The 2016 Christmas market in Birmingham. The city’s benefitted from huge investment but too little of it has reached the rest of the Midlands

Theresa May is utterly humiliated, forced to rely on the Democratic Unionists for a majority. Jeremy Corbyn has exceeded all expectations. Far from losing ground, Labour made a net gain of 30 seats – not just in the north and Wales but in London, the south-east, south-west and Scotland. Corbyn’s critics have been silenced and his position as leader is now in no doubt. Theresa May, by contrast, is on borrowed time, as is her government. The prospect of another general election, either later this year or early in 2018, looms large.

I was on the doorstep in Bedford – one of the seats that Labour unexpectedly gained – on election day and it was clear that something was afoot. To my surprise, no doors were slammed in our faces: instead Labour voters were turning out in their droves, as were former Tory and UKIP voters and those who’d never voted before.

But it’s easy to forget, amidst all the jubilation on the centre-left, that Labour lost this election. Badly. Labour won 262 seats, nine fewer than Kinnock did in 1992, and only three more than Gordon Brown in 2010. Given that both 1992 and 2010 are remembered as catastrophic defeats for Labour, it’s difficult to see the 2017 result as a triumphant success.

Like most observers my predictions were way out: in a blogpost on the eve of the election I predicted a Tory majority of 55. But the gist of my argument – that Theresa May would do better than she deserved to, and that Labour had no chance of winning – still held true. Exceeding expectations is not the same as winning.

Yes, Labour did do very well in terms of vote share: at exactly 40%, much higher than that achieved by Brown in 2010, Miliband in 2015, and even Blair in 2005. The trouble is that the weakness of the Lib Dem recovery (and the utter collapse of UKIP) meant the return of two party politics, south of the border at least. With 85% of voters now opting for the two main parties, 40% is no longer enough to win an election.

Another problem was that Labour’s advance was very unevenly spread. Labour did outstandingly well in London, easily holding marginals like Ealing Central, Ilford N and Eltham, and winning more prosperous seats like Enfield Southgate, Battersea and even Kensington from the Tories (the only exception was north London seats with a high Jewish population such as Hendon and Finchley & Golders Green, which both stayed blue).

Estuarine seats to the east of London that Labour held until 2010 – Thurrock, Dartford and the three Medway seats – stayed Conservative with majorities of up to 10,000. But Labour made extraordinary advances in bigger southern towns and cities – particularly those with lots of students or ethnic minority voters. Labour gained Ipswich, Stroud, Bedford, Peterborough, Reading E, Portsmouth S, Brighton Kemptown and Canterbury, and came within 350 votes of taking out Amber Rudd in Hastings. In Bristol, until recently a Lib Dem stronghold, Labour’s advance was seismic. The previously marginal Bristol W seat now has a rock-solid Labour majority of 37,000, and Labour won the leafier Bristol NW seat next door by nearly 5,000 votes.

****

So how on earth did Labour win seats like Kensington and Canterbury but not become the largest party nationally in 2017?

Firstly, Labour only made a modest recovery in Scotland (gaining six seats from the SNP, compared to the Tories’ 12). It’s still 33 down on the 41 Scottish seats won by Gordon Brown in 2010. Just over the border, Labour failed to win back two former safe seats, Carlisle and Copeland.

And further south Labour’s advance was very patchy indeed. There’s been a lot of discussion over the last three days about the 35 seats Labour gained, but very little about the five seats that Labour lost. One of these was in the north-east (Middlesbrough South & East Cleveland) but the other four were all in the Midlands: as always, a region that helps decide the outcome of general elections.

Labour’s Midland results last week were mediocre at best: far from gaining ground, Labour made a net gain of zero seats. Labour gained Lincoln, Derby N, High Peak and Warwick & Leamington. But it lost Mansfield (held by Labour continuously since 1923), Stoke on Trent S and Derbyshire NE (both held by Labour continuously since 1935) and Walsall N. The last result is particularly poignant: the loser, the well-respected maverick David Winnick, was first elected in Croydon S as long ago as 1966 (until dissolution he was the only MP to have sat in the Commons in the 1960s).

