When statistician Francis Galton asked eight hundred people to guess the weight of a slaughtered ox in 1906, he found out that the median of the guesses was surprisingly accurate: 1207 pounds. The true weight of the ox was 1198 pounds, which meant the crowd was accurate within 1%. This was better than nearly all of the individual guesses as well as of the guesses of the cattle experts.
In another experiment, American economist Jack Treynor asked 56 people to estimate the number of jelly beans in a jar. Guesses varied wildly when he asked people individually, but when he took the average of all the guesses, the estimate was once again remarkably accurate: 871 versus 850 actual beans in the jar. Only one individual of the 56 had made a better guess.
This phenomenon is known as the wisdom of the crowd: the average or median of the answers of a group of people is often far more accurate than the answer of any individual in that group. On the Internet, websites such as Wikipedia, Stack Exchange, and Quora rely on the wisdom of the crowd to deliver accurate information to the world.
The wisdom of the crowd lies at the heart of blockchain project Augur. Created in 2015 on the Ethereum blockchain, Augur is one of the oldest blockchain projects and it’s still going strong. Among its advisors are heavy hitters such as Vitalik Buterin (founder of Ethereum), Robin Hanson (father of modern prediction markets), and Elizabeth Stark (CEO of Lightning labs).
But what is Augur and what makes it unique? How does it work, what are some of its problems, and how will it look like in the future? Find out below.
What is Augur?
Augur is a decentralized prediction market protocol that’s owned and operated by the people that use it. Ultimately, Augur wants to be the place where people can go to estimate the probability of any imaginable event occurring. Want to know whether Bitcoin will reach $10,000 again by the end of the year? Augur it. Want to know whether we’ll have people on Mars by 2030? Augur it. Want to know if a category three or higher hurricane will hit Florida over the next three months? Augur it.
While the first two examples above are more for curiosity's sake, the final example is a genuine use case that can show the value of Augur in real life. If you’re a farmer in Florida, knowing the relative probability of a hurricane destroying your crops will give you time to prepare and to hedge yourself against that risk. With hedging, I mean protecting yourself against financial loss by betting on the event happening. In this case, by betting on a category three or higher hurricane to hit Florida over the next three months.
Another use case of Augur is company forecasting. Research has shown that a sizeable group of employees and people from the public can produce a company forecast that’s often more accurate than the people whose job it is to do the forecasting. Government agencies, too, are using the wisdom of the crowd for economic and geopolitical forecasts. In these cases, Augur can serve as the protocol and platform for those bets, while also giving those companies and government agencies a way to hedge against the financial risk of them not hitting their forecast targets.
Augur can also be used for more personal use cases. Imagine you’re hosting an outdoor party in London in a few weeks. You know that people won’t show up if it rains, so you ask Augur what the chances are that it’ll rain on the day of your party. After a while, various people will have given their estimate of whether it’ll rain that day, and you’ll have a reasonably accurate estimate. If it’s likely to rain, you can bet some of your own money on it raining.
When the day comes and it doesn’t rain, you’ll lose the money you bet, but at least the party will go through and people will show up. When it does rain, people won’t show up, but at least you’ll have gotten some money back, money that you’re likely to have spent buying food and organizing everything. It’s a way of insuring yourself.
Additionally, Augur gives power to people who might not otherwise have access to advanced financial markets. If you live in a developing country where it’s impossible to buy Apple shares, you can use Augur to predict that the Apple share price will go up or down. It’s a type of financial derivative, where you don’t own the shares, but can still buy or short where you believe the price will go. All you need is internet and some money.
How Does Augur Work?
A decentralized market prediction protocol is the main idea of Augur, but how does it work specifically? If you’re unfamiliar with betting or trading, it might take some time getting used to. To begin with, if you want to ask a question or make a bet, you’ll need to install the Augur app. Once installed, you’ll need to give it some time to sync up with the blockchain that it runs on. This can take a while, depending on your internet speed, but only needs to be done once. Alternatively, if you just wanna see what Augur is about, you can use its web app to follow along without syncing the full blockchain.
Once you’ve synced the full node, you can access Augur and you’ll see an interface similar to the screenshot above. As I’m writing this article, the top two questions are about one of the most popular topics on Augur: Donald Trump and U.S. politics. Other popular topics are cryptocurrency and soccer speculation.
