8JUL

Electoral Calculus projects first SNP majority since 2011

3 min read

10:13UTC

An Electoral Calculus Holyrood MRP published on 7 April 2026 projects the SNP on 67 seats, two above the 65-seat majority line, on fieldwork from 4,105 respondents taken 13-31 March.

← Back to Farage to quit Clacton to force by-election Jump to analysis ↓

PoliticsDeveloping

Key takeaway

Electoral Calculus's 7 April MRP is the first Holyrood model since 2011 to project an outright SNP majority.

Electoral Calculus published a Multilevel Regression and Post-stratification (MRP) model of the 6 May 2026 Holyrood election on 7 April 2026, based on a fieldwork sample of 4,105 respondents taken between 13 and 31 March. The model projects the Scottish National Party on 67 seats, two above the 65-seat majority threshold in the 129-seat chamber. All 67 projected seats come from constituency wins; Electoral Calculus allocates the SNP zero regional list seats.

The projection is methodologically unusual because the Additional Member System (AMS) was designed to prevent exactly this outcome. Regional list seats are distributed by a corrective formula that subtracts constituency wins from each party's vote share, meaning constituency over-performance normally drains a party's list entitlement. Electoral Calculus director Martin Baxter has argued the 67-seat projection requires the SNP to concentrate its vote above the model's historical tolerances, a scenario produced by five-party fragmentation of the anti-SNP vote rather than by any SNP surge.

The model is the first Holyrood projection of an outright SNP majority since 2011. No such majority has occurred since Alex Salmond's 2011 win, which produced the 2014 independence referendum on a 55-45 No result. A projected repeat lands on a chamber already losing a record 39 MSPs to retirement, and reframes the April campaign around a question the polling had not previously been asked to answer.

Deep Analysis

In plain English

Scotland elects its parliament using a system with two types of seats: 73 local constituency seats (won in head-to-head races) and 56 regional list seats (awarded proportionally to correct for how lopsided the constituency results were). The system was designed specifically to stop any one party winning a majority. Despite that design, the SNP won a majority once — in 2011 under Alex Salmond. That majority produced the 2014 independence referendum, which Scotland voted 55-45 to stay in the UK. A polling model published on 7 April 2026 projects the SNP winning another majority — 67 seats, two above the 65 needed. First Minister John Swinney has said this would give him a mandate to push for a second independence vote. The model works out this way because the parties opposed to the SNP are now splitting their votes between five different options, which reduces the built-in correction mechanism. One month is a long time in politics. The SNP has been projected to win majorities before polling day twice since 2011 and failed both times.

Deep Analysis

Root Causes

The projected SNP majority emerges from five-party fragmentation of the unionist and centre-right vote, not from any surge in SNP support. The Electoral Calculus model projects the SNP on a vote share similar to or below 2021, but with the anti-SNP vote split across Scottish Labour, Scottish Conservatives, Scottish Liberal Democrats, Scottish Greens and Reform UK.

This fragmentation reduces the efficiency of the AMS correction, because the formula distributes list seats proportionally among all parties, including smaller ones whose small vote shares produce small list-seat returns.

The deeper structural cause is the UK-wide realignment since the 2014 referendum. Scottish Conservative support rose to 31 seats in 2016 on Ruth Davidson's unionist platform, then has declined steadily. Reform UK's entry into Scottish politics at regional list level draws predominantly from the 2016-2021 Scottish Conservative voter base — voters who were never SNP-friendly but who are now distributing their votes across two right-of-centre options rather than one.

Escalation

The projection is directionally escalatory for Scottish independence politics: a confirmed majority would immediately test whether the UK government's position that a Section 30 order is required for a legal referendum can hold against a renewed SNP mandate.

The Scottish Government's 2021 Supreme Court referral on unilateral referendum legality produced a definitive ruling that Holyrood cannot legislate for a binding referendum without Westminster consent. A majority does not change the legal position; it restarts the political negotiation.

What could happen next?

Consequence
An SNP majority would immediately place a Section 30 request for a second independence referendum on John Swinney's agenda, requiring the UK government to formally respond.
Short term · 0.8
Risk
Reform UK on 14 Holyrood list seats would become the official opposition at Holyrood, with the staffing, scrutiny and profile rights that entails — and no constituency seat base in Scotland.
Short term · 0.65
Precedent
If the AMS majority is achieved despite the system's corrective design, it would prompt renewed academic and political debate about whether AMS as configured is fit for purpose in a five-party era.
Medium term · 0.72
Risk
Scottish Conservative collapse to nine seats would leave the centre-right unionist coalition in Holyrood without sufficient seats to sustain a credible opposition, potentially forcing a de facto Reform-Conservative alliance.
Short term · 0.68

Source Landscape

This story draws on neutral-leaning sources

Primary parallel: Alex Salmond's SNP won 69 seats in 2011 — the only outright Holyrood majority in the Parliament's history. The mechanism was identical to the one Electoral Calculus now projects: the opposition vote was fragmented enough (Labour, Conservative, Liberal Democrat all viable) that the AMS corrective formula failed to prevent constituency over-performance translating into a net majority. That majority produced the 2014 independence referendum, which returned a 55-45 No.

