What the evidence shows

The Greens won more than twice as many votes in Holborn and St Pancras in October 2026 as they did at the 2024 general election. Yet their share of the vote more than tripled. How can both be true? Because a vote share measures a party's slice of all valid votes cast. The Greens received more votes, and the total number of valid votes was smaller. Their slice therefore grew faster than their vote count. Labour still won the seat.

  • Green votes rose from 4,030 in July 2024 to 8,928 in October 2026.
  • The total of valid votes for all candidates fell from 38,602 to 27,211.
  • The Greens went from about 10 in every 100 valid votes to about 33 in every 100.
  • These results do not tell us which people changed parties or why.

What happened at the election

On 8 October 2026, voters in Holborn and St Pancras chose their MP in a by-election. Labour's Sagal Abdi-Wali won with 12,292 votes. Green candidate Zack Polanski came second with 8,928.

At the general election on 4 July 2024, the Green candidate in the same constituency, David Robert Stansell, received 4,030 votes. The party therefore gained 4,898 votes between the two elections.

That is a rise of about 122%, or just over twice the earlier number. This increase is real; it is not created by presenting the result as a percentage.

Sources: Camden CouncilCamden CouncilUK Parliament

A vote count and a vote share answer different questions

The vote count answers: how many votes did the candidate receive? Vote share answers: what proportion of all valid votes went to that candidate?

To calculate the share, divide the Green votes by the total votes for every candidate. Leave out rejected ballots because they were not accepted as votes for a candidate.

The council tables add up to 38,602 valid votes in 2024 and 27,211 in 2026. That is about 30% fewer valid votes overall at the later election.

Sources: Camden CouncilCamden Council

Holborn and St Pancras: share of valid candidate votes, calculated from the unrounded counts
Measure4 July 20248 October 2026
Green votes4,0308,928
All valid candidate votes38,60227,211
Green share10.44%32.81%

More Green votes, in a smaller overall total

In 2024, the Greens received 4,030 of the 38,602 valid votes: 10.44%. In 2026, they received 8,928 of the 27,211 valid votes: 32.81%.

Think of each result as 100 valid votes. About 10 were Green in 2024, compared with about 33 in 2026. This is a way to picture the proportions, not a claim that we followed the same 100 voters.

The party had more votes of its own, while there were fewer votes overall. Together, those changes explain why its share more than tripled even though its vote count did not.

For readers comparing the exact figures: the vote count became 2.22 times its previous size, while the share became 3.14 times its previous size. The change from 10.44% to 32.81% is a gap of 22.37 percentage points.

Sources: Camden CouncilCamden CouncilUK Parliament

What this result can and cannot tell us

"The Green vote share more than tripled" is an accurate description of these figures. "The number of Green votes tripled" would be incorrect. We are explaining the difference, not correcting an identified false post.

The two elections had different candidates and took place in different circumstances: one was a general election and one was a by-election. The result tables cannot tell us which voters changed parties, stayed home or joined the electorate, or why they did so.

Nor can one constituency show how popular a party is across the country. Parliament's records confirm that both elections used the constituency created in 2024, so a changed constituency is not the explanation for this comparison.

We checked the official candidate totals and calculated each share from those totals. The council and Parliament display different turnout percentages; our calculation does not depend on either turnout figure. The linked tables and calculations let you repeat the check.

Sources: UK ParliamentCamden CouncilCamden CouncilUK Parliament

What is established

  • The Green vote count rose from 4,030 to 8,928.
  • Its share of valid votes rose from 10.44% to 32.81%.
  • Labour retained the seat.

What remains unknown

  • Which voters changed parties, stayed home or entered the electorate.
  • Why voting patterns and participation changed.
  • What this single result means for support across the country.

The story, in order

General election
Green candidate David Robert Stansell received 4,030 votes out of 38,602 valid candidate votes

Parliamentary by-election
Green candidate Zack Polanski received 8,928 votes out of 27,211 valid candidate votes. Labour's Sagal Abdi-Wali won.

Camden published its result announcement
The council's announcement linked the full candidate results

YOUR TURN TO CHECK

Reproduce this investigation

Use the official candidate results to work out the Green vote count and share for each election. You can do this with a calculator.

You will need
A browser and a calculator, or Excel/another spreadsheet app for the candidate CSV

Candidate votes with sources · CSVCalculation results · JSONRecalculate in Python · ZIPStep-by-step guide · TXTSource notes · TXTTimeline · CSV
What you can and cannot repeat

You can repeat the arithmetic, but these tables cannot track individual voters or explain their decisions. Leave rejected ballots out of vote-share calculations. Published turnout percentages are a separate measure.

  1. Find the 2024 Green result

    Open Camden's 2024 general election results and select Holborn and St Pancras. Copy all 12 candidate vote counts. Find the Green candidate, David Robert Stansell.

