Here is a long overdue update of my UK opinion poll tracker and in this post, I going to take a closer look at the key trends over the last 12 months.
In a week’s time, Americans will go to the polls for what is known as the mid-term elections. Inevitably, the results will lead to much speculation on what it means for Donald Trump’s chances of re-election in 2020. However, I will be surprised if many commentators will look to history as a guide to 2020 and so I will fill in this gap with the help of fun 10 question quiz about US presidents.
As I write this, a plethora of economic forecasts are making the rounds in the news in the UK. In all cases, the forecasters have failed to publish their track record and these days, I will not pay attention to what they say unless their forecasts are accompanied by a track record. But, how does one go about presenting a forecasting track record to prove that one has forecasting skill? To demonstrate, I will analyse how well opinion polls have predicted General Elections in the UK and measure their track record. I must confess I was surprised at what I found out and I would urge all opinion pollsters to take note of my results.
For my latest Opinion Poll Tracker, I will explore 4 points currently being debated by political pundits.
- Have the Conservatives taken the lead in the polls?
- How much weight should we give to Survation polls who were the most accurate in 2017 and are the outlier poll today?
- Was there a “Youthquake” in the 2017 general election?
- What would the effect be of allowing 16 & 17 year olds to vote in general elections?
The last 3 general elections have seen some significant polling errors. In 2010, the Lib Dems were significantly overestimated, in 2015 the Conservatives were underestimated and last year saw the largest ever underestimate in the Labour vote. Whilst these errors suggest that the polling industry is struggling with general elections these days, a natural question to ask is “are all pollsters equally bad or are some better than others?”
Exactly one month ago, the UK woke up to the news that they had elected a hung parliament for the second time in 3 elections. For many forecasters including myself, this came as a surprise as I had been predicting a Conservative majority of 100 seats. In the event, the largest ever polling underestimate of the Labour vote was enough to see the Conservatives lose their majority.
At the beginning of my commentary on election night itself, I defined success for my forecasts as being how close the number of Conservative seats was to my forecast of 375. I also stated that if the number of seats was in the 340s I would consider this to be a prediction error. The final outcome was 317 seats so clearly that is a major prediction error.
Ahead of the 2017 general election, I predicted that the opinion polls would be wrong again and that the Conservatives lead over Labour would be underestimated by 2.6%. I based this on data provided by Mark Pack who has systematically recorded every opinion poll published since 1945. In the event, I was right that the polls would be wrong but instead of an error favouring the Conservatives, the polls recorded the largest ever underestimate of the Labour vote. As a result, election forecasters were blindsided yet again and the result was a hung parliament which few saw coming.
Friday 07:35 – I’ve just woken up to the result that we have a hung parliament which is miles away from my prediction. Clearly I need to conduct a post-mortem of my model but I did state in my first chart of this post (see bottom of post) that a hung parliament would occur if the CON-LAB lead was under 3%. Sure enough that is the outcome with the BBC currently saying the CON-LAB lead is 2.3%. What it also means that we have another epic polling error as the average of all polls in the week before the election showed a CON-LAB lead of 7.5%. Unlike 2015 when all pollsters got it, congratulations must go to Survation who called it spot on.
I am sure we will have a wave of people claiming to have predicted this error. As far as I am concerned a valid prediction is only one made in public in advance. Such a prediction would also need to explain why they expected a polling error on the scale shown in the chart.
The last time such an error occurred was 1983 and the error is on the scale of 1951! In fact the error is even more remarkable if you look at the 3 main parties.
Another polling post mortem beckons but this election will go down in history.
My official prediction using my Final Election Model is that the Conservatives will make a net gain of 45 seats resulting in a working majority of 105 seats.
My forecast uses data from my latest UK Opinion Poll Tracker and it is worth reading that post in conjunction with this post. At the bottom of this post is a spreadsheet containing my prediction for each seat. I am basing all figures in this forecast on the assumption that Conservatives will have a 9.5% lead over Labour on June 8th. I arrive at that figure by taking the current CON-LAB lead of 7% in the latest polls and adding an expected 2.5% underestimate in the Conservative lead over Labour based on my analysis of historical polling errors. A knock-on effect of this assumption is that I expect turnout to be 2pts higher at 68%.
Unlike the 2015 general election when the polls were essentially static (& wrong) throughout the election, the 2017 general election has seen some of the most extraordinary volatility in the polls that I can remember. If you are a Conservative supporter, the narrowing lead over Labour must be leading to anxiety and changed underwear. If you are a Labour supporter, you are probably starting to dream “can we? will we?!” It doesn’t help that your state of mind will depend on which poll you are reading and your memories of the pollsters’ failure in 2015 so how can you make sense of what is going on. I will show you how in 5 steps and to heighten the drama, I will leave the punchline to the end!