Counter-parallel: In 2016 and 2021, the SNP polled consistently ahead of the majority threshold but fell short on polling day — 63 and 64 seats respectively. Both elections saw late movement of list votes to The Greens as a pro-independence coalition partner, rather than to the SNP directly. The 2026 model projects a different mechanism, but the base rate of projected SNP majorities not materialising is two from two.

Consensus view: The British Election Study's Chris Hanretty, who pioneered MRP methodology for UK elections, has noted that AMS creates natural modelling uncertainty because the compensatory list mechanism produces non-linear relationships between vote share and seat total.

A constituency-heavy SNP over-performance, as projected, requires the regional list correction to be unusually small — which happens only when opposition votes are distributed efficiently across multiple parties rather than concentrated.

Counter-view: Reform Scotland and the David Hume Institute have observed that polling before Scottish Parliament elections has consistently overestimated SNP performance in the weeks before polling day since 2016, when the SNP fell short of poll-suggested majorities twice in succession. A 67-seat projection at T-30 should be read against a track record of late SNP underperformance relative to polls.

Key tension: Whether five-party fragmentation of the unionist vote is sufficiently geographically concentrated to prevent the AMS correction from clawing back SNP constituency gains via list seats — the specific mechanism Electoral Calculus's Martin Baxter acknowledges requires historically atypical vote distribution.

Sources:Electoral Calculus

Mentions:Electoral Calculus →John Swinney →Multilevel Regression and Post-stratification (MRP) →Alex Salmond →Prima →Reform UK →Green Party →Scottish Conservatives →Welsh Labour →Additional Member System (AMS) →Parliament →Scottish National Party (SNP) →

First Reported In

Update #1 · Wales rewrites parliament no voter has used

Electoral Calculus· 7 Apr 2026

Read original →

Causes and effects

This Event

Electoral Calculus projects first SNP majority since 2011

An outright SNP majority under AMS is mathematically difficult because the regional list compensates for constituency over-performance. The model projects one anyway.

Led to

More in Common puts SNP short of majority

The Electoral Calculus SNP 67-seat projection (2101) is contrasted with More in Common's 56, establishing the 11-seat uncertainty range.

Occurred 24 Apr 2026

Read story →

Tories Projected to Hold Zero Scottish Constituency Seats

Occurred 7 Apr 2026

Read story →

IFS Demolishes Both Scottish Opposition Manifestos

Occurred 8 Apr 2026

Read story →

Second MRP confirms SNP majority path

YouGov Holyrood MRP corroborates the Electoral Calculus MRP from 7 April, with both projecting SNP majority at 67 seats and Scottish Conservative constituency wipeout.

Occurred 11 Apr 2026

Read story →

Different Perspectives

Local Government Association

New chair Eamonn O'Brien broke the LGA's neutrality on reorganisation for the first time since December 2024, telling the incoming prime minister the 'unprecedented scale' of disaggregation risks statutory services and asking him to adjust the LGR timetable, aligning the sector's institutional voice with the judicial review claims Suffolk, Essex and Norfolk have already filed.

Reform UK

Reform's £12m Harborne windfall from 2025 predates the 25 March retrospective crypto-donation ban, so the new law does not reach it; a live police probe into £500,000 of Cottrell donations continues, while the standards inquiry into Harborne's undeclared £5m gift is frozen since Farage vacated Clacton on 8 July.

Incoming Burnham government

Burnham secured 349 of the backing the NEC's 81-MP threshold required by 13 July, becoming Labour leader on 17 July and prime minister on 20 July without a members' ballot. He now inherits the LGR timetable the LGA wants him to adjust and a finance bill his own backbencher, Liam Byrne, is trying to harden further via amendment NC34.

Rupert Lowe / Restore Britain

Lowe called Farage's Clacton contest a 'Reform-sponsored media circus' and kept Restore Britain out of it, pledging instead to stand in the second by-election he expects the Harborne inquiry to force later this year. The boycott doubles as a signal that Restore Britain, not Reform, will contest the 'real' vote.

Kemi Badenoch / Conservative Party

Badenoch dismissed Farage's Clacton contest as a 'fake by-election' and joined the boycott, but reserved the right to fight a genuine second contest if Greenberg's inquiry forces one. The Conservatives are betting their real fight comes in autumn, not July.

Rhun ap Iorwerth / Plaid Cymru

Ap Iorwerth defended a £145m NHS allocation at First Minister's Questions on 1 July, blaming a graduate-nurse shortfall on 2022 training decisions taken before his Plaid-led government took office in May. Answering for choices his government did not make gives his new administration its first real accountability test.