Local Council by elections 16th- 18th July

[Sibboleth]

Kun-koyd would be what we might think of as Correct Welsh, though that doesn't mean that this is the correct pronunciation because the usual rule there is 'whatever the locals say is right'.

[iain]

Jul 17, 2019 15:53:05 GMT Chris from Brum said:

Jul 17, 2019 15:47:29 GMT froome said:

From when i lived in Cardiff (which was a very long time ago now):

Cyn was pronounced somewhere between Kun and Kin, closer to the former.

Coed (Welsh for woodland) - oe was usually pronounced as one verb, similar to oy in English, though in Welsh you can sometimes hear the 'e' separately, as in 'oy-ee', so Koyd or Koy-eed.

"Y" is the only Welsh letter that has two different values depending on context.  In the word ysbyty (hospital) it is a schwa in the first two syllables and "i" in the third, thus "usbutty" to English ears.  I would have hazarded "Kincoyd" myself, given that "y" is often an "i" sound before an "n", but locals will know better.

I believe the usual rule is that ‘y’ is ‘uh’ normally, but ‘ih’ if it’s in the last syllable of a word - which also works in your ysbyty example.

[tonygreaves]

Kin Co-ed

So the first language Welsh speaker in our office says. But he's a North Walian!

[gwynthegriff]

Jul 17, 2019 20:54:03 GMT tonygreaves said:

Kin Co-ed

So the first language Welsh speaker in our office says. But he's a North Walian!

Certainly not co-ed in the way it's used to abbreviate co-educational.

[lbarnes]

Jul 17, 2019 8:30:14 GMT Davıd Boothroyd said:

Jul 17, 2019 8:16:59 GMT Chris from Brum said:

Middleenglander (above) would beg to disagree with you.

It is statistical rot, and sheer stupidity, to regard these two things as identical:

1) A party which polls 20% in one election and 30% in the next election
2) A party which did not previously have a candidate polling 10% in an election.

And I don't care who is doing it. It's bunk. Rubbish.

It really isn't.  There may be caveats to attach but it's a mathematical swing.


Your ignorance grows.

[Adam in Stroud]

Jul 17, 2019 22:09:02 GMT lbarnes said:

Jul 17, 2019 8:30:14 GMT Davıd Boothroyd said:

It is statistical rot, and sheer stupidity, to regard these two things as identical:

1) A party which polls 20% in one election and 30% in the next election
2) A party which did not previously have a candidate polling 10% in an election.

And I don't care who is doing it. It's bunk. Rubbish.

It really isn't.  There may be caveats to attach but it's a mathematical swing.


Your ignorance grows.

Aaargh!  Please let this lie.

[finsobruce]

Jul 17, 2019 22:34:24 GMT Adam in Stroud said:

Jul 17, 2019 22:09:02 GMT lbarnes said:

It really isn't.  There may be caveats to attach but it's a mathematical swing.


Your ignorance grows.

Aaargh!  Please let this lie.

Agreed. We need to move away from this now, please.

[Chris from Brum]

Jul 17, 2019 16:14:12 GMT iain said:

Jul 17, 2019 15:53:05 GMT Chris from Brum said:

"Y" is the only Welsh letter that has two different values depending on context.  In the word ysbyty (hospital) it is a schwa in the first two syllables and "i" in the third, thus "usbutty" to English ears.  I would have hazarded "Kincoyd" myself, given that "y" is often an "i" sound before an "n", but locals will know better.

I believe the usual rule is that ‘y’ is ‘uh’ normally, but ‘ih’ if it’s in the last syllable of a word - which also works in your ysbyty example.

Welsh placenames can be inconsistent in how many words they consist of.  I have a friend who lives near Ebbw Vale - his village is called Waunlwyd, Waun-lwyd or Waun Lwyd according to taste.  A certain village on Anglesey can be Llanfairpwllgwyngyll or Llanfair Pwllgwyngyll (or Llanfairpwll, which is the station name, or Llanfair PG, or Llanfairpwll...gogogoch, but the extended version is for tourists only).  I don't know whether Cyncoed could alternatively be Cyn Coed, but if it can, then the one syllable of the first part would also be the last syllable.

