Friday, 16 April 2010

Behavioural politics, day 10 of 30

That was much more exciting than I expected.

Following the leaders' debate on Twitter added a hilarious dimension - a peanut gallery for the oh-so-serious talk going on on screen. The side-effect was that I hardly noticed what they were saying, and could pay attention to the proper subject of a behavioural analysis: how they were saying it.

On this basis, Clegg undoubtedly came out on top. There were lots of reasons for this:
  • Expectations. For some reason people thought Cameron would do best (even though he's not a very inspiring speaker at all), and so he was compared to the standard of expectations instead of reality. Brown was expected to be dull and earnest, and mostly was - though a few jokes at Cameron's expense will have delighted Labour partisans. And most people hadn't given much thought to Nick Clegg, so it was the easiest thing in the world for him to beat those non-existent expectations.
  • Mean reversion. In a comment a few days ago, Min pointed out that mean reversion (or regression to the mean) is not a law of nature, but often simply a statistical artefact. The point is, it applies when there is no reason to expect an underlying difference between the things being compared. In modern politics, which is based on appealing to the largest number of voters rather than fighting for heartfelt principles, there's every reason to expect that the parties will all be about equally appealing to voters. To the extent that there are differences (e.g. the Tories' huge lead over Labour last year) these mostly arise from temporary effects and cognitive biases, and the more new facts or impressions people are exposed to, the more the differences will disappear.

    This is why there are "iron rules" like the one David Cameron broke:

    ...he faced down critics in his party who warned that pressing for a television debate would break one of the iron rules of politics – that a frontrunner should never agree to hold an election debate with opponents.
  • Novelty. The main "real" effect hurting Labour is that people are tired of them. On the surface, this should favour their main opponents, the Tories, but there are plenty of voters who still remember them too, and without fondness. Therefore, if the Liberal Democrats appear as a viable alternative, and one without the handicap of having actually been in government within living memory, our desire for novelty will support them.
  • Clever positioning. Nick Clegg did exactly what he had to do, which was to stand outside of the fray. His physical positioning supported the brand positioning, and he allowed Brown and Cameron to argue across each other before stepping in and offering "the third way".
Most importantly, Clegg has established the Lib Dems as having a real chance of doing well, which (as discussed yesterday) is a self-fulfilling prophesy. Yesterday I gave some speculative numbers:
It would represent a genuine earthquake in British politics if we were to wake up the next day with 150 Lib Dem MPs, 250 Tories and 200 Labour. But it's not impossible.
And what is the projected outcome from today's (spectacular) poll results?
Latest ITV/Comres poll of Con36/LibDem35/Lab24 would give Con 289, Lab 186, LD 145 HOC seats try it
More soon on why a Lib Dem lead of 11 points over Labour translates into fewer MPs, and whether that's actually a likely outcome. But on today's showing, while Gordon Brown has missed an opportunity to establish himself as a "safe pair of hands", David Cameron is the top loser.

If there is no immediate backlash against the Lib Dems, this could turn out to be the most important day in British politics for many years.

Ratings: Lib Dems 9/10, Labour 5/10, Conservatives 3/10.

Update: The Comres poll numbers above are not a representative sample - the numbers are only from people who watched the debate. The Lib Dems have made some progress in the polls today but not that much. I expect the picture to become clearer today and tomorrow.


Min said...

"one of the iron rules of politics – that a frontrunner should never agree to hold an election debate with opponents."

I had always thought of that rule as having to do with the fact that the front runner has more to lose from a debate. Your tie-in to mean reversion is quite interesting. :) Assuming a random result of the debate, the front runner will usually be perceived as having lost it. Similarly, the laggard will often be perceived as having made up ground.

I have to qualify that last because debates among several candidates are often structured to show the also-rans in a bad light. For instance, in a six-way race for mayor of New York City which featured Lindsay (liberal) vs. Buckley (conservative), Lindsay, Buckley, and Beame (Democrat) were seated, while the lesser candidates stood behind them. At one point Buckley referred to the standing candidates as the "Peanut Gallery", a reference to children guests on the TV show, "Howdy Doody".

Min said...

The also-rans could have their moments, as well. I heard a story about a debate in a four way race among two veteran politicians, a young lawyer, and a welder who liked to run for office. The more eminent veteran spoke first and dumped all over the other veteran and the lawyer, but said nothing about the welder. Then the other veteran dumped all over the first two, and said nothing about the welder. Then the lawyer dumped all over the veteran politicians, and said nothing about the welder. Then the welder took the podium and said, "Let me say that I agree with everything my opponents have said."