EU elections

Got anything else on your mind that isn't about the Warriors? If you do, this is the place to post.
Posts: 14426
Joined: Mon Sep 30, 2002 5:32 pm

EU elections

Post by DaveO » Tue Apr 23, 2019 6:15 pm

As it looks like we might be having these I thought I'd start a thread about them so people can cheer them on or vent their rage one way or the other.

One of the curiosities of these elections is they use a form of proportional representation (PR) known as the D'Hont method. What you do with this is divide up the number of seats based on a formula, not a simple percentage as a pure PR might suggest you do.

You put the seats on a list and they get allocated like this for a list of five seats assuming five parties in the election get the following share of the vote: P1 gets 30%, P2 gets 25%, P3 gets 20%, P4 gets 10% and P5 gets 5%.

P1 got the most votes so gets the first seat. Then their share of the vote is divided by 1+number of seats won, so they are now on 15%. 1 seat down 4 to go.

Do the the same again which means P2 on 25% is in the lead so gets the next seat. Divide their vote by 1+ number of seats, so 12.5%. 2 seats down 3 to go.

Do the the same again which means P3 on 20% is now in the lead so gets the next seat. Divide their vote by 1+ number of seats, so 10%. 3 seats down 2 to go.

Doing the the same again puts P1 back in the lead. They are on 15% having won the first seat and 15% is larger than what remains for P2 & P3 and is larger than what P4 and P5 got originally. So P1 gets another seat. Divide their vote by 1+ number of seats, so 10% left. 4 seats down 1 to go.

Doing the same again leaves us with P2 on 12.5% which is more than P1 and P3 have left and more than P4 and P5 had originally anyway so the final seat goes to P2.

Final result P1, two seats, P2, two seats, P3 one seat and P4 and P5 miss out.

The UK has 73 seats to allocate and earlier today a poll had the following projections (ignoring SNP/Plaid)
  • Brexit Party, percent of vote 23%
  • Labour Party, percent of vote 22%
  • Cons 15, percent of vote 17%
  • Green, percent of vote 10%
  • Lib Dems, percent of vote 9%
  • ChangeUK, percent of vote 8%
  • UKIP, percent of vote 6%
The UK complicates things by having regional lists not one large national list and there are different numbers of seats on each list based on population. The largest list has 10 seats on it (south east) and the two smallest 3 (East of England and Northern Ireland).

Given the different number of seats on the various lists that would translate into results of:
  • Brexit 23
  • Labour 21
  • Cons 15
  • Green 8
  • Lib Dems 5
  • ChangeUK 1
  • UKIP 0
Change UK gain their solitary seat in the South East which has the largest regional list of 10 seats to allocate. All the other lists have fewer seats so they are squeezed out.

These results also show how the regional lists and also how the opposition vote splitting distorts the result.

The confirmed anti-Brexit parties, Greens, Lib Dems and ChangeUK total 27% of the vote compared to, for example, the Conservatives who are on 17% yet the Conservatives end up with 15 seats and Greens, Lib Dems and Change a total of only 14 despite attracting 10% more of the vote.

UKIP are totally squeezed out on 6% of the vote.

If we didn't have regional lists but just one large one, the results would be
  • Brexit 17
  • Labour 17
  • Cons 13
  • Green 7
  • Lib Dems 6
  • ChangeUK 6
  • UKIP 4
  • Other 3
Note the poll data is ignoring SNP/Plaid but I have seen other data with the figure of 4% of the vote so "other" above is about right.

Of course not all regions will split the same way with Scotland featuring a strong SNP vote but I think what this shows is tactical voting still ought to be in play depending which region you are in as that dictates the number of seats up for grabs and how the polls look regionally is what is going to be important.

For example the East of England only has three seats to allocate so on current polling it's going to be Brexit, Labour and Conservatives with one seat each.

The North West Region has 8 seats to allocate and the current polling says the result would be:
  • Brexit 2
  • Labour 2
  • Cons 2
  • Green 1
  • Lib Dems 1
  • ChangeUK 0
  • UKIP 0
  • Other 0
So unless the percentages change substantially if you are pro-Brexit voting UKIP is a waste of time and if you are anti-Brexit voting ChangeUK is equally pointless. If you add all of UKIP to Brexit and all of ChangeUK to Green they each gain a seat at the expense of the Cons and Lib Dems. You tend to rob Peter to pay Paul so unless either the Tory and/or labour vote collapses the outcome as far as pro and anti Brexit is concerned will be broadly similar.