[MeirionGwril]

Re: Cyncoed. Last word on this. It's kin-coyd in whatever dialect. The precise value of the 'oe' may vary slightly from place to place, but it's kin NOT kun. As a prefix (meaning something like 'before', 'in front of') the 'y' in 'Cyn' is never the central vowel schwa (or 'uh') but always the front vowel, so it avoids the 'ysbyty' rule because it is a prefix.
Meirion Gwril (Welsh speaker, former Cardiff resident, phonetician).

[Chris from Brum]

... and here are Andrew's Previews for 18th July.

[carlton43]

Jul 17, 2019 22:34:24 GMT Adam in Stroud said:

Jul 17, 2019 22:09:02 GMT lbarnes said:

It really isn't.  There may be caveats to attach but it's a mathematical swing.


Your ignorance grows.

Aaargh!  Please let this lie.

These matters are not pure maths but an understood convention based on an abstraction. So they are rules based just like the conduct of cricket and the scoring system in tennis.

First there is the movement of actual votes.....

People who had not voted now voting
People who had voted and now not voting
That is differential TO
And differential TO within parties.

Then we have movement of votes...

People who voted A now voting B
People who voted B now voting A
And many other such movements
That is inter-party voter churn.

Then we have the classic understood 'Swing' which is solely calculated on the apparent movement of votes between the two majors of Labour and Conservative where it is an established given that the other parties are not relevant even if they have a higher proportion of the total TO vote to either of the majors or even both of the majors.

When it is contended that there is a swing from Conservative to LD it is really an assumed movement of votes that might be a very complex churn in many directions at once. It could comprise an actual series of movements that could be expressed as....

Non voter to Con 1.2%
LD to Con 0.2%
Lab to Con 1.1%
Green to Con 0.0%
BP to Con 0.0%

Non voter to LD 8.8%
Con to LD 1.3%
Lab to LD 7.8%
Green to LD 4.8%
BP to LD 0.0%

Non voter to Lab 0.2%
Con to Lab 0.5%
LD to Lab 0.3%
Green to Lab 0.7%
BP to Lab 0.0%

Non voter to Green 0.5%
Con to Green 0.1%
LD to Green 0.2%
Lab to Green 2.8%
BP to Green 0.0%

Non voter to BP 7.3%
Con to BP 18.9%
LD to BP 0.6%
Lab to BP 8.4%
Green to BP 0.0%

This shows the actual churn from reading the minds of all voters and relying upon total accuracy of those minds as a given for this purpose.

We can only compute the BP actual result here as it is the new party standing for the first time and thus losing nothing and with no back record for the base comparison. It scores 35.2% of the total vote and may be in for a shout for the win?

Let us assume this had been a firm Con seat and we see it has lost 20.8% and only gained 2.5% resulting in an 18.3% net loss.

The LD have seen gains of 22.7% and lost only 1.3% resulting in a net gain of 21.4%.
From those 'raws' we may assume that the Cons lost the seat and that the LDs probably won the seat? No matter. As regards the movement between Con and LD it can now readily be seen to be only a net 1.1% (1.3-0.2) but in these thread columns the movement will be computed on the relative change in the two new totals from the previous election results. That will tend to show an apparent change very much larger than 1.1%. The reason for that being the massive bleed of votes from Conservative to the BP which caused the Conservative total to materially drop but not on any swing to the LDs.


Thus we should not use swing to be employed as a term between any two parties other than Conservative and Labour where that movement between those two parties is always a true reflection of the movement of votes between them, in a manner that is hardly ever true when applied between either of them and any other party or between any two other parties.

[yellowperil]

Jul 18, 2019 8:51:38 GMT carlton43 said:

Jul 17, 2019 22:34:24 GMT Adam in Stroud said:

Aaargh!  Please let this lie.

These matters are not pure maths but an understood convention based on an abstraction. So they are rules based just like the conduct of cricket and the scoring system in tennis.

First there is the movement of actual votes.....

People who had not voted now voting
People who had voted and now not voting
That is differential TO
And differential TO within parties.

Then we have movement of votes...

