FindOutNowUK

[andrew111]

Dec 7, 2021 23:29:57 GMT batman said:

Dec 7, 2021 12:07:26 GMT andrew111 said:

It is hard to see how their method could possibly account for the Ashfield Independents. If they disappear then Labour do probably win Ashfield

Not sure why the Ashfield Independents would disappear. Basically they are the Lib Dems in all but name in that constituency.

I am not suggesting they will disappear, just trying to account for the results in this poll

[cogload]

Screams outlier to me here.

[bigfatron]

FindOutNow is a bit weird regards smaller parties; since August, while most pollsters are showing Lib Dems up a bit, FON are showing us down 2.

But I could imagine it being within MOE for FON - Labour appeared to be about 38, Tories around 32, LDems around 10 and Greens around 7, but FON always seem to be higher for Greens than most polls (along with YouGov)

[batman]

Dec 18, 2021 8:14:33 GMT cogload said:

Screams outlier to me here.

why? several polls last week were similar. The Tories may be a bit low perhaps, but not by very much. 

[The Bishop]

So what is their seat "projection" for this poll?

[bigfatron]

Dec 18, 2021 12:07:36 GMT The Bishop said:

So what is their seat "projection" for this poll?

It'll show the LibDems losing seats - it always does!

[The Bishop]

I'm just as interested to see if they have the Nats sweeping every Scottish seat again.

[batman]

On a strict application of the swing in this poll, the Liberal Democrats would not lose seats. They would gain 12 seats from the Conservatives but lose 2 to the SNP. The SNP would not win every Scottish seat ; they would gain all the Conservative ones, but would lose Orkney & Shetland and Edinburgh W to the Liberal Democrats, and Edinburgh South and Kirkcaldy & Cowdenbeath to Labour. There would be an Independent gain from the Conservatives in East Devon, and of course a very large number of Labour gains.

[jm]

[woollyliberal]

New Find Out Now/Electoral Calculus opinion poll 14-18 Feb
Lab 38% +3 since Dec
Con 32% -4
Lib Dem 13% +2
Green 7% -1
Reform 3% +1

Predicted seats
Lab 308 +105
Con 243 -122
SNP 59 (usual* pinch of salt needed)
LD 16 +5
PC 5 +1
Gr 1 =

*SNP will not get 59 of 59 seats in Scotland so Con, Lab and LD should be expected to get a few more each than stated.

Lab / SNP is a viable government. Lab / LD is probably just about possible given SNP not getting all 59. No Con government is possible.

[clyde1998]

Electoral Calculus have done another MRP, this one for the local elections (figures for England and Wales only unless stated) - based on two FindOutNow polls:

Wards:
Lab 3,514 (+819)
Con 978 (-548)
LDm 662 (+49)
Plaid 151 (-42)
Grn 23 (-26)
Oth 265 (-251)

Councils:
Lab 85 (+16)
Con 39 (-4)
LDm 14 (+3)
Ind 3 (nc)
Plaid 1 (nc)
NOC 26 (-11)

The equivalent national vote share is estimated as: Lab 39%, Con 24%, LDm 15%. Around 45% of those who responded to the poll say they wouldn't vote.

[gwynthegriff]

May 4, 2022 15:38:20 GMT clyde1998 said:

Electoral Calculus have done another MRP, this one for the local elections (figures for England and Wales only unless stated) - based on two FindOutNow polls:

Wards:
Lab 3,514 (+819)
Con 978 (-548)
LDm 662 (+49)
Plaid 151 (-42)
Grn 23 (-26)
Oth 265 (-251)

Councils:
Lab 85 (+16)
Con 39 (-4)
LDm 14 (+3)
Ind 3 (nc)
Plaid 1 (nc)
NOC 26 (-11)

The equivalent national vote share is estimated as: Lab 39%, Con 24%, LDm 15%. Around 45% of those who responded to the poll say they wouldn't vote.

The "others" presumably are largely Independents. Given the general anti-politics mood I'd be rather surprised to see them losing (net) half their existing seats.

[jmagosh]

Think there will be a hat-eating feast if that FindOutNow poll is correct.

Not that it will be, because EC are never correct, it seems.

