🚨 New paper out!
We analyze arbitrage strategies on @Polymarket and show how traders extracted $40M over one year by exploiting inefficiencies.
With @lululixious , Vahid Ghafouri, Guillermo Suarez
1/🧵 Thread 👇
looks like this polymarket exploit could have been worse:
the compromised admin wallet wasn't just holding USDC and POL
it had `resolveManually()` rights on the UMA Adapter which bypasses the oracle and forces any market outcome on polymarket
the attacker could have:
- bought large positions in some markets
- flagged them for manual resolution
- waited out the 1 hour safety window
- then called resolveManually() to resolve the markets in favor of their positions
- cashed out their wins on all markets they resolved themselves
(for markets that were already in a manual resolution window, there wouldn't even be a safety window)
its scary because it took the polymarket team roughly over an hour to get this situation under control
in that time the exploiter could have caused serious damage to active markets (over $453m in open interest)
1/10 days ago I posted about how Kalshi and Polymarket are scaling sports betting under the prediction market label. That thread was about what's wrong. This one is about what should be built instead.
another good reason why people would want to copy trade other traders on prediction markets is due to their domain expertise.
let me explain:
many successful traders in prediction markets have specialized knowledge or deep insights into specific fields or events
> politics,
> finance,
>sports,
> tech ... the list goes on
this expertise allow them to make more accurate predictions.
by copy trading, the less experienced traders can tap into this expertise without needing to have deep knowledge in specific domains.
It's really that simple.