As any lockdown online quizzer will tell you, the collective noun for a group of unicorns is a blessing. For psephologists, the same term can also be applied to opinion polls. With two polls in two days saying very different things, in order to try and make sense of the situation it is first worth recapping on the data.

BBC Poll

The BBC polling was done by ICM Unlimited and was based on a representative sample of 1,001 people aged 16+ interviewed by telephone from 28 January to 21 February 2021.

The poll produced a Senedd which looks like this:

  • Labour 30
  • Plaid Cymru 15
  • Conservatives 13
  • Liberal Democrats 1
  • UKIP 1

This is based on uniform national swings linked to percentages shares in each of the constituencies and regions.

  • Constituency: Lab 39%, Con 24%, Plaid 24%, Lib Dem 4%.
  • Region: Lab 37%, Con 22%, Plaid 22%, UKIP 4%, Abolish 4%.

On a constituency basis, the modelling predicts Labour winning back Rhondda from Plaid Cymru.

The BBC poll also asked about options for the constitutional future of Wales and produced the following options:

  • Wales should become independent, separate from the UK – 14% support
  • The Welsh Parliament should have more powers than it currently has – 35% support
  • The powers the Welsh Parliament currently has are sufficient and should remain as it is now – 27% support
  • The Welsh Parliament should have fewer powers than it currently has – 3% support
  • The Welsh Parliament should be abolished and Wales governed directly from Westminster – 15% support

Western Mail Poll

The Western Mail polling was done by YouGov and based on a representative sample of 1,004 people engaged online from 19 to 22 February 2021.

The poll produced a Senedd which looks like this:

  • Labour 24
  • Conservatives 16
  • Plaid Cymru 14
  • Abolish the Assembly 5
  • Liberal Democrats 1

Again, this is based on uniform national swings linked to percentages shares in each of the constituencies and regions.

  • Constituency: Lab 33%, Con 28%, Plaid 22%, Lib Dem 4%, Reform UK, 4% Green, 5% other.
  • Region: Lab 29%, Con 25%, Plaid 24%, Abolish 9%, Green 5%, Reform UK 3%, UKIP 2%, Lib Dem 2%, and 1% for both the Communists and Wales National Party.

On a constituency basis, the modelling predicts Labour losing the Vale of Glamorgan, Vale of Clwyd, Wrexham and Gower to the Conservatives; and both Llanelli and Blaenau Gwent to Plaid Cymru, but gaining a third list seat in Mid and West Wales.

On the matter of independence for Wales, 50% oppose, 14% are undecided, and 9% say they would not vote. Among those giving a definitive answer, 33% say they would vote Yes on independence for Wales, while 67% say they would vote No.

Conclusions

One day you get an opinion poll telling you that Labour is on course for its best ever result in a Senedd election. The next day comes an opinion poll telling you that Labour is heading for its worst ever result. Apart from demonstrating huge volatility in the electorate, what does this all actually mean?

The first conclusion must surely be that Labour looks certain to ‘win’ the next Senedd election, though the number of champagne corks popped cannot be predicted. It will all come down not just to the results themselves, but to the results of those results when the Senedd is convened in order to take nominations for First Minister. Back in 2016, the Conservatives and UKIP named Leanne Wood as their preferred choice, forcing a second ballot for Carwyn Jones before he won. Next time round, will all the opposition parties be inclined to work together around promoting a single alternative?

The answer here must surely be who that alternative actually is, and that depends on who comes second. The BBC polling has the Conservatives and Plaid neck and neck in terms of percentages, but the vagaries of the list system on their modelling would give more seats to Plaid. The Western Mail poll puts the Conservatives in a clear second place. But would Plaid back a Conservative candidate in the same way the Conservatives backed Plaid in 2016?

There is also the issue of what happens if there are three or more candidates for First Minister in a contest in the Senedd. The Standing Orders are clear on this process:

Standing Order 8.3: If more than two Members have been nominated and no Member receives more than half of the votes cast by roll call, the candidate who has received the smallest number of votes must be excluded and further votes by roll call taken until one candidate obtains more than half of the votes cast; and the Presiding Officer must declare that Member to be the nominee. If there is an equality of votes between the two remaining candidates a further vote by roll call must take place.

Looking at the matter of constitutional futures, most of the media focus has been on the rise of the independence option to 14% compared to 11% last year, or the marginal change in the abolition option from 14% last year to 15% this year. Yet within the polling the biggest shift has been the drop of a full 8% from 43% to 35% for those who believe Senedd should have more powers than it currently has. There are also of course 27% of respondents who believe that the powers the Senedd currently has are sufficient and should remain as it is now. With most analysis and reporting usually pointing to the extremes of abolition or independence, it is worth noting that even together these options only comprise 29% of the electorate, which is virtually the same as those who want no change whatsoever. Indeed, the Welsh electorate seems to be split into thirds – incremental shift in powers, no change, and total change.

So, what sort of blessing are these opinion polls? Like the unicorns, they may just be mythical blessings in name only.