Bottom line is: We are interested in forecasting. Prediction markets like Polymarket and Kalshi are one of the best methods of forecasting. They tend to be better than complex statistical models, polls or sports betting. A person may have specific information or insight, he uses that information to rattle the prediction closer to the truth. Just like other financial markets and commodity markets where the aggregate of people’s bets decide the price of the commodity. The price aggregates people’s estimates of that probability. Prediction markets share lot of similarities with traditional markets. They are easier to participate and people have opinions. You might be already investing or want to try your luck starting small (like me). I would watch out for these booby traps. Total bro science here:
THIS IS NOT INVESTMENT ADVICE
Look out for thick markets with enough variance. Don’t invest in thin markets. Thin markets: less people, small bets.
Try to gather information off-chain. If you have a secret or local knowledge that is not accessible by others or accessible to a few people. For example, conduct a survey to gain people’s estimate about what everyone else’s prediction would be. (This is better than asking their own prediction, for reasons I’ll leave up to your imagination.)
Check whether there is an organic incentive backing the prediction where optimistic people with fruitful outcomes are participating. Not just sharks. Again, think thick markets.
Never invest in a bubble.
There could be markets created in order to produce information. These are often thin with big players participating as underdogs creating an illusion of leaking information. Observers think that others seem to know something, and follow them – a false wisdom of crowds.
Predicting the outcome of a prediction, aka second order prediction, is seen as reputation building because you are betting on the accuracy of someone else’s bet. I believe this is not true in the context of price-driven markets. Second-order predictions can behave like synthetic CDOs if the market is a bubble itself, or more prominently, when the market is created to incentivize prediction.