Let’s take the top question as an example to explain how Augur works. Will Donald Trump be re-elected in 2020? Currently, 49.90% of people think he will be. This question has a trade volume of almost 289 ETH (approximately $39,000 in today’s price), open interest of 175 ETH (this means money at stake, $23,000), an estimated fee of 0.26% and an expiry date of 3 AM UTC+1, Dec 6, 2020.
So on the 6th of December 2020, which is when we’ll know the answer to the question, at least $23,000 will go to the people who predicted the answer correctly, with 0.26% of the winnings going to the person who asked the question.
Each question can be clicked on to give more detail about the question and to show the trade interface. To understand this better, think of the question in terms of shares. The price of the share fluctuates between 0 (no, he won’t be re-elected) and 1 (yes, he’ll be re-elected).
If you think he’ll be re-elected, you can buy one share for an average of 0.4695 ETH, which is the price between what buyers are willing to buy for and sellers willing to sell for. If you think he won’t be re-elected, you can short the share and it would cost you 1 - (0.4695 ETH) = 0.5305 ETH. When the result comes in on Dec 6 2020, you’ll either win 1 ETH per share or lost everything you bet.
This both gives you an indication of the likelihood of the event happening and the possibility to trade on that likelihood, all in a way where the users retain access over their funds and where no third party has power. This isn’t limited to yes/no questions like the above, but can also be used for multiple choice questions or numerical ranges.
Problems with Augur
Augur officially launched in July 2018, and its first big milestone is that it works. Between one to twenty new questions are being asked on the platform every day since its launch, and the platform hasn’t gone through any major hiccups where people accidentally lost money. In the relatively young blockchain industry, that’s an achievement already.
But Augur still wrestles with a few big problems, the first of which is its accessibility. Downloading the app and syncing it to the Ethereum blockchain is slow and difficult for people with irregular access to internet. On top of that, you need to buy Ethereum, which is still a cumbersome process that is likely to discourage those unfamiliar with cryptocurrencies.
Secondly, if you’re unfamiliar with trading, it’s difficult to use Augur. Consumers aren’t traders. They want to make a simple bet without having to think about price, bid price, ask price, trade volume, and all the other terms associated with trading. Will Donald Trump be re-elected in 2020? Yes? Here’s how much it’ll cost you and how much it’ll yield you. No? Same thing. That’s how easy it should be.
This is somewhat made better with the many applications built around Augur. After all, Augur is a protocol and doesn’t necessarily have to be a front-end interface. It can run in the background too. Predictions Global is an excellent website that provides a much cleaner UI for all the questions asked on Augur, while also providing an easier way to sort through the questions.
Thirdly, most questions asked on Augur do not have enough liquidity. This means it’s not easy to find people to buy shares from or sell shares to. Considering the wisdom of the crowd relies on as many people as possible for an accurate prediction, this defeats the purpose of using Augur. While the vision is to have a platform where anyone can ask any question, currently that’s mostly limited to soccer, politics, crypto prices, and the Oscars.
Fourthly, some of the questions aren’t without controversy. Augur is a blockchain project and the company that maintains the protocol, the Forecast Foundation, goes to great lengths explaining on their website that the responsibility and legality of using Augur lies in the hands of the end user (considering betting is illegal in many countries).
As such, people are free to ask whatever they want, and assassination markets have sprung up. Do you think [insert name] will die before [insert time]? Given enough money at stake, this is effectively placing a bounty on someone’s head, without authorities being able to know who’s behind the question (the only info about the market maker is a crypto address).
Users can police the questions asked by marking them as invalid, to ensure no one receives a payout. This is how the platform is currently regulated, and it’s again relying on the wisdom of the crowd to determine which questions are wrong and which are not. So far, it works, but it’s yet to be seen whether it’ll continue to work once some of these questions start seeing significant amounts of money flowing through them.
A Glimpse of the Future
What’s been described above is only the tip of the iceberg. The Augur whitepaper describes the inner details about Augur in much greater detail. The team behind Augur is dedicated and active, and they have some high-profile advisors backing them. They have a strong vision and they’re continuously improving the protocol. While they still have serious problems to overcome, the official platform hasn’t been out for a year yet. That’s the equivalent of Facebook in 2005, Google in 1999, Twitter in 2007. There’s a long way to go, but it’s a young project.
Augur gives a glimpse of the future of prediction markets. A place where anyone can ask any question and receive a relatively accurate prediction about it. Where anyone can have access to advanced financial instruments to invest or hedge their risk. Where aspects of life become a bit more knowable. It’s controversial, innovative, and incredibly interesting.