This was not quite the election or result I expected: The decisive return to two party politics; that young people really did turn out in large numbers; that the once formidable Tory election machine could prove so incompetent; that Jeremy Corbyn could do so much better than any of his leadership rivals could have done; that Arlene Foster's DUP could suddenly become hugely powerful.

When the facts change we have to change our underlying models – and this is one of those moments.

Once again, the resisters have made the headlines; a large part of the electorate unwilling to go along with the script of the establishment powers that be. In doing so, they have upset conventional wisdom. 

Over the last couple of years, it looked like we were heading to a period of one-party rule in both London and Edinburgh. Instead, the election signals a return to vigorous competition, which must be healthy in the long-run even if it gives us confusion in the short-term.

Politicians will surely have learned a big dose of humility – and will be much more wary of being too clever by half in trying to manipulate the electorate

The experts were almost uniformly wrong. The main print media – whose owners lined up strongly behind Theresa May - were simply ignored by many of their readers. And we’re reminded that winning and losing are quite odd concepts in politics – parties can win the votes, and the seats, but still be seen to have lost.

Once again we saw both the vigour of democracy – particularly in the fantastic turnout from young people – and the contradictions and flaws of elections that aren't always the best expression of a society's collective intelligence. One negative aspect of the election was that it was even further from being an evidence or fact based election than other recent ones.  

Main parties were much less concerned to prove that their sums add up

It was also an oddly contradictory result. What sometimes looked like a kamikaze version of Conservatism – able to snatch defeat from the jaws of victory – nevertheless won a huge share of the vote. Meanwhile, the kumbaya-ish Labour party – sometimes appearing to believe that no problem couldn’t be solved with a bit more spending and goodwill – is in ebullient mood yet essentially lost a third election in a row, and was, paradoxically, partly boosted by voters’ expectations that there was only a slim chance of Jeremy Corbyn becoming Prime Minister.  

Both are good reasons why we will continue to promote democratic innovations that help to improve the quality of deliberations and arguments, and allow voters to make more nuanced inputs than is allowed by an occasional cross on a ballot paper.

I predicted to many people earlier in the year that Theresa May would probably switch from being uniquely popular and powerful to becoming uniquely unpopular and weak – but thought it would happen after the election not during it.  

In other countries, too, we’re seeing extraordinary fluidity – with Macron’s likely landslide in the parliamentary elections just the latest example. These are all reasons to be wary of overinterpreting the result. The next election - which could come soon - might deliver a very different result.

My biggest anxiety of the last year was that the UK government leadership would be inflexible and unimaginative in its handling of Brexit 

I was against Brexit, but at least hoped that smart enough politicians could mitigate the worst risks.  

A common criticism of Theresa May was that she lacked the fluency and agility that’s so vital to navigate complex issues. The recent manoeuvring, unfortunately, has done far more to unite Europe against us than to lay the ground for a good deal.

The really big question now is whether we end up with a leadership that is smart enough, subtle enough and agile enough to give us a Brexit result that really is in the UK national interest, and appropriately reflects public opinion that remains roughly split down the middle.

Consistency is sometimes a virtue in leaders. But we also need leaders who, when the facts change, change their models and approach. Today, I have no idea whether we'll get them.