The first big event on Thursday night is the close of poll at 22:00. At this instant, election day reporting restrictions are lifted and the results of the big exit poll are announced. For the last few elections, and again in 2019, the broadcasters have prioritised getting the estimated number of seats right and devoted a huge amount of polling resources, expert advice and sophisticated statistical analysis to that end. The efforts have paid off with some very accurate predictions in the last few elections. It is unwise to mock the exit poll – its surprise prediction of a hung parliament in 2017 was correct, and in 2015 the late Paddy Ashdown promised to eat his hat if the Lib Dems did as badly as the exit poll said. They did, and good sport that Ashdown was, he did. But still, the exit poll is not perfect. That 2015 exit poll significantly understated the Conservatives’ eventual victory – it predicted a hung parliament with the Tories a little short of an overall majority but when the votes were counted they had won a small outright majority. Similarly in 2019, there is a grey area around a central prediction of, say, 321 Conservative seats – the range of normal-sized error covers both a Conservative overall majority and a hung parliament with the possibility of a Labour-led government. Pity the broadcasters trying to summarise that scenario.

The Press Association compiles a very useful list of the times at which declarations are expected from all 650 constituencies. All the media use these for planning their coverage on election night but as the PA points out this timetable can only ever be indicative. The timings are provided by the local authorities who conduct the count and are based on past experience and staffing levels available on the night, but they announce when the final count has been signed off rather than waiting until the time they give to PA. Some of these timings look a bit cautious; the count is often concluded quickly in the Wandsworth constituencies and those three (including marginals Battersea and Putney) may well come in much earlier than 02:00.

Sometimes the counting hits a snag and the announcement must be put off; when the result is very close the candidates have the right to request a recount. A check of the voting bundles usually suffices if there is a majority into the hundreds, but a full ballot-by-ballot recount seriously delays the announcement. In the closest elections, like Kensington in 2017 and Harlow in 2005, the final result is announced on the Saturday. The chances are that one seat will be in this limbo for a day or two, but there is no way of telling which seat it will be.

The first results are rarely exciting. For several elections, Sunderland were the undisputed masters of the quick count but their sworn rivals at Newcastle have got in on the act. It is an opportunity for a bit of publicity and perhaps to burnish an image as an efficient council. In 2017 Newcastle Central beat Houghton & Sunderland South to the punch. There is only a limited amount that we can conclude from these very early results, due around 23:00 or shortly after. Neither is very typical, and if the North East as a whole is diverging from the national picture the swing may not be a reliable indicator. But in 2017 they did suggest what was happening. In Newcastle Central there was a 2.1 per cent swing to Labour, reflecting the movement in core cities and places with young, educated populations. Traditional working-class Houghton & Sunderland South went the other way – a 3.4 per cent swing to the Conservatives. Watch both the size of the swing and the divergence between the two seats, but be careful in concluding too much.

Broadcasters now have another hour and a bit to fill, and the continuing uncertainty means that politicians, unless they are retiring and demob-happy, will try very hard to say nothing of interest.

At 01:00 things start getting more interesting. The first results from the much-discussed ‘Red Wall’ (a term whose prominence I regret but with which we seem to be stuck) will start to come through. The first couple will probably be Makerfield and Wigan, very safe Labour seats despite their Leave vote in 2016 – Wigan has been Labour since 1918 and Lisa Nandy’s majority in 2017 was 16,000 (34 per cent); Makerfield and its predecessor Ince have had an uninterrupted Labour history since 1906 but are slightly less monolithic (29 per cent majority). YouGov’s MRP model predicts the Labour lead in 2019 as 17 per cent in each, a meaty swing of 6-9 per cent to the Conservatives. If either seat is at all close, it promises cataclysmic defeat for Labour in some former strongholds later on in the evening, starting with the west Cumbria seat of Workington. The affluent Northern Ireland seat of North Down, where Lady Sylvia Hermon stands down, will probably be a DUP gain but Alliance are competitive. Rutherglen & Hamilton West in south Lanarkshire was a surprise Labour gain from the SNP in 2017 (although Rutherglen had been a Labour seat from 1964 to 2015). If the SNP fail to take it back, they will be disappointed.

