It begins lawyerly and philosophical, turns darkly religous, and then ends with this astonishing paragraph:
With malice toward none, with charity for all, with firmness in the right as God gives us to see the right, let us strive on to finish the work we are in, to bind up the nation’s wounds, to care for him who shall have borne the battle and for his widow and his orphan, to do all which may achieve and cherish a just and lasting peace among ourselves and with all nations.
(I should have posted this on March 4th, but was away from home at the time.)
"With malice towards all, with charity for none..." could have been the start of Trump's inaugural.
On the surface this is an arcane argument between statistics anoraks. In reality Reform are troubled. Their real problem being not YouGov - who like everyone else have a method to analyse the raw data - but the Reform downward curve which is, I think, universal among polling companies.
Can we expect a Reform 'Polling companies lie' campaign in true Trump style?
It is more about Yougov trying to poll tactical voting by asking how respondents will vote in their own constituency rather than who they want to win. This results in Reform being about five points lower than with other polling companies. If valid, and this is as important for betting as for politics, it shows many voters are willing to vote tactically to keep Reform out of power. (And by valid, I mean valid in the technical sense that a test or poll measures what it purports to measure.)
That doesn’t make sense. People changing away from a preference to vote Reform might be tactical voting but it cannot be tactical voting against Reform!
It could make sense if the change in the question affects how likely people are to vote. A disaffected voter may express low likelihood to vote on the first question, but be motivated by the thought of keeping out Reform and give a higher likelihood on the second question. That would reduce Reform’s percentage after weighting.
Not saying that is what’s happening. But it’s a possibility.
It begins lawyerly and philosophical, turns darkly religous, and then ends with this astonishing paragraph:
With malice toward none, with charity for all, with firmness in the right as God gives us to see the right, let us strive on to finish the work we are in, to bind up the nation’s wounds, to care for him who shall have borne the battle and for his widow and his orphan, to do all which may achieve and cherish a just and lasting peace among ourselves and with all nations.
(I should have posted this on March 4th, but was away from home at the time.)
"With malice towards all, with charity for none..." could have been the start of Trump's inaugural.
It would still have been a lie given his affection for Putin.
On the surface this is an arcane argument between statistics anoraks. In reality Reform are troubled. Their real problem being not YouGov - who like everyone else have a method to analyse the raw data - but the Reform downward curve which is, I think, universal among polling companies.
Can we expect a Reform 'Polling companies lie' campaign in true Trump style?
It is more about Yougov trying to poll tactical voting by asking how respondents will vote in their own constituency rather than who they want to win. This results in Reform being about five points lower than with other polling companies. If valid, and this is as important for betting as for politics, it shows many voters are willing to vote tactically to keep Reform out of power. (And by valid, I mean valid in the technical sense that a test or poll measures what it purports to measure.)
There's got to be more to it than that.
Imagine we have a FPTP election with 25 voters, 12 of whom support Reform, 10 of whom support Labour, and 3 of whom support the Greens. If they all vote for their first choice party, Reform will win. But if the Green voters hate Reform, and can be convinced the only way to stop them winning is to vote Labour, then the actual election result is Labour 13, Reform 12.
This changes the winner, and changes the Green and Labour vote shares. The only thing is doesn't do is change the Reform vote - they get 12 votes in both scenarios.
So I'm not sure what's going on in the You gov polling, but the one thing that can't be happening is for them to be reducing the Ref voteshare because they are picking up more accurately on anti-Reform tactical voting.
It's worth noting that if my scenario above happens in two constituencies, but with the Labour and Green vote shares and tactical squeeze the other way round in one of them, an opinion poll of both seats asking about tactical voting will have the same polling numbers as without asking who the voters support - you'll have 22 Reform, 13 Labour, 13 Green across the two seats either way, it's just that tactical voting alters the outcome from 2 Reform wins to 1 Labour win and 1 Green win.
It's possible that they are picking up anti-left tactical voting, where Reform voters are voting Tory to avoid Labour and the Greens winning - that would reduce the Reform vote and inflate the Tory vote, but you'd expect that to be at least partially cancelled out by Tories voting tactical for Reform.
And all of this assumes voters know who they'll vote for tactically anyway. My current constituency was traditionally a Tory/Lab marginal, which went Labour by a large margin at the last election, having been a narrow Tory win in 2019. I've literally no idea who best to vote for if I wanted to vote tactically left or right - I could imagine it plausibly being won by any one of Reform, Tory, Labour or Green, probably on about 25% of the vote. About the only party unlikely to get a look in are the LibDems. I don't know what I'll actually do on polling day, but if I was asked how I'd vote for an option poll right now, I'd just say Reform.
Of course you are entirely right, it’s an expensive pipe dream! However in brute political terms it taps into the disgust a clear majority of Brits feel toward Trump, which is quite shrewd. And by the way less of the snide LD’s in government nonsense; we did a good job in coalition, stabilised the economy, kept the hard left and right nutters out of government and delivered 70% of our manifesto. All the while with the Mail on the right and Guardian on the left lambasting us for being useless. Hmm, if that was the case we seemed to upset both of them an awful lot; perhaps because we stopped their respective nonsense….
Comments
Not saying that is what’s happening. But it’s a possibility.
NEW THREAD
Imagine we have a FPTP election with 25 voters, 12 of whom support Reform, 10 of whom support Labour, and 3 of whom support the Greens. If they all vote for their first choice party, Reform will win.
But if the Green voters hate Reform, and can be convinced the only way to stop them winning is to vote Labour, then the actual election result is Labour 13, Reform 12.
This changes the winner, and changes the Green and Labour vote shares. The only thing is doesn't do is change the Reform vote - they get 12 votes in both scenarios.
So I'm not sure what's going on in the You gov polling, but the one thing that can't be happening is for them to be reducing the Ref voteshare because they are picking up more accurately on anti-Reform tactical voting.
It's worth noting that if my scenario above happens in two constituencies, but with the Labour and Green vote shares and tactical squeeze the other way round in one of them, an opinion poll of both seats asking about tactical voting will have the same polling numbers as without asking who the voters support - you'll have 22 Reform, 13 Labour, 13 Green across the two seats either way, it's just that tactical voting alters the outcome from 2 Reform wins to 1 Labour win and 1 Green win.
It's possible that they are picking up anti-left tactical voting, where Reform voters are voting Tory to avoid Labour and the Greens winning - that would reduce the Reform vote and inflate the Tory vote, but you'd expect that to be at least partially cancelled out by Tories voting tactical for Reform.
And all of this assumes voters know who they'll vote for tactically anyway. My current constituency was traditionally a Tory/Lab marginal, which went Labour by a large margin at the last election, having been a narrow Tory win in 2019. I've literally no idea who best to vote for if I wanted to vote tactically left or right - I could imagine it plausibly being won by any one of Reform, Tory, Labour or Green, probably on about 25% of the vote. About the only party unlikely to get a look in are the LibDems. I don't know what I'll actually do on polling day, but if I was asked how I'd vote for an option poll right now, I'd just say Reform.
Halcyon days 😀