European elections - how does it all work

Screenshot of D'Hondt Method election result calculator

EU elections are now almost certain to happen.

They use a version of Proportional Representation known as the D'Hondt Method.
I won't try and re-explain it here, as there are already various good explanations available such as this one.

I've made a google spreadsheet partly based on some other ones I found online, to let people play around with different voting patterns and see how it affects the results.

I created it to test various questions people were asking on social media:


  • If we have say 4 new remain parties, does it split the vote?
  • If UKIP and Farage's Brexit party pick up equal votes, does this affect their total seats or the remain seats?
  • If all remain voters vote Labour do they get a better allocation of seats?
  • Etc...

(The above are merely examples I've seen people ask - don't read anything into them as being suggestions of things to do or not do).

Anyone can view the file, but to change any of the data, it will ask you to make a copy of it to your own google account.

I've already included the 2014 EU election data for the UK except Northern Ireland (which uses a completely different system).

You only need to edit the sections of the document marked in yellow.

To repopulate the sheet with sample data, use the light blue pull-down menu and pick a region of the UK. It may take a couple of seconds to complete as it is not as responsive as a spreadsheet in Excel.

I've assumed that all parties have enough candidates to fill all seats (because looking at 2014, all parties that were remotely likely to get any seats fulfilled this criteria).

The percentage of votes received is colour coded - green high, red low. Ditto for the number of seats won by the party.

I've done a comparison of share of seats and share of votes. Typically, the biggest discrepancies are for the party that picks up the last seat and the one that would have picked up the next seat if one was still available.

I've also looked at the average number of votes parties needed to gain seats, as well as the maximum and minimum.

Below are some thoughts this I've had in response to queries from people and my own testing of some different scenarios.

Some of these are Berkshire related - You'll need to check the figures for your own area if you are located elsewhere.

The system is fairly proportional, but favours larger to mid sized parties. The tiny ones don't get a look in.

The bulk of the difference last time (from proportion of votes received to proportion of seats allocated) was from the last seat taken and the one that would have been taken after this if there was another available - and it's very hard to predict which those will be.

Across all the parties that won seats in SE England, the average difference from votes allocated to seats allocated is 2.5% - so they actually get a slightly better vote allocation overall than they would if the allocations were absolutely proportional.

Far more important in boosting the remain vote will be turnout.

The figures below are from 2014:

Area Turnout
Bracknell 33%
RBWM 35%
Reading 33%
Slough 37%
West Berkshire 35%
Wokingham 39%

We need to encourage Remainers to get out and vote.

Based on demographics, the UKIP voters (on average older) are also the group who votes most reliably, giving them a boost that does not reflect the reality.

One thing that helps us (a bit) is the UKIP splitters. Last time round they made up about half the options (if you include other rival right wing groups like BNP). This time it's likely to be more if anything as so many of their 2014 MEP intake have flounced off in a huff for various reasons (they started with 24, they now have 7).

We also need to get people to register to vote and in many cases to make them aware that they are able to vote (for EU citizens and Commonwealth citizens particularly).

In Slough, it's estimated that only 81% of those eligible to be on the electoral register are. In some other areas it's better, but there is still a resource there to be tapped.

I've done some images on Twitter / Instagram recently to encourage people to register for the local elections. I'll be making debranded versions to use in a countdown to the EU elections and put them online for anyone else who wants to use them.

I was asked whether the same issue with the UKIP splitters applied to any remain splitters.

To an extent it does matter, although we will never reach the crazy level of splitters that the far right always seem to achieve based on minor differences that seem of relatively little consequences to outsiders.

In 2014, if you count all the minor parties (not just far right ones) that got no seats in SE England, they took 120,000 votes between them.

The next seat not allocated (to Labour) would have needed 105,500 votes to take it - so if there were 11 seats (rather than 10) and all the small groups had got together they could just have managed to get a seat. That won't happen though - I'm fairly confident of that.

Lib Dems, Greens and Labour don't worry me as much here as Change UK (formerly The Independent Group) and Renew (there may well be a range of other similar ones we have barely heard of), some of which seem more like a publicity stunt for their founders than a serious attempt to get elected.

Renew stood in Newport West last week, which was a by-election (admittedly in a leave area) that would normally mean very focussed campaigning as the whole party's countrywide resources can be poured into that one seat.

They got 3.7% of the votes though, finishing 7th.

There is some recent polling for the EU elections in the UK here.

I sense there is going to a big battle going on between UKIP and Farage's new Brexit party. We need to make the most of any confusion that this may cause splitting voters, while at the same time not letting it fill the news stories at the expense of the many sensible remain minded parties out there (who just get on with stuff rather than creating dramas).

If I take the most recent polling (which is country-wide, so not accurate for the South East England constituency) and allocated the votes for the main parties last time to the same main parties plus the Brexit party, the outcome would be:
Labour 5 (+4), Tories 3 (=), Brexit party 1 (+1) and then either Lib Dems or Green get 1 (based on the current polling it's a dead heat).
So on that basis, UKIP would get none (-4), but Farage would still be around, but with nothing like his current number of seats in the SEE constituency from last time.

To try another scenario:
In the previous one, the Lib Dems & Greens were a dead heat.
If one dropped out and all those votes went to the other, they would still only get 1 seat, so there is no clear advantage to this.
If the UKIP & Brexit party then clubbed together in that same scenario, then they'd get 2 seats (+1 from before, but -2 from 2014) by pinching one from Labour - but I don't see any possibility of the latter outcome occurring.

To take one final variant on the scenario, If UKIP and the Brexit Party clubbed together in the original scenario (Lib Dems and Greens entering separately), they'd still get an extra seat (-2 from 2014), but Lib Dems / Greens would both get 0 (-2 from 2014, -1 from current polling) and Labour would keep their 5 (+4 from 2014).

Labour have some good MEPs at present - people like Seb Dance and Richard Corbett - we should probably wait to see their party lists to see who are on them and their EU credentials. Normally, the currently sitting MEPs are at the top of any new list, so get the first shot at it unless they are standing down.

I'm sure more will transpire in the coming weeks - in the mean time though, feel free to play about with the spreadsheet and get a bit used to how PR systems work in practise.