Another problem is, people may actually want to bet on random outcomes, because of money laundering or simply because this is how gambling essentially works. That huge account could be an insider or a billionaire with a few hundreds k to burn. Or maybe they want to orient people’s opinions towards a certain outcome.
Claiming that price movement in a prediction market reveals some amount of truth implicitly assumes that:
- people bet on something they believe to be true, and not to sway other people’s opinions or simply to burn money,
- people bet on something they believe to be true because they have specific private information (e.g. I bet on the Red Sox not because I think they’re good but because I know things other don’t about their opponents, their physical conditions and so on).
- their belief is actually correct (eg if I’m in the CIA and I know that the Soviets are about to launch a nuclear missile I can bet on it… but I don’t know that an officer down the line will refuse to do that).
Even if this was true, there is an issue of timing and consequences. Example: imagine it’s 2011 and some CIA or DoD officer makes huge, sudden bets on the fact that Bin Laden will be caught. Some AQ people get wind of this and move Bin Laden somewhere else. Congrats, your price movement signaled non public information to the market!
Another issue is that these bets tend to rely on public sources, news reports and so on. A journalist in Israel was threatened to change his news reports so that certain people didn’t have to lose on a prediction market. This could become more and more common, and with the advent of AI generated pictures who are you going to believe? Are you losing money because you bet on the wrong outcome or simply because someone with enough resources ensured that your outcome was never going to be reported?
