Paul's Pictoral Parliament Predictor

[ciphergoth]

The position on the diagram indicates the proportion of votes cast; the colour of the circle indicates composition of the resulting parliament. The hollow circles indicate hung parliaments, and the ones with white dots indicate a 3/5ths majority.

The white triangle indicates the region where no party has a majority of the votes, with the subdivisions showing which has plurality, so the point where the three lines meet is the point at which all three have equally many votes.

Updated to use labels Three points on the diagram are labelled; one marks the results of the 2005 general election from which the entire diagram was extrapolated, and two of them mark results from recent polling data and what sort of Commons this model predicts from that data.

The means of extrapolation is pretty ropey - I'd be interested to know where to look to find better ones.

This is of course a work in progress.

Comments

[dennyd.livejournal.com]

Is the resemblance to a woman's bikini region my own sordid invention, or a cunning part of your plan? :)

[barakta]

No, that was my initial thoughts too!

[softfruit.livejournal.com]

I hang my head in shame that I didn't see the link between a hung parliament and knickers until I read this thread!

[barakta]

Sounds like you need some pervery rebootage!

[postmodern-minx.livejournal.com]

So clever, i really enjoyed your historical version that includes a time axis also.

pm

[softfruit.livejournal.com]

My own bias of course means I'm more interested in the spread of seats within a parliament rather than just which is the largest party. That said, I do like how your triangle neatly shows how much the current system favours Labour - to the degree where a Lib Dem national vote share over 50% produces an overall majority for Labour (or even for a couple of dots with a majority for the Conservatives).

Would I be right in thinking that a line from the last result to the current polling prediction would map the currently indicated swing - and how much of that swing would lead to a hung parliament rather than majority?

I'd also be interested to see previous GE results plotted on the chart, for all that boundary changes have subtly altered the figures. Speaking of which - are your calculations based on the 2005 seats or the predicted 2009/10?

[ciphergoth.livejournal.com]

My own bias of course means I'm more interested in the spread of seats within a parliament rather than just which is the largest party.

Mmm - if only I could think of a way of illustrating it...

[silverclear.livejournal.com]

I really want to understand what this shows, it looks fascinating, but I don't know what it's trying to represent! Can you explain first of all what one dot means? And what does the whole picture represent? Am I being dumb?!

[softfruit.livejournal.com]

I'll have a go and hope Paul can correct me, which will help as I'm not totally sure on the odd point :D

Starting with "why?"

You've probably seen traditional election swingometers. Big arrow pointing at the floor, swings left or right showing how many % the vote has changed since last time, and along the arc the arrow would point to you mark all the seats in the election so you can see that if it swings *yay* far in that direction, these seats change from one party to another.

That works pretty well if you have a 2 party contest going on, so until 1997 it worked quite well for UK elections. (let's not get into scottish parliament elections here, it will get way too confusing)

The trouble is, there's now about 10% of the seats and 20% of the vote going to the LibDems, and that confuses matters. Because at the last election, the Tories stayed pretty much as they'd been before in share of the vote, but the shift between people voting Labour and LibDem meant that all sorts of seats changed hands. And on that left-right swingometer you just can't show what the heck is going on, because if you like the arrow isn't moving left or right but on a different axis - out of the page towards you.

Paul's diagram seeks to tackle that problem.

That's the "why", now the "what does it mean?"

The square has a left-right axis (x) and up-down axis (y).

How far up the y axis you go tells you the Lib Dem share of the vote. At the bottom it's 0%. At the horizontal white line, that's the Lib Dems getting 50%.

How far across the x axis you go from the middle shows the relative strengths of the Labour and Tory votes. The vertical line in the middle is the point where they both get the same share of the total votes cast - but it's NOT 50% except at the bottom where the Lib Dems get 0%.* But to the left of that line it means more Labour votes than Tory, and vice versa.

So- "how to read it"

Start at the left hand circle. This is the result last time: Lib Dems about 23% so halfway between the bottom and that horizontal white line; Labour about 5% ahead of the Tories so to the left of the middle line.

Now, let us assume the Lib Dems stayed on the same vote share and all the swing was between Labour and Tory. That would mean you move from the dot circled as the 2005 result to the left or right - this shows that even if Lab and Con got the same number of votes, it's a solid red dot, Lab would have an overall majority. You have to go quite a way to the right of the middle line for the Conservatives to actually have an overall majority, there's quite a wide range of possible election outcomes where you'd have a hung parliament.

If instead the Lab-Con split stayed the same and the swing was from Lab to Lib Dem, you move up the chart. As you can see you can move much further that way before you get to a hung parliament (black dot).

So, given an opinion poll, you can look at this chart and say: it means a swing yay far across to the Tories and yay far away from the Lib Dems which puts us about... here... and see what colour dot you have to immediately see which party would have the most seats (colour of circle), whether it would be enough to have an overall majority (is circle coloured in rather than black), and if it's a substantial majority like Blair had in 97 - 2005 (white circle).

That about right Paul? Anything I've missed?


* this is a bit of a simplification, as Paul's model doesn't show about 10% of the vote going to Other parties - Plaid, SNP, Greens, BNP, UKIP etc.

[softfruit.livejournal.com]

I can see one small error in this description, but can't work out how to describe it easily without a diagram! Here goes:

A straight Lab-LibDem swing movement on the chart wouldn't mean going directly upwards on the y axis. It would be moving in parallel with the line that runs from the bottom middle of the picture to the right hand side, so the movement would be up-and-right-a-bit or down-and-left-a-bit.

[silverclear.livejournal.com]

Thank you very much! This makes much more sense now! I hadn't understood that the coloured spots were showing 'potential' result points and that the white on top were the actual data.

[ciphergoth.livejournal.com]

No apology needed - this is hard to understand and I didn't take the time to explain it, but softfruit has it just right!

[sibelian.livejournal.com]

These days, of course, if you showed this diagram to any ordinary member of the general public they'd run away in fear and swear never to vote again because it's too complicated.

[ciphergoth.livejournal.com]

It's maddening. People are very resistant to anything that complicates the system by which we get a House of Commons. So instead of a complex system that produces simple and predictable results, we have a simple system with complex and often stupid results.