People who voted A now voting B
People who voted B now voting A
And many other such movements
That is inter-party voter churn.

Then we have the classic understood 'Swing' which is solely calculated on the apparent movement of votes between the two majors of Labour and Conservative where it is an established given that the other parties are not relevant even if they have a higher proportion of the total TO vote to either of the majors or even both of the majors.

When it is contended that there is a swing from Conservative to LD it is really an assumed movement of votes that might be a very complex churn in many directions at once. It could comprise an actual series of movements that could be expressed as....

Non voter to Con 1.2%
LD to Con 0.2%
Lab to Con 1.1%
Green to Con 0.0%
BP to Con 0.0%

Non voter to LD 8.8%
Con to LD 1.3%
Lab to LD 7.8%
Green to LD 4.8%
BP to LD 0.0%

Non voter to Lab 0.2%
Con to Lab 0.5%
LD to Lab 0.3%
Green to Lab 0.7%
BP to Lab 0.0%

Non voter to Green 0.5%
Con to Green 0.1%
LD to Green 0.2%
Lab to Green 2.8%
BP to Green 0.0%

Non voter to BP 7.3%
Con to BP 18.9%
LD to BP 0.6%
Lab to BP 8.4%
Green to BP 0.0%

This shows the actual churn from reading the minds of all voters and relying upon total accuracy of those minds as a given for this purpose.

We can only compute the BP actual result here as it is the new party standing for the first time and thus losing nothing and with no back record for the base comparison. It scores 35.2% of the total vote and may be in for a shout for the win?

Let us assume this had been a firm Con seat and we see it has lost 20.8% and only gained 2.5% resulting in an 18.3% net loss.

The LD have seen gains of 22.7% and lost only 1.3% resulting in a net gain of 21.4%.
From those 'raws' we may assume that the Cons lost the seat and that the LDs probably won the seat? No matter. As regards the movement between Con and LD it can now readily be seen to be only a net 1.1% (1.3-0.2) but in these thread columns the movement will be computed on the relative change in the two new totals from the previous election results. That will tend to show an apparent change very much larger than 1.1%. The reason for that being the massive bleed of votes from Conservative to the BP which caused the Conservative total to materially drop but not on any swing to the LDs.


Thus we should not use swing to be employed as a term between any two parties other than Conservative and Labour where that movement between those two parties is always a true reflection of the movement of votes between them, in a manner that is hardly ever true when applied between either of them and any other party or between any two other parties.

Okay we have been asked by the mods, and others, to let this lie, but with characteristic arrogance this is ignored with a long post which seems to me total claptrap.So without going in to the detail of it, the essence of it seems to be that real swing is only between Lab and Con because they are the  Only True Parties. Well, I'm sorry but if that was ever true it certainly isn't true at the moment, and while it is possible that the present world of multi-party  fluidity with 4 parties at roughly 20% each may be a very short lived phenomenon it could well be that if we get back to the sort of duopoly that FPTP tends to produce it is quite possible that one or maybe both of the old duopoly will be missing, Calculations about what constitutes swing may the have to be rethought from first principles.

[Pete Whitehead]

To be fair its marginally more interesting than how to pronounce silly Welsh place names

[timmullen1]

Jul 18, 2019 9:48:42 GMT Pete Whitehead said:

To be fair its marginally more interesting than how to pronounce silly Welsh place names

Really?

[carlton43]

Jul 18, 2019 9:19:53 GMT yellowperil said:

Jul 18, 2019 8:51:38 GMT carlton43 said:

These matters are not pure maths but an understood convention based on an abstraction. So they are rules based just like the conduct of cricket and the scoring system in tennis.

First there is the movement of actual votes.....

People who had not voted now voting
People who had voted and now not voting
That is differential TO
And differential TO within parties.

Then we have movement of votes...

People who voted A now voting B
People who voted B now voting A
And many other such movements
That is inter-party voter churn.

Then we have the classic understood 'Swing' which is solely calculated on the apparent movement of votes between the two majors of Labour and Conservative where it is an established given that the other parties are not relevant even if they have a higher proportion of the total TO vote to either of the majors or even both of the majors.