[The Bishop]

Utterly risible stuff, little more to be said.

[iainbhx]

May 4, 2022 16:40:19 GMT jmagosh said:

Think there will be a hat-eating feast if that FindOutNow poll is correct.

Not that it will be, because EC are never correct, it seems.

FindOutNow could be correct, but Baxter is never correct especially in Scotland.

[london(ex)tory]

May 4, 2022 15:38:20 GMT clyde1998 said:

Electoral Calculus have done another MRP, this one for the local elections (figures for England and Wales only unless stated) - based on two FindOutNow polls:

Wards:
Lab 3,514 (+819)
Con 978 (-548)
LDm 662 (+49)
Plaid 151 (-42)
Grn 23 (-26)
Oth 265 (-251)

Councils:
Lab 85 (+16)
Con 39 (-4)
LDm 14 (+3)
Ind 3 (nc)
Plaid 1 (nc)
NOC 26 (-11)

The equivalent national vote share is estimated as: Lab 39%, Con 24%, LDm 15%. Around 45% of those who responded to the poll say they wouldn't vote.

Obviously precisely how changes in seats translate to changes in councils will depend on where exactly the seats are lost, but I would be utterly gobsmacked if a net loss of 548 seats only saw a net loss of 4 councils, particularly as on my figures the Tories are defending 22 councils with a majority of 10 seats or fewer (so only 5 losses for NOC).

I have repeated the largely pointless but very interesting exercise of trying to model what seat numbers / changes there would be if people and places voted exactly the same way they did in 2019.

Taking the 2019 results where the same ward is up now is easy, but there was some jiggery pokery required to deal with boundary changes, the councils that didn't have elections in 2019 (including London but I haven't included Scotland and Wales) I've done a forecast for each ward. It's not a prediction of what I think will happen today, it's just my best effort at modelling a very particular scenario.

Anyway, the big spreadsheet comes up with a net loss of 183 seats and 8 councils in England. I first wanted to create this as a kind of benchmark for me to comment on whether today's results should put further pressure on Boris, by comparing these results to May's 2019 disaster.

[jmagosh]

May 5, 2022 14:43:01 GMT iainbhx said:

May 4, 2022 16:40:19 GMT jmagosh said:

Think there will be a hat-eating feast if that FindOutNow poll is correct.

Not that it will be, because EC are never correct, it seems.

FindOutNow could be correct, but Baxter is never correct especially in Scotland.

The headline local election voting intention on the Find Out Now local election poll was 39% Labour, 22% Conservative, 12% Lib Dem, 9% Green, 2% Reform, 7% SNP, 1% Plaid, 7% Independents/others.  The poll covered all of GB, and looks like it's weighted to how many different local authorities are voting in those regions, as there's hardly any sampled in the East Midlands, for instance. 

This isn't a very good comparison to 2018, because there were only elections in England then, rather than Wales and Scotland as well (and indeed some of the councils being fought in England are different too). But for reference, back then it was Lab 41%, Con 32%, LD 14%, Green 6.5%, UKIP 1%, other 5%.

Nonetheless, Find Out Now say this constitutes a '6 point swing'. Which, even despite the fact the Scottish and Welsh seats were last fought in 2017 (when Con was much stronger and Lab weaker), feels too large to me. That's a long winded way of saying I remain skeptical of Find Out Now's polling, as well as EC's projection. Even without EC's dodgy projection, these numbers would imply substantially bigger losses for the Tories than anyone else is predicting, I think.

Tables for the FindOutNow polling can be downloaded here.

[Tony Otim]

So they have both Lab and Con down vs 2018 and Green up but Green losing half their seats...

How does that even work?

[jmagosh]

May 5, 2022 17:36:11 GMT Tony Otim said:

So they have both Lab and Con down vs 2018 and Green up but Green losing half their seats...

How does that even work?

Well indeed.

[clyde1998]

May 5, 2022 17:36:11 GMT Tony Otim said:

So they have both Lab and Con down vs 2018 and Green up but Green losing half their seats...

How does that even work?

Boundary changes? A lot of councils seem to be reducing the number of elected members, which could reduce the Green seats despite increasing the vote share (or the vote is distributed differently than 2018) - assuming 2018 is the baseline.