Darlington (01:30) has a marginal tradition – it was Conservative in 1983 and 1987. A Conservative gain would signal their best result since those two triumphs.

The results really start flowing in as we approach 02:00. This batch will really give us a good idea of what is going on, although if it is close between a Conservative majority and a hung parliament we probably can’t say yet which side of the line we will land on. Labour defend marginal Battersea (if it’s close or a surprise Conservative gain, the YouGov MRP was too optimistic for Labour) and try to take Putney from the Tories. Most of the potential traffic is in the other direction, however – Leigh, Labour since 1922 and requiring a pro-Tory swing of over 10 per cent, was certainly regarded as in play by both sides during the campaign. YG-MRP had Labour a slim 4 points ahead at the end. Bury council is another quick counter, and its two marginals Bury North and Bury South declare around now. Labour gained North in 2017 – will they hold on? South has been Labour since 1997 for former MP Ivan Lewis who is standing as an independent but has urged a vote for the Tories. Bury South, too, seems to be going sour on Labour (and politics in general). West Bromwich East and the more decisively named West Bromwich West declare – both had large Labour majorities in recent elections but have been highly competitive. Wrexham also reports in. Three-way Scottish marginal Lanark & Hamilton East is a potential Conservative gain from SNP.

If you’re still up at 05:00 you’re probably a happy partisan

You will not be able to keep up with the results coming in during the peak period between 02:00 and 04:00. But unless this is a very unusual election, the broad trends should be visible not too long after 02:00 and from here on it is about illustrating those trends and noting the exceptions – and also, if it is close, seeing if there is going to be a majority or not. There will be a large number of shaky Labour defences coming through from the likes of Birmingham Northfield, the Wolverhampton seats, Bolton North East, Great Grimsby, Blackpool South and Hyndburn. It’s worth seeing how much resistance Labour are putting up to a concerted Conservative assault. Results from the Conservative-Lib Dem battlefront are also coming in at around 03:00 – the central London tactical dilemmas of Chelsea & Fulham and Chuka Umunna’s target of Cities of London & Westminster (known for short as CLAW to this election analyst, at least) and other targets such as Eastbourne (for the Tories) and Cambridgeshire South and Wokingham for the Lib Dems. Wokingham sees a rare contest between two incumbent MPs, John Redwood and Phillip Lee. Stirling is a key Scottish Conservative defence against any SNP resurgence. The big Northern Ireland contest, between Nigel Dodds of the DUP and John Finucane of Sinn Fein in Belfast North will also declare. About now we will hear from Jeremy Corbyn’s count at Islington North and Jo Swinson’s at Dunbartonshire East. Corbyn is in no danger, but Swinson might be.

Things quieten down after 04:00. We will hear from Bassetlaw, a former mining seat where a bungled selection process has worsened Labour’s already substantial problems, and a few more marginals. If the Conservatives are winning well, look at Rother Valley, Erith and Thamesmead and see if Labour holds on against the tide in Canterbury. If the Conservatives are falling seriously short, look for a Labour hold in Crewe & Nantwich and a gain in Harrow East.

City of Chester declares at around 04:30 but you will probably be watching Boris Johnson’s count at Uxbridge & South Ruislip .

If you’re still up at 05:00 you’re probably a happy partisan. If the Tories are doing well, it’s worth their while to stay up to watch Ashfield, the Stoke-on-Trent Central and North seats and Bolsover turn blue. If it’s gone the other way, Labour people will want to see if they get any joy out of Milton Keynes North or Hendon. However disappointed the Lib Dem viewer might be, it’s probably worth seeing the dawn and the result in Richmond Park.