Are bookies having a change of heart about the Scottish independence referendum?

‘I’ll give yer 6/1 on a Yes victory round the back.’ Phil Burns, CC BY-SA

Debate in the referendum continues to rage over to what extent we can trust the opinion polls. While the Yes side have most recently being putting their faith in the circa one million undecided voters who will supposedly swing the outcome their way, one group that arguably deserves more attention as we approach September 18 is the betting industry.

Bookmakers decide odds through a combination of demand from customers and by monitoring a wide range of communication channels, such as newspaper reports and polls, to produce a daily running analysis of where sentiment is heading.

What follows is an analysis of what 23 leading bookmakers have been telling us about the outcome over the past five months (April 1 to August 31). It should provide some sobering reality to those Yes supporters, while also offering them hope for the weeks ahead.

Slide in August

The odds of the major bookmakers have tended to be fairly consistent with each other. Yet they are starting to diverge, with August 31 showing the lowest at 4 in decimal odds for Coral (meaning about 3/1) and Betfred and the highest at 5.46 (around 9/2) for Betfair. There’s also been a general trend downwards for the No vote in the past couple of weeks, which possibly reflects the recent opinion polls.

Average odds for Yes vote (Apr to Aug 2014)

Bill Buchanan, Author provided

The two TV debates between Alistair Darling and Alex Salmond both had a marked effect on the betting odds. Before the first debate, which was generally seen as a Darling victory, the odds for a Yes vote had been narrowing. After it, they rose steeply up.

In the second part of the month, the Yes odds started to come back in. After another lurch upwards just before the second TV debate, the Yes odds came in sharply on the back of what was seen as a victory for Salmond.

Odds for Yes vote (Aug 13 to Sept 1)

Bill Buchanan

Time to bet the farm?

By August 31 the odds for a No vote were 1/5, and approaching 3/1 for Yes. This is quite a change from a couple of weeks ago. On August 11 some bookmakers were quoting a No vote at 1/10, which is the kind of betting you would get for Manchester City against a team about seven leagues below – let’s say Tiverton Town.

In other words, the vote seemed certain for a No at that point. The odds indicated that the bookmakers wanted to limit the number of punters putting money on a one-horse race. But now the odds suggest that the bookmakers are less sure, and can see benefits in reducing their No vote odds and taking on bets.

The Yes vote has gone in the opposite direction. It drifted out to about 11/2 after the first debate, much wider than the current price. At 1/5 for No and 3/1 for Yes, the football analogy is now more like Manchester City v Luton Town – two teams separated by three leagues. If you can bear to switch sports for a moment, the bookmakers are highlighting that it’s now back as a two-horse race.

Know what ah mean? Neil Dorgan, CC BY-SA

If this trend continues for the next two weeks, the Yes vote will sit at 2.47. This is not much higher than evens. At that point, the No campaign has a real game on its hands.

Keep watching the betting market over the next couple of weeks, as they are more likely to get it right than anyone else. The next few days should show if this trend continues. And keep a close eye on whether the Yes odds narrow beyond the 3/1 barrier or settle at that figure. As betting experts will know, that 3/1 level is often seen as decisive in determining whether a two-horse race is too close to call or has a clear favourite – I’ll update the latest odds daily at this link.

As things stand, the bookmakers obviously still strongly favour No, but the speed at which the odds are changing will make the betting prices essential viewing between now and September 18.

We produce knowledge-based, ethical journalism. Please donate and help us thrive. Tax deductible.