As Augur and other prediction markets take off, we may see some big arbitrage opportunities.
What is Arbitrage?
Arbitrage is when you buy or sell a financial asset like a stock or cryptocurrency on multiple exchanges at the same time in order to capture price differences. The type of arb we’ll look at here is taking both sides of a trade on different prediction markets.
For instance, let’s say that Prediction Market A prices Dwayne Johnson as having a 73% chance of being elected president in 2024 while Prediction Market B forecasts a 76% chance. You could buy shares in Prediction Market A for ~73 cents on the dollar while selling shares in Prediction market B for ~ 24 cents on the dollar (100 minus 76).
If Dwayne wins, you’ll get a dollar for each share you bought in Prediction Market A, and 0 for each share you sold in Prediction Market B. This means you’ll gain 27 cents for each share in Prediction Market A but lose only 24 cents for each share in Prediction Market B.
Meanwhile, if Dwayne loses to Kanye or AOC, you’ll gain 76 cents for every share you sold in Prediction Market B while only losing 73 cents for each share you bought in Prediction Market A. Either way, you come out 3 cents ahead for every pair of shares you bought i.e., a 3% gain.
Here’s a breakdown of the P/L (profit/loss) for each possible outcome:
Why Do Arbitrage Opportunities Exist?
There are many reasons why arb opportunities may exist in prediction markets…
Varying types of users between prediction markets with differing biases or levels of knowledge.
2. Markets responding to and pricing in news and other developments at a varying speeds. For example, due to gas prices and low liquidity, Augur may lag behind a market like PredictIt.
3. Differences in betting limits, fees and other traits between platforms
4. Large, arbitrary buy or sell orders clearing out the order book on a prediction market, especially if the market has low liquidity.
5. Lag in market creation date. If Market A was created a month ago it may already be accurately (or at least, maturely) priced. If Market B comes along a month later it may take a few minutes (or hours or days) to play catch up.
6. Minor differences in the framing of the question or structure of the markets spurring different prices due to cognitive bias.
7. Discrepancies in broader market offerings between different platforms creating different hedging impulses. For example, imagine on PredictIt there was a market on whether Dwayne Johnson will run for president in ’24 and a separate market on whether he will actually be elected, while Augur only had the latter outcome to trade on. Traders on PredictIt betting that Dwayne would not run might hedge their bet by buying YES shares in the win-the-election market. This may lead to PredictIt pricing his odds of being elected higher than on Augur.
Proceed With Caution…
When considering an arbitrage trade, a few things to keep in mind:
If you’re posting orders rather than filling existing bids or asks, there’s a chance your order may go through on one market but not the other. The safest approach is to only fill existing orders. Alternatively, you can post a buy or sell order on one market and if and only if it gets filled, you can then fill an existing order on another market for the other side of the trade.
On Augur, gas and trading fees may offset an arb spread.
Make sure to factor in the risks of holding Eth (necessary to trade on Augur at this point) and early platform risks i.e., Augur is still young and potentially susceptible to hacks.
Keep in mind the risk of invalid market resolution on Augur. Successful prediction arb requires taking both sides of a trade, but if a market ends up resolving “Invalid” rather than YES or NO, then you’re effectively only taking one side of the trade! Make sure the Augur market, including the details section, is framed unambiguously and has a reliable, verifiable resolution source.
This is a less likely but still possible flavor of arbitrage. On Augur, anyone can create markets for relatively low fees. So you end up with some redundant markets betting on the same or similar outcomes. You also end up with some markets that are subsets of others, e.g., will Bitcoin exceed 20k at the end of 2018 vs. will Bitcoin exceed 20k at any point before the end of 2018. These types of redundancy create the possibility for pricing gaps between different markets on Augur.
Thanks for reading. To stay ahead with fresh insights on the future of prediction markets, join The Augur Edge.