When it is contended that there is a swing from Conservative to LD it is really an assumed movement of votes that might be a very complex churn in many directions at once. It could comprise an actual series of movements that could be expressed as....

Non voter to Con 1.2%
LD to Con 0.2%
Lab to Con 1.1%
Green to Con 0.0%
BP to Con 0.0%

Non voter to LD 8.8%
Con to LD 1.3%
Lab to LD 7.8%
Green to LD 4.8%
BP to LD 0.0%

Non voter to Lab 0.2%
Con to Lab 0.5%
LD to Lab 0.3%
Green to Lab 0.7%
BP to Lab 0.0%

Non voter to Green 0.5%
Con to Green 0.1%
LD to Green 0.2%
Lab to Green 2.8%
BP to Green 0.0%

Non voter to BP 7.3%
Con to BP 18.9%
LD to BP 0.6%
Lab to BP 8.4%
Green to BP 0.0%

This shows the actual churn from reading the minds of all voters and relying upon total accuracy of those minds as a given for this purpose.

We can only compute the BP actual result here as it is the new party standing for the first time and thus losing nothing and with no back record for the base comparison. It scores 35.2% of the total vote and may be in for a shout for the win?

Let us assume this had been a firm Con seat and we see it has lost 20.8% and only gained 2.5% resulting in an 18.3% net loss.

The LD have seen gains of 22.7% and lost only 1.3% resulting in a net gain of 21.4%.
From those 'raws' we may assume that the Cons lost the seat and that the LDs probably won the seat? No matter. As regards the movement between Con and LD it can now readily be seen to be only a net 1.1% (1.3-0.2) but in these thread columns the movement will be computed on the relative change in the two new totals from the previous election results. That will tend to show an apparent change very much larger than 1.1%. The reason for that being the massive bleed of votes from Conservative to the BP which caused the Conservative total to materially drop but not on any swing to the LDs.


Thus we should not use swing to be employed as a term between any two parties other than Conservative and Labour where that movement between those two parties is always a true reflection of the movement of votes between them, in a manner that is hardly ever true when applied between either of them and any other party or between any two other parties.

Okay we have been asked by the mods, and others, to let this lie, but with characteristic arrogance this is ignored with a long post which seems to me total claptrap.So without going in to the detail of it, the essence of it seems to be that real swing is only between Lab and Con because they are the  Only True Parties. Well, I'm sorry but if that was ever true it certainly isn't true at the moment, and while it is possible that the present world of multi-party  fluidity with 4 parties at roughly 20% each may be a very short lived phenomenon it could well be that if we get back to the sort of duopoly that FPTP tends to produce it is quite possible that one or maybe both of the old duopoly will be missing, Calculations about what constitutes swing may the have to be rethought from first principles.

ITEM  I had not seen the request by the Mods when I posted this and it was not arrogance and it was not a deliberate affront. I don't do that and have been very supportive to the Admin ever since the change. That was a low blow and quite unfair of you.

ITEM There are differences of opinon on the nature, the meaning and the structure of voter movement and how we may effectively represent it in records and in the definition of the word 'Swing'. It is again unkind and unhelpful of you to categorise my post as claptrap. Read it again. It is not an assertive position but my attempt to explain why referring to 'A Swing between Conservative and LD' is at best only a measure and a form of ratio that may well hide more than it states.

To say there was a swing of 16.9% from LD to Conservative (or the reverse) when in fact there was no such movement at all between those named parties is for me a signal mistake as it implies and states an untruth unless they were the only two parties on each occasion when the staement is a complete truth.

If one party loses a lot od votes to another party or other parties or to not voting at all, permitting a third party to come good in comparison, despite there being little or no actual transfer of votes between those two parties, it does make the word 'Swing' seem inappropriate. Yes that is an opinion but one I think and I hope is shared by many others.

This is the sort of discussion I thought was the original and main purpose of this Forum? I cannot see why it is so contentious and why it cannot be contested without accusations and personal abuse?

