Election results 2017: 'Clearly not a good night for pollsters'
The polling industry had much to prove in 2017.
The 2015 election had resulted in a proverbial car crash for pollsters, with almost half the campaign polls registering narrow Labour leads when the outcome was a seven-point Conservative lead.
All of the companies reviewed their methodologies as a result.
Yet I have never seen such a degree of churn in campaign polls as in this election.
It is interesting where the change in the polls has occurred this time.
Nine polling companies who published eve-of-poll figures also polled at the start of the campaign.
Taking the averages of these first and then final polls, we find the Conservatives down from 46% to 44%; Labour up from 26% to 36%; the Lib Dems down from 11% to 8%; and UKIP from 8% to 4%.
- Election results live updates
- Seat-by-seat result forecasts
- What happens if no-one wins the election?
- Full UK results breakdown
The estimated vote share outcome puts the Conservatives at 44% and Labour at 41%.
The poll that came closest to this was Survation's final telephone poll, which suggested Conservatives at 41% and Labour at 40%, followed by Kantar's online poll which had Conservative at 43% and Labour at 38%.
The remaining polls gave Conservative leads of between 13 and 7 points.
This has clearly not been a good night for polling.
All the serious work that has gone into repairing the damage in 2015 has now to be started again.
In five of the last seven general elections, the polls have overstated Labour's share.
This time, as in 2010, they have understated it.
It seems that whether people voted Leave or Remain in 2016's European referendum played a significant part in whether they voted Conservative or Labour this time.
Did the 2017 campaign polls factor this sufficiently into the modelling of their data?
If younger voters came out in bigger numbers, were the polls equipped to capture this, when all experience for many years has shown this age group recording the lowest turnout?
Also, the variations in voting across Britain at this election appear much more complex than in previous elections and that poses real challenges to polls seeking to reflect the national picture.
We have seen great movements in Scotland; Labour taking Canterbury and Enfield Southgate but losing a core seat such as Mansfield and sustaining a big swing in Bolsover.
The future seems very challenging for the pollsters.
The main message that emerged from the painful experience of 2015 was that they were facing increasing problems in reaching fully representative samples of the electorate.
It does not seem unfair to suggest that they need to go back to the drawing board.
Is there some important missing ingredient that, once found, will restore their past record of successes?
Or is there a systemic problem about achieving accurate samples today that no amount of massaging or modelling of data is ever going to resolve?
It is important for all of us that the pollsters succeed.
If their reputation collapses then a space is created that will be filled by any number of spinners and fraudsters.
Politics, like nature, abhors a vacuum and if polls are discredited then any number of people will step in to give their partial and partisan interpretation of events.
National samples will be elbowed aside by the wisdom of tweets and Facebook friends.
I hesitate to call for inquiries - after all, we had one post-2015.
But I think post-2017 we have to allow more time for proper scrutiny of the various methodologies.
In 2015 there was great pressure for a quick answer: 2017 suggests there is no such thing.