Winner of the New Statesman SPERI Prize in Political Economy 2016

Tuesday 10 December 2019

Why you should vote tactically, and how to do it.

In this election we have a choice. We can choose a party led by an inveterate liar, which is happy to appeal to the racist or xenophobic vote, happy to take us out of the EU with no deal with all the consequences for public services that entails, happy to see ever longer waiting times for GPs and A&E, happy to see more homeless people on the streets and more food banks.*** Instead we can choose a hung parliament with a Labour minority government that may not even be led by Corbyn, with their power curtailed by the smaller parties, including the Liberal Democrats, that can actually do something about the problems we face. But we can only make that choice by voting tactically.

But I have an ethical problem with voting tactically

There is one serious argument against tactical voting. It is the idea that if you have a choice between voting for a good party, a not so good party and a terrible party you should always vote for the good party. (The ranking is what matters here, not my description.) The argument is that if you vote for the not so good party, you are in some senses endorsing the bad things that party or its leaders have done in the past.

That argument applies to those who prefer to vote Liberal Democrat because of Labour’s antisemitism problem, or those who think they cannot vote Liberal Democrat because they were part of the austerity Coalition. Both arguments are wrong, because they simply do not apply in a FPTP system when the only two possible winners in the constituency you live in are the not so good party and the terrible party.

In a FPTP system your vote for the good party will simply have token value. However you could have done something towards preventing the terrible party coming to power, but you chose not to. In that way you become responsible, in a small way, for what happens when the terrible party comes to power. By your inaction, you will have contributed to the terrible party coming to power.

If you vote for the not so good party, will you be responsible for the not so good things happening if that party wins? No, because all you could have done differently is let the terrible party win. There is no way your voting for the good party will influence anything. It is a wasted vote because you will have wasted an opportunity to make the world a better place. The Kantian ‘do no evil’ idea does not apply because doing no evil actually means allowing more bad things to happen.

You may say this is a consequentialist argument, and your philosophy is different. You might think about the fact that your philosophy would allow tyranny to come to power, just because you had some problems with the opposition to tyranny. If you think that is far fetched, you need to note that part of their manifesto gives the Conservatives a mandate to change our constitution so that the executive has complete control over parliament.

There is a weaker argument for voting for the good party, and that is thinking about your vote as part of a repeated game. The argument suggests that by voting for the not so good party, you are encouraging it to remain not so good. But this fails for the obvious reason that if you do not do everything you can to stop the terrible party coming to power, you are encouraging the terrible party.

And if all that doesn't convince you, use where you can actually help the party you prefer win something. 

But who should I vote for. It’s so confusing.

For most people its not. Have a look at this website, for example: or will see in most cases there is total agreement about who to vote for tactically. Of course some constituencies are such safe Tory or Labour seats that your vote is highly unlikely to achieve anything. But my rough list of constituencies were tactical voting matters (England and Wales only I’m afraid) is as follows (with the party to vote for in brackets). It is a long list because I have been deliberately pessimistic about what the two parties could lose and optimistic about what they could gain, because there are always surprises in any General Election.

Why did I bother doing this you might ask. It is just a by-product of some work I did out of my own curiosity looking at marginals, and I thought I might as well share it. I have waited until now to make these suggestions because I think a lot of confusion has been caused by sites that have made calls earlier on in the campaign and then had to revise them. Many a LibDem bar chart has been based on that misleading information. Fine if the polls don’t move, but they were always going to move to the two main parties in this election.