Labour cemented its grip on the big Midland cities – Birmingham, Nottingham and Leicester – but failed to win enough seats outside them, including several it had only narrowly lost in 2010 and 2015: Corby, Redditch, Nuneaton, Telford (just a few miles from Corbyn’s childhood home), Dudley S, Northampton N, Warwickshire N and Sherwood, for example. In 1997 Labour won five out of the six constituencies in my home county of Northamptonshire; in 2017 all seven of them (an extra seat has been added by boundary changes in the meantime) stayed blue.

The irony is that Jeremy Corbyn himself is a Midlander. Although he was born in Wiltshire, the Corbyn family moved to Shropshire when Jeremy was seven. He was brought up in a hamlet called Pave Lane, near the Staffordshire border, and went to prep school and then grammar school in Newport, a small town nearby. He later became active in the Wrekin constituency Young Socialists and worked briefly as a reporter for the Newport and Market Drayton Advertiser, before spending two years doing Voluntary Service Overseas in Jamaica and then moving to London. If Corbyn – like Brixton-born John Major in 1992 – had made more of his Midland origins in the campaign maybe Labour would have fared better here.

Above all, Labour did not have a compelling story to tell in the Midlands. Though the Midlands are home to many important universities – Birmingham, Leicester, Nottingham, Warwick and Loughborough, among others – Labour’s message on tuition fees was muted here. Research by the Higher Education Policy Institute shows that of the 9 English constituencies with a student population of more than 25%, just one (Nottingham South) is in the Midlands.

Outside Birmingham, the Midlands have below-average levels of unemployment, but while jobs are plentiful many of them are in low-wage sectors like logistics, retail and call-centres, with manufacturing still in decline. Apart from Stratford-upon-Avon and the Peak District few tourists visit the region. There’s a lot of untapped potential in the Midlands but little co-ordination, few advocates and precious little vision. The disillusionment felt in Stoke-on-Trent has been widely reported thanks to the recent by-election there, but the malaise is felt much more widely.

Austerity has hit the region hard. Of the six NHS trusts with the worst crises last winter, two are in the Midlands. Ofsted says that the East Midlands has the worst schools in the country, while in 2015 and 2016 the West Midlands had the highest level of gun crime in England. Sandwiched between the economic hothouse of London and the Northern Powerhouse, the Midlands feel neglected: other than HS2, widely perceived as a boost for Birmingham rather than the region as a whole, little new infrastructure is planned.

East-west transport links remain very poor: where I live, in Northamptonshire, there are no direct public transport links at all to Cambridge, the boomtown thirty miles to the east. The promised rail link between Oxford and Cambridge, which would skirt the southern edge of the Midlands, has fallen years behind schedule, and it’s not certain the crucial section east of Bedford will ever be built at all. Meanwhile the electrification of the Midlands Mainline through Leicester, Derby and Nottingham has been repeatedly postponed.

Labour’s defeat in the West Midlands Metro Mayor contest in May should have been a wake-up call, but in the general election campaign Labour concentrated on London, Yorkshire, the north-west and the north-east, not the Midlands. Labour’s strategic decision to only pour resources into seats with Tory majorities of less than 2,000 meant that many crucial Midland marginals were ignored.

Labour fought a good campaign elsewhere but it needs to ask why it failed to make any advance at all in the Midlands. Labour now has 32 fewer seats in the Midlands than it won in 1997. Even if it had won just half these 32 Midland seats last Thursday, without any extra gains elsewhere, Jeremy Corbyn would probably now be in Downing Street as head of a Labour-Lib Dem-SNP coalition.

At the next election – whenever it may be – Labour may well struggle to win many more seats in Scotland, given the Tory revival and the underlying strength of the SNP there. In London, the north-west and Yorkshire most of the marginal seats it can realistically win are already in the bag. Labour won’t win by piling up even bigger majorities in its inner-city safe seats, but by winning marginal, unglamourous constituencies that don’t have universities, Waitrose supermarkets or Jamie’s Italian restaurants. Labour still has a mountain to climb. That mountain’s called the Midlands and Labour should get its crampons on now.