Local Council by-elections 20th January 2022

[jamesdoyle]

Jan 18, 2022 17:19:34 GMT andrew111 said:

Jan 18, 2022 16:39:32 GMT jamesdoyle said:

Yes, good point. I'll have to go back and look at the calculations.
My model takes account, in multiple-seat elections, of the variance in each party's vote, and that may have caused some weirdness here with the extreme range for Labour, against a single Tory candidate.
Tbf, I am feeling a bit that there should be an entirely different model for STV elections, but I'm not sure I can be bothered when there are so few to be considered.
If I can find a workable hybrid, I will!

ETA: Nope, it's not that! I've had a chance to have a quick look, and I somehow introduced an error into the calculation of some figures at the start of January. Will correct it later.

Well, we do know what to do with STV elections, you add up the first preferences for all the candidates for each party, and that is the vote for each party. It is actually much simpler than multiseat FPTP elections, where there is no "right" way to do it (top vote vs average vote). This is because in STV each person only has one vote (the clue being in the S)

I do know of that, but that doesn't help what I'm trying to do. But as you have fairly pointed out, and I have agreed, my current approach is unsatisfactory. I'll go back to the drawing board. Luckily, it looks like there aren't any more Scottish by elections in the near future so I've got some time to think about it.
I've corrected my spreadsheet error in the table posted earlier, which at least gives a 'better' picture.

[minionofmidas]

Jan 18, 2022 18:50:59 GMT jamesdoyle said:

Jan 18, 2022 18:13:17 GMT minionofmidas said:

Anyways I don't see how a party can score above 0 on seats they don't contest. Even where a case can be made that they should have.

I'll say this (yet) again: the Importance Index is not an end-product; it is a by-product from an intermediate stage of calculating the good week/bad week index. The Importance Index includes as a factor how many seats a party has on a council and how close it is to have control of the council. If a party was a seat away from taking control of a council, how important would it be to them to win a by election for a vacant seat? And if they failed to stand a candidate, your view is that suddenly that by-election is unimportant to them?

Not so much my view as my deduction.
Unless they just messed up the nomination papers. Or are effectively supporting someone with a different label (as appears to be what's happening in Loughborough?) But as long as there's no way of accounting for that in the final stage...
Of course, a consistent pattern of a party standing fewer and fewer candidates, either nationally or in a specific region where they used to be active, tells you a lot. But I'm not sure that can be reduced to a single figure in this way - byelections are too rare and too randomly distributed for that.
Meanwhile there are hundreds, probably thousands, of wards in the UK that the LDs, the Greens, the non-locally-dominant of Labour and the Tories only contest occasionally, and where their failure to appear at a probably unwinnable byelection though they ran last time means nothing whatsoever. Even on councils where they regularly have representation. The Lindsey one in this set would seem to qualify.*

I find what you're trying to do here very interesting. Please by all means keep at it! But this is a feature that seems like a bug to me.


* edit: not at all, but it probably qualifies under 'supporting someone with a different party label'.

[andrew111]

Jan 18, 2022 18:10:46 GMT minionofmidas said:

Jan 18, 2022 17:19:34 GMT andrew111 said:

Well, we do know what to do with STV elections, you add up the first preferences for all the candidates for each party, and that is the vote for each party. It is actually much simpler than multiseat FPTP elections, where there is no "right" way to do it (top vote vs average vote). This is because in STV each person only has one vote (the clue being in the S)

there is a problem lurking in by-elections to seats won by other than the largest party though...

Well, the problem is that a "gain" is not really a gain, very often. It is an AV election where only one person is elected, rather than several. Still, counting up all the first preferences is a good first step at a baseline for prediction of AV, and looking at the STV transfers as done by ballot box Scotland gives a fair prediction of what would have happened in an AV election last time.

[timmullen1]

Here are Andrew’s Previews for this week:

www.britainelects.com/2022/01/20/previewing-the-council-by-elections-of-20-jan-2022/

[jm]

Are they all counting tonight?

[agbutler]

Jan 20, 2022 22:11:12 GMT jm said:

Are they all counting tonight?

England is, Scotland tomorrow.

[jm]

Byram and Brotherton

Con 215
Lab 207
Green 25

[carolus]

Selby:
Con 215, Lab 207, Green 25

[Davıd Boothroyd]

SELBY Byram and Brotherton

ASHTON, Georgina (Conservative Party Candidate) 215
BURTON, Matthew Keir (Labour Party) 207
WATERS, Cherry Elizabeth (The Green Party) 25

[Chris from Brum]

Jan 20, 2022 23:16:10 GMT Davıd Boothroyd said:

SELBY Byram and Brotherton

ASHTON, Georgina (Conservative Party Candidate) 215
BURTON, Matthew Keir (Labour Party) 207
WATERS, Cherry Elizabeth (The Green Party) 25

I guess that's technically Con gain from YP. And while Labour might fulminate against the Greens for splitting the vote, perhaps they might have wished us to stand, as we tend to attract disaffected Tories in a way that the Greens never can and which are more difficult for Labour

[jm]

Jan 20, 2022 23:15:54 GMT jm said:

Byram and Brotherton

Con 215
Lab 207
Green 25

%

Con 48.1 (+13.4)
Lab 46.3(+26.2)
Green 5.6

[Davıd Boothroyd]

Charnwood's verification for Loughborough Shelthorpe found 1,242 ballot papers so the turnout is 18.7%.

[phil156]

CON GAIN IN Selby

[minionofmidas]

First Tory gain of the year, holds are still beyond them though (if Labour can't gain here where will they ever?)

[jm]

A large swing to Labour and the highest Labour vote (as a %) that has ever been recorded in this ward. The Labour vote is actually higher than when we last held the ward in 2015. Bearing in mind that this is the kind of area that will have swung heavily to the Tories in 2019, not a bed result.

[Robert Waller]

We have some very good results-hounds on this forum, but I suspect that the Chapel St Leonards result is the type of contest which eludes early reporting (only one of the main parties etc) - so while it should have been easy to count quickly (2 candidates), congrats to anyone who can find out what happened!

[timmullen1]

[listener]

About 8 years ago, there was a town council by-election in East Lindsey, which was counted on Thursday night. The electoral staff took Friday as holiday in lieu, without leaving a record of the election result. I rang the Town Clerk and the Returning Officer's secretary, but neither of them were informed of the result and the elections office was closed for the day. I had to wait until Monday to get the result.

[phil156]

Jan 21, 2022 0:02:41 GMT listener said:

About 8 years ago, there was a town council by-election in East Lindsey, which was counted on Thursday night. The electoral staff took Friday as holiday in lieu, without leaving a record of the election result. I rang the Town Clerk and the Returning Officer's secretary, but neither of them were informed of the result and the elections office was closed for the day. I had to wait until Monday to get the result.

[timrollpickering]

There's a public group on Facebook for Chapel St Leonards. Here's the most recent relevant post:

www.facebook.com/groups/chapelstleonards/permalink/4628004543915736/

"Anyone know result of election today."