Theresa May resigns

[AtW]

<placeholder>

[WTFH]

<placeholder>

She promised she wouldn't.

Oh, no, that was the last Tory PM.
She only promised not to have a snap general election.

...but remember, lying about that is nothing compared to making a mistake about some figures, then apologising and correcting yourself.
Lying is OK if it's Tory lying.

[Paddy]

She promised she wouldn't.

Oh, no, that was the last Tory PM.
She only promised not to have a snap general election.

...but remember, lying about that is nothing compared to making a mistake about some figures, then apologising and correcting yourself.
Lying is OK if it's Tory lying.

May comes across as the type of person who will never admit that they are wrong and will blame everyone else
for their own mistakes and wrong doing. Furthermore, her husband is making millions out of her position and influence. If she goes, it will be due to a mass revolt in the Tory Party. A week after the election?

[AtW]

The Tory Party is already revolting, but a week after the election they will be rebelling...

[Bee]

Was this your dream last night?

[BlasterBates]

Poll gives Tories 1% lead

Tories go "nuclear"

[AtW]

ToryGraph begs to differ...

Exclusive Telegraph ORB poll: Conservatives&#39; last-minute climb as ‘Leavers’ losing faith in Labour and turning to the Tories

Poll's target audience the staff serving Barlays Brothers palace...

[Mordac]

She promised she wouldn't.

Oh, no, that was the last Tory PM.
She only promised not to have a snap general election.

...but remember, lying about that is nothing compared to making a mistake about some figures, then apologising and correcting yourself.
Lying is OK if it's Tory lying.

Unless you're Diane Abbott, and then it gets put down to an inability to deal with complicated numbers.
Or unless you're Gordon Brown, and then it gets put down to something very similar.

[jamesbrown]

Poll gives Tories 1% lead

Tories go "nuclear"

I really have no idea why anyone listens to these polling companies anymore. They are so fundamentally dependent on turnout filters and other fudge factors (turnout almost entirely explains the systematic differences between polling companies), and the past is now a questionable guide to the future on turnout. Take the ICM poll tonight from the Mirror/Indy, for example. The headline lead is 11% CON. Before the turnout filter, it's 2% CON. Some of these companies are going to look very stupid come Friday. The problem is that a different set of companies look stupid every time

If you really want decent polling, you need to pay for it (large, aggressively pursued/randomized, sample), and sacrifice the temporal resolution.

[AtW]

If you really want decent polling, you need to pay for it (large, aggressively pursued/randomized, sample), and sacrifice the temporal resolution.

All published polling is paid for

HTH