[Pete Whitehead]

Jul 18, 2019 9:50:30 GMT timmullen1 said:

Jul 18, 2019 9:48:42 GMT Pete Whitehead said:

To be fair its marginally more interesting than how to pronounce silly Welsh place names

Really?

Well yes - this is an election discussion forum. Discussing swings and such like are our meat and drink. That said I couldn't be botehred to read beyond about the third line of carlton's post and this particular discussion is getting old, but it is surely more broadly an issue that is of interest to all on here. How to pronounce a place name or word orally when in all likelihood one is only ever going to have to write the word, is by contrast of very limited relevance or interest

[yellowperil]

I have no desire to get back into prolonged debate with carlton43 or indeed any one else on this when I may be close to leaving the forum; as far as Carlton is concerned our world view is so far apart there is little point in continuing this debate. All I will say is that I totally accept that any understanding of swing as a concept needs to be based on accepting churn as a vital part of the story. Even if you are interested only in swing between Con and Lab, a lot of that swing would involve churn involving what you would no doubt think of as minor parties. I see swing as the averaging out of all the complexities of churn, but if that as a concept is beyond you so be it.

[Deleted]

Jul 18, 2019 10:11:21 GMT yellowperil said:

I have no desire to get back into prolonged debate with carlton43 or indeed any one else on this when I may be close to leaving the forum; as far as Carlton is concerned our world view is so far apart there is little point in continuing this debate. All I will say is that I totally accept that any understanding of swing as a concept needs to be based on accepting churn as a vital part of the story. Even if you are interested only in swing between Con and Lab, a lot of that swing would involve churn involving what you would no doubt think of as minor parties. I see swing as the averaging out of all the complexities of churn, but if that as a concept is beyond you so be it.

I know you have other matters on your mind, but I do hope you are not thinking of leaving over this spat, which is even more incomprehensibly overblown than most. Log off for a couple of days and see how you feel then.

[carlton43]

Jul 18, 2019 10:11:21 GMT yellowperil said:

I have no desire to get back into prolonged debate with carlton43 or indeed any one else on this when I may be close to leaving the forum; as far as Carlton is concerned our world view is so far apart there is little point in continuing this debate. All I will say is that I totally accept that any understanding of swing as a concept needs to be based on accepting churn as a vital part of the story. Even if you are interested only in swing between Con and Lab, a lot of that swing would involve churn involving what you would no doubt think of as minor parties. I see swing as the averaging out of all the complexities of churn, but if that as a concept is beyond you so be it.

I am attempting to convey that the word swing has no point except with Conservative-Labour for these sound psephoogical reasons

For decades only those two parties have been in a position to form a government.
They are the only parties to fight all seats (except NI) in every GE since 1945 (rare exceptions).
They are often the only serious contenders in a seat.
Historically they are the measure for the ebb and flow from right to left in the nation.

It is not because I disparage other parties or don't think they count or matter. Just at present the BP has the potential to cause my party massive damage perhaps without winning many seats itself.  I know that 'Classic Swing' has its limitations, but it still has a purpose and a primary interest. If either of the majors implodes in discord and stops contesting all seats then that classic measure will be over.

[yellowperil]

Jul 18, 2019 10:18:28 GMT @goodoldcause2 said:

Jul 18, 2019 10:11:21 GMT yellowperil said:

I have no desire to get back into prolonged debate with carlton43 or indeed any one else on this when I may be close to leaving the forum; as far as Carlton is concerned our world view is so far apart there is little point in continuing this debate. All I will say is that I totally accept that any understanding of swing as a concept needs to be based on accepting churn as a vital part of the story. Even if you are interested only in swing between Con and Lab, a lot of that swing would involve churn involving what you would no doubt think of as minor parties. I see swing as the averaging out of all the complexities of churn, but if that as a concept is beyond you so be it.

I know you have other matters on your mind, but I do hope you are not thinking of leaving over this spat, which is even more incomprehensibly overblown than most. Log off for a couple of days and see how you feel then.

No, just to clarify my little aside, then - no, I would not leave over a some overblown spat- I'm not that precious!  My point rather was that I may be unable to continue shortly depending on how things turn out in the next week or two, so had no desire to end it in some undignified dispute over nothing of significance.That's all.