Bookies v opinion polls: who will get it right on the Referendum?

With only a week to go until the EU referendum, we are witnessing an interesting spectacle. No, it’s not the painfully underwhelming campaign on either side, although it is quite remarkable how low politicians can go in giving up intelligent, fact based arguments, unanimously resorting to emotional blackmailing. No, what appears to be at least as striking is the difference between the predictions of public opinion polls and betting agents.

Public opinion polls now invariably predict that Leave will win on 23 June. The average over all polls shows a four point lead for Leave (52-48).


Bookies on the other hand anticipate a conspicuously different outcome. Almost all the major online betting agents offer 1/2 odds for Remain (i.e. for every £1 you bet you win £0.5) as opposed to the 13/8 odds of Leave (i.e. for every £1 bet you win £1.625).


The rhetorical question is: if bookies read opinion polls, why do they think Remain is more likely and why are they offering higher payouts for betting Leave?

The main public opinion research companies are publicly listed private companies (or their subsidiaries). When they get it wrong their reputation suffers and potentially their stock value suffers. But even if they get things wrong, it is not so bad if everyone else gets it wrong. It is as if their reputation only suffers to the extent they get it worse than their competitors. On the other hand betting agents have a real skin in the game. If they get their odds wrong, they directly lose money – irrespective of what their competitors are doing.

The 23rd of June will answer an interesting question: who can we trust more, those who have less to lose from getting it wrong, or those who have a real skin in the game?


2 Responses to Bookies v opinion polls: who will get it right on the Referendum?

  1. I don’t think the two sets of figures are incompatible. Polls measure opinion at a point in time. Betting markets focus on a future point in time. In one-off referendums there is considerable volatility in people’s vote intention, and there’s a well documented tendency for people to revert to the status quo, even on the day of the election (see That, or something like it, seems to be the theory on which betting markets are operating.

    • orb75 says:

      Yes, good point Chris. My question: given what you said, what is the market value of a poll taken for example at t-23 days? More interestingly, if what you say is true, how come bookies recognise the limitation of a poll taken 2 weeks before the referendum (and do not adjust their odds) but markets don’t (and hence we’re witnessing a fall in GBP and stock markets)?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: