The House | Misunderstood, over interpreted, unrepresentative: How to read the local election results

5 min read23 min

If there is one thing most pollsters and analysts can agree on, it’s that the UK local election system is a bit of a mess. Some councils have all-outs, some go in thirds. Mayoral elections run on different cycles, and different parts of the country use different electoral systems and even have different electorates.

From an analysis perspective, it’s hard to think of a slipperier basis on which to take stock of the country as a whole. And yet, every May, this is what we are called to do. In the first 24 or 48 hours, the fog of poll descends, and political pronouncements are made on incomplete, unrepresentative or misinterpreted results. Given the Prime Minister’s difficult political situation, this is even more likely to be the case this year.

That being the case, there are a few points worth bearing in mind as we head into that chaotic post-election period. The idea is to guard against incomplete or premature arguments and to share what, from Persuasion’s extensive research, we already know that we know.

Remember, Reform started from a baseline of nothing

The relevant baseline for most of the elections happening this year is 2022. In 2022, Reform UK as a party barely existed (in England, it made a gain of precisely two seats). This is worth remembering when Reform starts clocking up lots of councillors early on in the night (areas more favourable to them report earlier on Friday morning). It’s almost certain they’ll gain thousands of seats as these wards, frozen in the mists of 2022, become updated to the politics of 2026.

The scale of Reform’s gains will be large but likely consistent with a party that has gone from a rounding error to a quarter of the national vote. This is far from irrelevant, of course, but it’s not new information – it’s there in the national polls, and it was there at last year’s locals. We learnt in 2025 that Reform’s post-2024 surge was real rather than a phantom of polling error.

For what it’s worth, if you really want to use councillor gains as a metric, respected local elections analyst Stephen Fisher judges that “Reform need more than +2270 net gains to really provide convincing evidence of the improvement that opinion polls suggest they have made since last year.”

Wait for the NEV

In time-honoured tradition, the best analysis will probably arrive too late – after a million terrible hot takes have already spawned. That said, it is worth waiting for it anyway. 

The national equivalent vote share (NEV), or projected national vote share (PNS), will probably come on the Saturday or Sunday after polling day. This is a piece of analysis, usually from the BBC and Sky, that works out national vote shares implicit in the results, adjusting for the fact that not every area is holding elections. It is essentially the only clean basis on which you can extrapolate the results of the May elections to a wider national picture.  

Using this, we’ll be able to more cleanly compare the directional movement of each party. The performance of Reform and of the Greens – two parties set to make gains mostly at the expense of Labour – will indicate whether their good national polling has legs. It’s also distinctly possible that, despite making significant net gains (and therefore being able to spin a narrative of a successful set of elections), they could underperform their polling. 

Another thing to watch out for here is the Restore number. While Rupert Lowe’s party picking up a handful of seats will tell us little about their national viability, if they can clock anything here, that would be a cause of significant concern for Farage and co.

To make things messier, sometimes PNS and NEV differ slightly – and only cover local elections, not devolved elections happening simultaneously in Scotland and Wales. They also obviously can’t predict national elections held in three years. But with all those caveats, they’re still by far the best metrics we have.

Labour losing seats to Reform does not equate to Labour losing votes to Reform

Finally, there’s the risk of a more classic ecological fallacy as results come in. That is, if a party loses in one area and another party gains in the same, the assumption is that the losing party has lost because their voters have switched to the winning one. This will be true in some areas, but not in others.

It’s especially worth remembering when Labour loses wards to Reform. Previous Persuasion work has shown that even in areas where the Greens or Liberal Democrats are in no position to win, Labour has primarily been shedding votes to progressive parties rather than directly losing votes to Reform. Likewise, turnout matters hugely here – it’s very likely a lot of Labour voters will simply stay at home, while Reform and Green voters will be more motivated to turn out. This will account for a lot of seat movement, rather than direct switching.

Likewise, while geographic variations – Red Walls, Blue Walls, and so on – do matter, it’s important not to overstate them. Right now, Labour is getting creamed pretty evenly right across the country, just with different parties benefitting, directly or otherwise. 

Steve Akehurst is the director of the polling and research organisation Persuasion UK