> Are there any concrete examples of these markets being accurate There are prediction markets for US elections, which tend to be pretty good. In general you s

> Are there any concrete examples of these markets being accurate There are prediction markets for US elections, which tend to be pretty good. In general you should only expect markets with high liquidity to be very hard to beat, and the US elections have relatively low liquidity, so IIRC Nate Silver still outperforms them (but only by a small amount). The reason you should expect high-liquidity markets to be hard to beat is that, if they were easy to beat, you could easily get rich by betting on them. So the common-sense idea of "there's no easy way to get rich" implies that they cannot be easy to beat. The main exception to this principle is when there's a huge amount of money that intentionally places "bad" bets on the market, usually because they want to hedge against something. (e.g. I don't want this to happen, so I'll bet that it will happen, that way at least if it does happen I'll make some money.) This is why the TIPS spread, which can be interpreted as a prediction market for US inflation, consistently over-predicts inflation. Entities that stand to lose from high inflation use TIPS to hedge, which causes the TIPS spread to be too high. This could be corrected with better market design, but the treasury hasn't done that yet.