It's good to share

Trading the election BannerThe 2015 was the closest and most interesting election for a generation. Thanks to the fixed term parliament act it was also the first time the campaign stretched into months rather than the few short weeks normal to the UK system.

During the 6 exciting months, I have traded the opinion polls, taking positions on each of the main 5 parties to try to follow every manifesto launch, panicky policy announcement and bacon related gaffe along the way. This page summarises the results, my portfolio’s return and the lessons learned.

You can follow each weekly update here and find the original rules of the game here. The polling data is from the ukpollingreport website.

Hunting “alpha”

So how did I do? In short, pretty well overall as you can see below. My eventual return was nearly 27% over a 20 week period – impressive by any measure, particularly in a “market” of pretty low volatility.

Trading return by week


Poll of Polls Party by party FINALMy long, short position on the Tories and Labour eventually paid off handsomely once the ballot boxes closed. The only question, the same one vexing the polling industry as a whole, is why it took the actual election for it to become clear the Tories were the clear winners? The Tories managed to achieve a greater share of the vote than they polled throughout 2015 and Labour the lowest – by some margin.

Poll of Polls FINAL


Its not all good news though. I called Ukip wrong at the start of the campaign, believing they could poll at about 15-16%, far higher than they ended at. Although its hard to determine precisely why this is, it may have a lot to do with a kind of “return to seriousness” when people actually go to the polls – a scenario that hurt Labour too.

The Lib Dems also confounded my expectations at every turn. I started short, without any idea why they existed as a party, only for them to poll better than expected week after week. I then went long only to find their vote collapse at the polls into the irrelevance I thought they were all along.

Finally the Greens did not surge in any way, in fact they collapsed. So although I made a decent return, I was wrong as much as I was right. The lesson in this – be right about the big things and reserve your mistakes for the little things.

What did we learn?

1. Its better to help people succeed than protect them from failure. So why did the Tories win and confound expectations? There are several theories of course, you can read mine here. Above all I think that Jim Messina, architect of Obama’s 2008 victory called it – optimism wins the day, and the pollsters saw it too but overlooked it.

2. It’s still the economy, stupid. Another thing the pollsters called and ignored was the strength of the Tories lead on the economy. Cameron and Osborne were consistently ahead on leadership and economic competence. In the end it showed.

3. The Russell Brand effect.  Poor Russell Brand. Firstly he was anti-politics and anti-voting. Then he was wooed by Ed Miliband and Brand endorsed him just in time for his electoral collapse. Finally, following the election he “withdrew from politics” again, his sole accomplishment the endorsement of a failed candidate. Or was it his only accomplishment?

Under the hood of the polling data there is a neat and ignored little nugget. In terms of percentage vote share the polls got it wrong. In terms of raw votes however, they were close on the Tories, but way off on Labour. The turnout was only 66% compared with a projected 80%.

The missing voters were overwhelmingly Labour. Well done Mr Brand for pointing out an important lesson, people must vote.

4. No narrative, no votes. The Lib Dems have concretely proved that no-one votes for “the coalition” party. Next time a politician gives the “we’re campaigning for a majority” response to questions of coalition negotiations, you’ll know why.

5. Perception matters. Ed Miliband was never Prime Ministerial material, most people could see that. In future every candidate should go through a “Putin test”. If you don’t believe he could stand up to a negotiation with Vlad, he shouldn’t be on the ticket.

So there you have it, the lessons of the campaign from a polling point of view. A fun little exercise and one I’ll carry on with next time around. Let me know our feedback or ideas for next time.