Augur predictions market goes live, confidently makes safe bets
Augur is worth watching to see what works, what doesn't work and what is just plain weird about using crowd wisdom.
In 2001, the Defense Advanced Research Project Agency (DARPA), which might be thought of as the United States' official mad scientist agency, started experimenting with crowdsourcing predictions of future terrorist attacks and other extreme events.
The system they conceived was essentially a way of using cash prizes to motivate a lot of people to bet on future events. It was swiftly shut down under a barrage of criticism from people who criticised it for wasting taxpayer dollars on a "terrorism betting parlour."
At the time, the exact same structure was already being used inside Hewlett-Packard to let executives anonymously gamble on future revenue and sales – and to win by being correct. It was seen as a way of cutting through office politics and encouraging people to go against the grain if they really believed in it to formulate alternative sales predictions.
The effectiveness of crowd wisdom as a way of cutting through noisy variables to find real data, like in Hewlett Packard, or just a way of getting statistically impressive accurate predictions, like DARPA hoped, has been well established.
The system has received new life in the cryptocurrency age, with Augur being one of the first platforms to go live with its very own terrorism, sports, market and everything else betting parlour.
It's only been live for about a week, with users collectively and accurately, making sane, sensible and correct predictions around the World Cup and other sporting events, and laying into modest medium-term cryptocurrency price predictions especially heavily.
The good news
You can find the current bets by installing the app or by going here.
The main highlight so far is its Ethereum price prediction because it has more money behind it than any other bet. The prediction is still evolving quickly with more money arriving all the time and has another five months left before paying out.
Augur collectively reckons, with a fair degree of confidence, that Ethereum's price will exceed a modest $500 by the end of the year.
It's worth noting that it's above $500 right now, so the 23% currently betting on "no" are probably expecting something quite dire to happen to Ethereum. Either that or they're heavily invested in ETH and just want to hedge against a drop.
An equally sane prediction has emerged for the Augur REP token itself. It hit its all-time high at over $100 in January and is also currently at about $32. It's similarly hard to imagine it dropping too low without something bad happening to Augur itself.
The platform has also proved adept at sports betting, with a relatively high degree of accuracy and a decent amount at stake across many events. When there's a decent amount of money and enough people involved, the predictions seem to take themselves sufficiently seriously to be useful.
There's also the possibility of using the platform as intended – to use what you know (or think you know) to win bets.
The 45% yes vote for whether BAT will be listed on Coinbase seems quite low considering Coinbase has announced that it's currently looking at adding BAT.
The bad news
The bad, but unsurprising, news is that the vast majority of predictions never attract any interest or bets, with most interest instead being concentrated around a handful of predictions.
The other bad news is that the potential for manipulation and the use of Augur as just another advertising platform is already very apparent.
For example, there's a prediction that asks whether a specific YouTube video will hit 5,000 views by 6 August. This immediately jumps out as straightforward advertising, and even if it's not, it will still be relatively easy for anyone to swing the results to yes.
It gets especially easy to calculate exact anticipated return on investment with crypto bounty markets, some of which have basically turned into crypto multi-level marketing hubs where people can get paid with a coin for shilling the coin they're about to get paid with.
If you have a good idea how much you'll win by accurately predicting a certain amount of likes, shares or other social signals, and know how much it will cost to buy those signals, it could mean easy money as long as you can find someone to bet against it. That aforementioned YouTube video view prediction is standing at only 40% yes at the time of writing, with $260 at stake. It might be genuine, or it might be someone trying to artificially seed some interest on the no side for additional profits when they manipulate the results to yes.
The meta-bets, such as the one below, might be the epitome of those born-to-win predictions.
The weird news
There's no shortage of conflicting predictions, and it might be that people simply start a new bet if they don't like the way a current one is turning out or how it's phrased. For example, you'll get conflicting predictions depending on whether you bet on a specific president being "re-elected" in the next election, or "winning" the next election.
Some of the predictions also highlight the need for clearly defined language guidelines to avoid ambiguity and preloaded assumptions.
As in... are Tether prices being manipulated? Yes, it's a stablecoin and that's how it works. Or does it mean manipulating bitcoin prices? In that case, it's operating with the loaded assumption that Tether hasn't already been found to be statistically likely to be manipulating bitcoin prices.
More esoterically, one might also consider how different phrasings and options can influence outcomes, especially on a large scale with enough participants.
- It started as "Do you agree that Scotland should be an independent country? Yes/No," but that was scrapped due to the potential bias towards yes implied by the phrase "do you agree."
- It was then changed to "Should Scotland be an independent country? Yes/No."
Brexit phrasing went through a similar evolution.
- It started as "Should the United Kingdom remain a member of the European Union?" with Yes and No as the options. This was then contested as favouring remain, with arguments that people might have an inherent bias to "yesness," and that only the Yes option clearly highlighted the results of that vote, while the No side didn't really have a follow-up.
- With those objections in mind, it was rephrased as "Should the United Kingdom remain a member of the European Union or leave the European Union?" and voters were given two options: to "Remain a member of the European Union" or to "Leave the European Union" rather than yeses and nos.
Augur is an open market though, where people can pick and choose to bet on the questions with the phrasings they like, however they like.
In this case, the scuffles over phrasing might be among those trying to use it to settle a bet. But anyone who wants to produce as neutral a question as possible will probably have to think carefully when they do.
Disclosure: At the time of writing, the author holds ETH, IOTA, ICX, VET, XLM, BTC and NANO.