Skip to content

“So such markets were, and perhaps are, subject to bias from deep pocketed people who may be expressing preference more than actual expectation”


Geoff Buchan writes in with another theory about how prediction markets can go wrong:

I did want to mention one fascinating datum on Brexit: one UK bookmaker said they received about twice as many bets on leave as on remain, but the average bet on remain was *five* times what was bet on leave, meaning more than twice as much money was bet on remain.

Clearly weathier people, most likely pro remain, would be able to bet more, and I strongly suspect a similar bias exists in prediction markets, which, the last time I dabbled in them, had quite small open interest (think total money at stake). So such markets were, and perhaps are, subject to bias from deep pocketed people who may be expressing preference more than actual expectation.

Here we are in December (I wrote this in July but, y’know, bloglag) so youall have probably forgotten what Brexit even is. The general point still holds, though.


  1. hgfalling says:

    Are there people claiming that prediction markets aren’t subject to bias? I mean, the stock market is subject to bias, too, and it’s a lot more closely watched/followed/exploited for inefficiencies than prediction markets.

    The reason prediction markets are valuable is that they can reveal information that would otherwise remain hidden because they provide real incentives (monetary) and anonymity to the possessors of said information. The reason they empirically perform well compared to, say, pundits spewing hot air about the future is that the participants have skin in the game that with rewards for accuracy, the opposite of punditry. But it’s not a theorem that prediction markets are infallible.

  2. Garrett M says:

    This seems like a liquidity issue. If betting markets had the liquidity of stock markets, I think this sort of effect would be minimal.

  3. Z says:

    Can we also have some dog pictures?

  4. Slugger says:

    Do rich people actually make more or bigger bets than not-so-rich people as this article states? This information about the composition of the betting community should be available. My personal observations from standing in line at convience stores and watching people buy lottery tickets are that buyers do not skew toward the upper classes.
    The citation from Mr. Buchan has several assumptions. These include that rich people make larger bets than the less rich, that rich people favored remain over Brexit, and that the rich made their bets in order to influence opinion rather than win a bet. The longer the chain of assumptions, the likelier a weak link in that chain.

  5. Michael says:

    Isn’t that because the odds for remain were much lower? People tried to get to some threshold of payoff.

  6. anon says:

    As far as I know there are basically two theoretical mechanisms why prediction markets might work (predict well).

    1. Really intelligent and informed people who can predict really well will use the uninformed stupid noise traders to make a profit. In so doing, they drive the price to what they think is the real price. Because they can predict so well, the market ends up predicting so well.

    2. The market price is some sort of weighted average of the beliefs of all the traders and it’s this averaging of ‘models’ that makes the market predict well. In that context, having the weights be proportional to a trader’s wealth would be seen as making the market price more likely to predict well. Wealth correlates with being right.

  7. 27chaos says:

    Any explanation of why prediction markets sometimes fail also needs to be compatible with the reasons they sometimes succeed. If we’re seeing a bias in the markets due to the wealthy on issues like Trump and Brexit, why hasn’t that bias also showed itself on other questions?

    • Andrew says:


      The prediction markets did not fail in the U.S. general election. The probability of approx 70% for Clinton was in line with polls and expert opinion. According to news reports, both the Clinton campaign and the Trump campaign expected Clinton to win.

  8. Simon Gates says:

    > youall have probably forgotten what Brexit even is

    Unfortunately not. We’ll be living with that particular mess for some time yet.

  9. Dzhaughn says:

    If the idea about the bias is convincing, no one will succeed in replicating the results for long, because the prediction market will move to correct the arbitrage.

  10. Sports books are influenced by large bets, but only if they recognize the source as smart money. A large bet from one of the “sharps” will move the line even if a lot more public money is going the other way. Presumably, the sharps win more often than they lose, which is why they’re still in business and why the books pay attention to them. (Of course, there can be cat-and-mouse games, as when a sharp will put down a bet against the team he really likes, and then when the line moves, he makes a larger bet the other way at the more favorable line.)

Leave a Reply