Reform and Plaid Cymru neck and neck in latest Senedd election poll

Martin Shipton

Reform UK and Plaid Cymru are neck and neck in our exclusive new poll ahead of the 2026 Senedd election, with Labour some way behind in third place.

According to the Beaufort Research poll for Nation.Cymru, Reform are on 27% (down 3 since September 2025), Plaid Cymru on 26% (up 4), Labour on 21% (down 2); the Conservatives on 12% (down 1), the Greens on 9% (no change), the Liberal Democrats on 3% (down 1) and other parties on a total of 2% (up 1).

Using the Cavendish Cymru seat projector, these results would put Reform and Plaid Cymru both on 30 seats, Labour on 24 seats, the Conservatives on 9, the Green Party on 2 and the Liberal Democrats on 1.

Such an outcome would almost certainly see Plaid Cymru leader Rhun ap Iorwerth elected as First Minister, either as head of a formal coalition with Welsh Labour or in some kind of cooperation agreement that would not entail there being Labour ministers.

There were variations in party support by region, gender, age, social grade and whether those surveyed were Welsh speakers or not.

For the purposes of the poll, Wales was split into three regions: North and Mid Wales; South West Wales and Valleys; and Cardiff and South East Wales.

In North and Mid Wales, Reform led with 29%, followed by Plaid Cymru on 24%, Labour 17%, Conservative 13%, Greens 12%, Liberal Democrats 3% and other parties 2%.

In South West Wales and Valleys, Plaid Cymru led with 31% followed by Reform on 25%, Labour 22%, Conservative 11%, Greens 6%, Liberal Democrats 3% and others 3%.

In Cardiff and South East Wales, Reform led with 31%, with Labour on 24%, Plaid Cymru 22%, Conservative 12%, Greens 8%, Liberal Democrats 1% and others 2%.

Reform leads among men, with 29% of the vote against 26% for Plaid Cymru, 23% for Labour, 9% for the Conservatives, 8% for the Greens, 3% for the Lib Dems and 3% for other parties.

Plaid Cymru leads among women, with 27% against 26% for Reform, 19% for Labour, 15% for the Conservatives, 9% for the Greens, 2% for the Lib Dems and 2% for other parties.

In terms of age, 16-34 year-olds were most likely to support Plaid Cymru. Some 31% did so, against 28% for Labour; 17% for the Greens; 14% for Reform; 4% for both the Conservatives and Lib Dems and 2% for some other party.

Reform leads amongst those aged between 35 and 54, with 36% support; followed by Plaid Cymru on 24%; Labour on 19%; the Conservatives and Greens both on 8%; the Lib Dems on 3%; and others on 1%.

In the 55 and over age group, Reform led with 27%; followed by Plaid Cymru on 26%; Labour 19%; Conservatives 16%; Greens 6%; Lib Dems 2%; and others 3%.

Prosperous

Those in the more prosperous ABC1 social grades were most likely to support Plaid Cymru, which had 30% support; followed by Labour on 23%; Reform on 20%; the Conservatives 12%; Greens 9%; Liberal Democrats 4% and others 3%.

By contrast, those in the less prosperous C2DE grades showed most support for Reform at 37%; followed by Plaid Cymru on 22%; Labour on 18%; Conservatives 12%; Greens 8%; Liberal Democrats 1%; and others 2%.

Among Welsh speakers, 41% backed Plaid Cymru; Labour 26%; Reform 15%; Greens 8%; Conservatives 6%; and the Lib Dems and other parties both on 2%.

Amongst those who cannot speak Welsh, Reform led with 31%; followed by Plaid Cymru on 22%; Labour on 19%; the Conservatives on 14%; the Greens 9%; Lib Dems 3%; and others on 2%.

* A total of 1,000 interviews conducted between November 10 and November 30 2025 were completed and analysed. The headline voting intention data is based on 505 respondents who fulfilled all the following criteria:

- Those aged 16+.

- Those who stated a preference for a party (ie excluding those who would not vote, don’t know

how they would vote or prefer not to say).

- Those who said they are 9 or 10 out of 10 in terms of their certainty to vote in a Senedd election (ie. excluding those who say they are 8 out of 10 or below).

Constituency seat projections

Afan Ogwr Rhondda: Reform 2; Labour 2; Plaid Cymru 2

Bangor Conwy Môn: Plaid Cymru 3; Reform 1: Conservative 1: Labour 1

Blaenau Gwent Caerffili Rhymni: Labour 2: Plaid Cymru 2; Reform 2

Brycheiniog Tawe Nedd: Reform 2: Plaid Cymru 1; Labour 1: Lib Dem 1; Conservative 1

Caerdydd Ffynnon Taf: Labour 2; Reform 2; Plaid Cymru 1; Green 1

Caerdydd Penarth: Plaid Cymru 2: Labour 2; Reform 1; Green 1

Casnewydd Islwyn: Reform 2; Labour 2; Plaid Cymru 1; Conservative 1

Ceredigion Penfro: Plaid Cymru 3; Reform 2; Conservative 1

Clwyd: Reform 2; Conservative 2; Labour 1; Plaid Cymru 1;

Fflint Wrecsam: Reform 2; Labour 2; Plaid Cymru 1; Conservative 1

Gwynedd Maldwyn: Plaid Cymru 4; Reform 2

Gŵyr Abertawe: Reform 2; Labour 2; Plaid Cymru 2

Pen-y-bont Bro Morgannwg: Reform 2; Labour 2: Plaid Cymru 1; Conservative 1

Pontypridd Cynon Merthyr:Reform 2; Labour 2; Plaid Cymru 2

Sir Fynwy Torfaen: Reform 2; Labour 2; Conservative 1; Plaid Cymru 1

Sir Gaerfyrddin: Plaid Cymru 3; Reform 2; Labour 1

In the interests of transparency, here are further notes from Beaufort Research explaining the methodology of how the poll was conducted:

The Omnibus sample is designed to be representative of the adult population resident in Wales aged 16 and over.

Interviews were self-completion and undertaken online using the Cint online panel exchange platform.

The Cint platform and its products comply with ESOMAR, MRS, ARF, MRIA, AMA, AMSRO and Insights Association standards.

Cint also complies with ISO 20252. Multiple data quality checks are built into the Cint system including GEO IP check and CAPTCHA at registration, unique respondent identification and fraudulent behaviour checks. On top of this Beaufort builds in its own quality control questions and measures within the survey and excludes respondents who fail these checks.

The survey was subject to interlocking demographic quota controls of age within gender. A further separate quota control was set on social grade and questionnaires were completed by residents of every local authority in Wales.

Online surveys were configured for PC/tablet and smartphone completion and English and Welsh versions of each were offered to every respondent.