to give you an idea of what the future may hold. It’s there that I discovered that, according to the platform’s users, Brett Kavanaugh has an 79 percent chance of being confirmed to the Supreme Court.

As it struggles to attract a large number of users placing bets in a diverse range of markets, Augur faces increasing regulatory scrutiny because of the markets that it does have.

When the cryptocurrency market matured in 2016 only to head parabolically skyward in 2017, Augur—whose ICO, which closed three years ago on Monday, netted $5 million in REP tokens at 60 cents—became an example of the revolutionary potential of the blockchain. In October 2017, Forbes referred to Augur as one of the crypto projects that “will reshape capitalism as we know it.” Two months later, the value of the REP token briefly touched the hundred-dollar mark, prompting another Forbes writer to call the platform “the future of prediction markets.”

One year later, it seems that Augur’s frantically rushing tides have stilled. REP is currently trading at about $13, and, like any pool of standing water, it’s stagnant and full of bugs. And potentially illegal options trading. And also news stories about people using it to bet on murders. Oh, and to top it all off, its founders are currently facing a $152 million lawsuit from a former partner who claims he was forced out of the company right before everybody else got rich from the ICO. (According to documents filed with the San Francisco Superior Court, the parties are currently negotiating a settlement agreement.)

As it struggles to attract a large number of users placing bets in a diverse range of markets, Augur faces increasing regulatory scrutiny because of the markets that it does have. All of which begs the question: Are Augur’s current struggles baby steps on the path to a new, more market-oriented world, or are they harbingers of a shit-show that will go down in history as an unholy combination of Silk Road and

Prediction markets have long held interest in academia because of their relative accuracy as a form of forecasting. “Academics prefer prediction markets to punditry because pundits are never punished for being wrong,” said Kevin Munger, a postdoctoral fellow at the Princeton Center for Democratic Politics. (Disclosure: Munger is a friend and former colleague.) “For predictions about the future to be meaningful,” he adds, “you have to put a level of confidence in them.” In this case, that means plunking down some cash that you’re willing to lose if you’re incorrect, incentivizing judicious forecasting in a way that Munger said is similar to the (pro-crypto) academic Nassim Taleb’s principle of having “skin in the game.”

But according to Augur’s own markets, the future might not be so rosy. One open market on the platform places the chances that the Forecast Foundation, Augur’s Estonia-based nonprofit parent organization, will face regulatory action before the end of the year at 65 percent. However, this 65 percent figure is just the opening offer of the person who made the market—no one has bothered to put any money on the possibility either way. This reveals what is perhaps Augur’s greatest current flaw: Not that many people are on it. Of the 1,200 or so bets listed on Predictions.Global, only a tenth or so are liquid, meaning that someone has taken the other side of the bet.

“Markets are social technologies,” said prediction market researcher and developer Ben Goldhaber of Parallel Forecast. “You need the platform, but more importantly you need the people.” While Augur is certainly an interesting project featuring modern, tasteful design (check the website if you need proof), it’s hard to use. First, you have to download the Augur app, and then connect your computer to an Ethereum node to log into the site. To create a market, you must own REP, which is also awarded to those who correctly judge the outcome of an event. Meanwhile, to bet on something, you must own ETH, which involves connecting a wallet such as Trezor, Edge, or MetaMask to your Augur account—a lot of steps to take to bet on whether Chance the Rapper will put out an album or not. As Goldhaber put it, “Augur has problems with user experience and liquidity that make it a high-friction process.”

When an emerging technology is so complex that it effectively shuts out casual converts, it leaves itself open to being used for other purposes. Consider Vine. The platform was certainly not designed with a new genre of short-form comedy in mind, but that’s ultimately what the six-second video format provided. The users determined the purpose of the product instead of its creators prescribing it. While the worst that Vine gave us was the occasional flash of unfunny racist or misogynistic jokes (and also Jake Paul), this same hands-off approach can also leave you with a community full of existentially disturbing moral reprobation—as in Augur’s “assassination markets.”

These markets, as you’d assume, involve people betting on whether or not public figures will die by a certain date. While little to no money was actually at stake on such markets, that didn’t stop outlets such as Mashable, The Independent, and Newsweek from writing about them, bringing Augur its first rush of mainstream press. Even if they’re all the work of some troll, there’s something intuitively queasy about such markets’ sheer existence, as if they tap into our most morbid impulses and put them on the internet for all to see.

When I reached out to the Forecast Foundation asking if there was any sort of plan to do anything about Augur’s assassination markets, their press representative stressed the degree of separation the nonprofit has from the Augur platform and referred me to’s FAQ section, which claims, “The Forecast Foundation has no role in the operation of markets created on Augur, nor does it have the ability to censor, restrict, control, modify, change, revoke, terminate or make any changes to markets created on the Augur protocol.”

“The main advantage to putting a prediction market on the blockchain is and always will be escaping regulation.”

After following up with more questions, including one asking how a market might be removed from the Augur platform once created, Augur co-founder Joey Krug told me, “There are a few possibilities. One is that reporters in Augur can resolve a market as invalid, which returns funds to the traders. The other is that Ethereum miners could choose to censor the transactions.” In other words, Augur relies on the same sort of community governance model as something like Wikipedia, in which users decide what stays and what goes. (In June, Augur’s developers publicly destroyed its “kill switch,” meaning they truly do not have special access to the network.)

According to Predictions.Global, the vast majority of money on Augur is concentrated in a handful of markets that have nothing to do with murder. Instead, each speculate on the price of specific cryptocurrencies, including the second-largest market by money at stake, which concerns the future price of Augur’s own REP token. (The only non-crypto market in the Predictions.Global top ten is on whether Trump will win the Nobel Peace Prize.)

As Bloomberg noted in a July 26 story, “creating a way to bet on the value of goods in the future is basically the definition of a type of derivative known as a binary option,” and this unregulated end-around for investors has put Augur at risk of a crackdown from the Commodity Futures Trading Commission. Also in July, the CFTC announced that it had extracted a total of $3 million in civil penalties from exchanges offering binary trades made without CFTC approval. On the other hand, the CFTC explicitly allows the trading of virtual currency futures contracts, which places Augur in something of a legal gray area. Goldhaber told me that he sees the situation as a feature, not a bug, saying, “The main advantage to putting a prediction market on the blockchain is and always will be escaping regulation.”

There could still be a version of the Augur story that ends happily. Just as bitcoin may become a decentralized, open-source corollary to the U.S. dollar, Augur could become a corollary to gambling, the stock market, and even the determination of objective facts. Already, there are clever proposals that could accomplish all sorts of things. One of the most interesting markets on Augur is one that asks, “Will a pee tape featuring Donald Trump emerge before the end of his first term?” The market, started by Aaron Lammer of the podcast Coin Talk, is certainly not tasteful. It stipulates that “urine must either be visible or discussed in [the] tape.”

Disgusting? Undoubtedly. But innovative? Well, maybe. Goldhaber says such a market “could encourage people to find the pee tape. Or at least create a deep fake.” In other words, if there were enough money riding on the pee tape market, someone who either knew of the pee tape’s existence or was capable of creating a convincing counterfeit could bet on “Yes,” release the tape, and collect a crowd-sourced bounty. And because Augur’s judges are rewarded in REP tokens for verifying the outcome of markets on the platform, having a small army of both experts and laypeople trying to vet the authenticity of a gross sex video suggests that a radical new way of assessing reality in an increasingly unreal world could be just around the corner. As Goldhaber put it, when you create a market on Augur, “You’re incentivizing people to really search for the truth.”