How Augur v2 Tackles 3 Critical Trading Risks
Posted On March 25, 2020
Trading on Augur carries three risks, right now. Risks, here, meaning cases in which you could be right but still lose money. For example, if I buy YES shares in the Giants winning a football game, and they win, it’s possible I could still come out at a loss.
Augur v2 has the potential to address all three of these risks. In fact, each of the three main contract-level innovations of v2 attempts to tackle one of these risks.
Each of these risks corresponds to a different layer of decentralization:
Currency Risk is a result of decentralization at the market trading level and is addressed by using DAI rather than ETH for trading.
Capital Lockup Risk is a result of decentralization at the market resolution level and is addressed by shorter reporting windows, immediate disputes, and preemptive staking.
Invalid Risk is largely a result of decentralization at the market creation level and is addressed by improved Invalidity bonds and tradable Invalid shares.
Augur v1 uses Ether for trading. So whenever you hold a position in an Augur market, you are also holding Ether, a volatile currency.
So effectively, you have two positions: your long or short position on some real-world outcome and a long position on the currency (Ether) that you’re using to place the bet.
Even if your bet is right, you could still come out at a loss in dollar-denominated terms, if the price of Ether drops more than the value of your bet rises.
In Augur v2, trades take place in DAI, a dollar-denominated stablecoin, rather than ETH. The upside is potential to attract more liquidity from more bettors, not just those who want to be long ETH. Potential downsides, include technical risks around DAI and its dollar peg, possibly lower liquidity for attaining DAI, and the inability to get full ETH exposure while trading on Augur, for those who wish to be long ETH.
On net, I see DAI as advantageous, since it is the clearest path right now for Augur to minimize currency risk and start to reach beyond “crypto-natives.” However, in the event that DAI fails, we may need to find other ways to let users hedge currency risk e.g., by taking short positions in ETH when obtaining Augur shares.
Capital Lockup Risk
Market settlement and payouts on Augur v1 can be slow and the timing is unpredictable. A market may go undisputed and resolve in around a week or be disputed for weeks or even months before settling.
While slow market resolution does not result in direct loss of funds, it may offset gains and incur indirect losses via forfeiting the earning capacity of said funds while they are locked up.
Time Value of Money is the idea that money is worth more now than later, due to its ability to accrue interest and earn in other ways e.g., using it to place other trades on Augur. Put simply, $100 now is worth more than $100 in a year. Or phrased in reverse, there is an opportunity cost to locking up capital over a period of time.
To take an extreme example, if my Augur position gains 2% in value, but I must wait months to redeem its payout, that gain would be an effective loss, since I could have collected over 2% interest on those funds in that same duration.
In v2, markets will resolve faster due to shorter initial reporting and dispute windows, the ability to immediately dispute outcomes, and pre-staking. For more info on these improvements, see here.
Services like STLD built on top of Augur will also provide instant settlement for many markets, letting traders redeem winning shares for DAI at a minor fee, at the time that a market enters reporting or even as soon as the real-world outcome is known.
As a result of contract-level changes and instant settlement services, I expect capital lockup risk will be trivial on v2 for the vast majority of markets.
On Augur, a market resolves “Invalid” if its terms or outcome are ambiguous, subjective, or unverifiable. This may be due to an accident like sloppy or imprecise market terms, a deliberate scam on the part of the market creator, or some unpredictable real-world event (or breakdown in the resolution source) that renders the outcome in question unknowable at the time of market reporting.
In Augur v1, if a market resolve Invalid, all outcomes pay out equally. For example, a binary (YES/NO) market pays out .5 ETH for each YES share and .5 ETH for each NO share. If a trader spends more than .5 ETH on a share, they lose the difference, in the event of Invalid resolution.
In v2, Invalid is a discreet, tradable outcome, meaning traders are able to buy and sell shares that pay out if the market resolves Invalid. If the market resolves Invalid, each Invalid share pays out one DAI, while shares in other outcomes e.g., YES or NO, pay out nothing. If the market resolves to any outcome other than Invalid, then Invalid shares will pay out nothing.
This scheme is called “Tradable Invalid,” since traders will be able to buy and sell shares in Invalid just as they can buy and sell shares in other market outcomes.
Tradable Invalid, in theory, may help signal and hedge Invalid risk. I’m skeptical, however, that in practice, the current incentive structure and UI will be sufficient for Invalid insurance and risk signaling (a rabbit hole I plan to dive into in an upcoming piece).
v2 also has improved Invalidity bonds. When users create markets on Augur, they put up “Invalidity” bonds that they get back at the time of market resolution unless the market is Invalid. The bond rate is automatically calibrated over time based on the number of markets that resolve Invalid.
In v1, the rate fell too quickly at times, making it cheap to create Invalid markets. The formula is improved in v2 such that the bond can fall no more than 15% in a single week. This may help deter Invalid market creation while letting users filter markets by the bond rate the market creator paid, with the idea that markets with higher rates are less likely to be Invalid.
Finally, in v2, the market creator does not collect trading fees if the market resolves Invalid, further disincentivizing the creation of such markets.
Thanks for reading. To stay ahead with fresh insights on the future of prediction markets, join my weekly newsletter.
Thanks Micah Zoltu for reviewing a draft.