    Green votes = 4,030
    Open Camden's 2024 table

    Keep in mind Do not use the separate Hampstead and Highgate table on the same page.

  2. Find the 2026 Green result

    Open the 2026 by-election table. Copy all 15 candidate vote counts and find Zack Polanski's Green result. Do not add the 108 rejected ballots to the candidate votes.

    Green votes = 8,928
    Open Camden's 2026 table

    Keep in mind Rejected ballots do not contribute to the share of valid candidate votes.

  3. Add up the valid votes in each election

    In the downloadable CSV, filter election_date to one election at a time, then add the votes column. Keep the two years separate.

    2024 total = 38,602
2026 total = 27,211

    Keep in mind The total counts votes accepted for candidates, not everyone who was eligible to vote.

  4. Work out the Green share

    For each year, divide the Green votes by that year's total valid votes, then multiply by 100.

    2024: 4,030 / 38,602 * 100 = 10.43987358%
2026: 8,928 / 27,211 * 100 = 32.81026056%

    Keep in mind Keep unrounded values for subsequent calculations.

  5. Compare the two changes

    First divide the 2026 Green vote count by the 2024 count. Then divide the 2026 share by the 2024 share. These tell you how many times larger each measure became. To find the gap in percentage points, subtract the earlier percentage from the later one.

    Count multiple = 8,928 / 4,030 = 2.21538462
Share multiple = (8,928 / 27,211) / (4,030 / 38,602) = 3.14278332
Share gap = (8,928 / 27,211 - 4,030 / 38,602) * 100 = 22.37038697 percentage points

    Keep in mind A rise from 10% to 30% is a 20-percentage-point gap and a tripling of the share. Those are different ways to describe the same change.

  6. Check the scope of the comparison

    Cross-check the 2026 counts against Parliament's table. Use its election history to confirm the same constituency. Remember that this compares different candidates and election types, not a national opinion poll.

    Cross-check Parliament's resultCheck constituency history

    Keep in mind This article explains the numbers. It does not establish deliberate deception.

Sources you can check

TALLYCAT01 calculations use all candidate vote rows and exclude rejected ballots from the vote-share denominator. All ratios are calculated before rounding. The council and Parliament display different turnout percentages; the explanation does not depend on either reported turnout. A separate signed PDF declaration was not recovered, so we do not claim to have inspected one. The complete official HTML result tables answer the count/share question.

Inspect the evidence

Source notes include the links and verification limits. The timeline opens in Excel or any spreadsheet app.

Source notes · TXTTimeline · CSVCandidate votes with sources · CSVCalculation results · JSONRecalculate in Python · ZIP

Photo credit and licence

Zack Polanski (cropped), 1 October 2022. Bristol Green Party; photographer metadata credits Gary Lawson / Wikimedia Commons. CC0 1.0. Resized for display and cropped to fit. Homepage photographs also have a dark gradient and text overlay.

Carousel photograph credits and licences

  • Zack Polanski (cropped)
    Bristol Green Party; embedded photographer credit @garylawsonphotography.com. Archive portrait of Zack Polanski, photographed on 1 October 2022. The photo is contextual and does not depict the 2026 election.
    CC0 1.0. Cropped, resized and darkened for the carousel.
  • House of Commons Chamber 1
    UK Parliament. Archive photographic frame of the empty House of Commons chamber, from a UK Parliament film published in 2012. Previously enhanced for brightness and contrast on Commons.
    CC BY 3.0. Cropped, resized and darkened for the carousel.
  • Bath 2019 general election count
    Rwendland. An archive photograph of the Bath constituency count in the 2019 general election. Workers sort papers at tables beneath team labels and a BATH sign; used only to illustrate the counting process. The adapted slide is also offered under CC BY-SA 4.0; logo trademark rights remain unaffected.
    CC BY-SA 4.0. Cropped, resized and darkened for the carousel. Original and photographic adaptation under CC BY-SA 4.0.
  • Holborn Tube Station Logo
    Roland Tanglao (Flickr name roland). A 2005 archive photograph of the Holborn Underground station roundel, used as location context only.
    CC BY 2.0. Cropped, resized and darkened for the carousel.
  • Ballot paper for the 2021 United Kingdom local elections (Coventry, Westwood ward)
    domdomegg. An unmarked Coventry City Council local-election ballot photographed in April 2021, used as an archive illustration of one candidate choice. It is not a ballot from either Holborn parliamentary election.
    CC BY 4.0. Cropped, resized and darkened for the carousel.
  • Vote- Polling station for UK's 2017 general election
    Descrier. A 2017 general-election polling-station sign in Camberwell, London, photographed by Descrier. Archive context only; no voters are shown.
    CC BY 2.0. Cropped, resized and darkened for the carousel. Photographer website.

These credits cover the carousel photographs. Original stock files are not distributed here. The logo remains separate brand artwork.

Research and production assisted by AI. Read our approach

Have evidence that could change this finding? Contact TALLYCAT01