Aberconwy (Lab)
Alyn and Deeside (Lab)
Ashfield (Labour)
Barrow and Furness (Labour)
Bassetlaw (Lab)
Bath (LibDem)
Battersea (Labour)
Beaconsfield (Independent - Dominic Grieve)
Bedford (Lab)
Bermondsey and Old Southwark (Lab)
Birmingham Edgbaston (Lab)
Birmingham Northfields (Lab)
Bishops Auckland (Lab)
Blackpool North (Lab)
Blackpool South (Lab)
Blyth Valley (Lab)
Bolsover (Lab)
Bolton North East (Lab)
Bolton West (Lab)
Bradford South (Lab)
Brecon and Radnorshire (LibDem)
Brentford and Isleworth (Lab)
Bristol North West (Lab)
Burnley (Lab)
Bury North (Lab)
Bury South (Lab)
Calder Valley (Lab)
Camborne and Redruth (Lab)
Canterbury (Lab)
Cardiff North (Lab)
Carlisle (Lab)
Carmarthen West and South (Lab)
Carshalton and Wallington (LibDem)
Cheadle (LibDem)
Chelsea and Fulham (LibDem)
Cheltenham (LibDem)
Chingford and Wood Green (Lab)
Chipping Barnet (Lab)
City of Chester (Lab)
Clwyd West (Lab)
Colne Valley (Lab)
Copeland (Lab)
Corby (Lab)
Crawley (Lab)
Crewe and Nantwich (Lab)
Croydon Central (Lab)
Dagenham and Rainham (Lab)
Darlington (Lab)
Delyn (Lab)
Derby North (Lab)
Dewsbury (Lab)
Don Valley (Lab)
Dudley North (Lab)
East Devon (Independent)
East Worthing and Shoreham (Lab)
Eastbourne (LibDem)
Eastleigh (LibDem)
Enfield Southgate (Lab)
Esher and Walton (LibDem)
Filton and Bradley Stoke (Lab)
Gedling (Lab)
Gower (Lab)
Great Grimsby (Lab)
Guildford (LibDem)
Halifax (Lab)
Harrow East (Lab)
Hastings and Rye (Lab)
Hazel Grove (LibDem)
Hendon (Lab)
High Peak (Lab)
Hyndburn (Lab)
Ipswich (Lab)
Keighley (Lab)
Kingston and Surbiton (LibDem)
Lewes (LibDem)
Lincoln (Lab)
Loughborough (Lab)
Mansfield (Lab)
Middlesbrough South (Lab)
Milton Keynes North (Lab)
Milton Keynes South (Lab)
Montgomeryshire (LibDem)
Morecambe and Lunesdale (Lab)
Moreley and Outwood (Lab)
Newcastle under Lyne (Lab)
North Cornwall (LibDem)
North Devon (LibDem)
North East Derbyshire (Lab)
North Norfolk (LibDem)
Northampton North (Lab)
Northampton South (Lab)
Norwich North (Lab)
Oxford and West Abingdon (LibDem)
Pendle (Lab)
Penistone and Stocksbridge (Lab)
Peterborough (Lab)
Plymouth Moor View (Lab)
Portsmouth South (Lab)
Preseli Pembrokeshire (Lab)
Pudsey (Lab)
Putney (Lab)
Reading East (Lab)
Reading West (Lab)
Richmond Park (LibDem)
Rossendale and Darwen (Lab)
Rother Valley (Lab)
Rushcliffe (Lab)
Scarborough and Whitby (Lab)
Scunthorpe (Lab)
Shipley (Lab)
Shrewsbury and Atcham (Lab)
South Swindon (Lab)
Southampton Itchen (Lab)
Southport (Lab)
St Albans (LibDem)
St Ives (LibDem)
Stevenage (Lab)
Stockton South (Lab)
Stoke-on-Trent North (Lab)
Stoke-on-Trent South (Lab)
Stroud (Lab)
Sutton and Cheam (LibDem)
Telford (Lab)
Thurrock (Lab)
Totnes (LibDem)
Truro and Falmouth (Lab)
Uxbridge and South Ruislip (Lab)
Vale of Clwyd (Lab)
Vale of Glamorgan (Lab)
Wakefield (Lab)
Walsall North (Lab)
Warrington South (Lab)
Warwick and Leamington (Lab)
Watford (Lab)
Weaver Vale (Lab)
Wells (LibDem)
Welwyn Hatfield (Lab)
Westmorland and Lonsdale (LibDem)
Winchester (LibDem)
Wirral West (Lab)
Wolverhampton North East (Lab)
Wolverhampton South West (Lab)
Worcester (Lab)
Workington (Lab)
Wrexham (Lab)
Wycombe (Lab)
York Outer (Lab)

If I have made any mistakes or bad judgements in this list, do let me know.

There are a few marginal seats which are so safe that you get a choice. In Arfon, or Carmarthen East and Dinefwr, you can vote Plaid or Labour without any fear of letting the Tories win. In Ceredigion the same applies between Plaid and the LibDems. In Bermondsey and Old Southwark, or Cambridge, or Leeds North West, or Sheffield Hallam you could vote for Labour or the LibDems.

Then we have eleven difficult seats, all in England, where both Labour and the LibDems have a claim to be the party to vote tactically. It is important to remember here that over the last months the national polls have been moving from LibDems to Labour, so polls or recommendations made in early or even mid November should be treated with caution. However the latest poll of polls suggest Labour are still below their 2017 total, and the LibDems are significantly above, with little movement over the last few days.

One issue I found it hard to assess was the finding in some local polls that Labour voters are more open to tactical voting than Liberal Democrats. I would hope this does not apply to any Liberal Democrat voters who might take any notice of what I say, because if it does they have not been reading my other posts. An argument that should have equal weight is that, if Johnson is not to romp home, the polls need to be a little biased against Labour because they underestimate the youth vote.   

Berwick-upon-Tweed A poll at the end of November had LibDems slightly ahead of Labour (which makes sense given 2017 result and subsequent national movement) but the Tories winning easily. More recent B4B MRP poll puts Labour ahead. TV sites split. So Recommendation Labour

Broxtowe (Lab) Anna Soubry was the Tory candidate, now running as an Independent with the Liberals not standing. But this is a Tory/Lab marginal, so Recommendation Labour

Cities of London and Westminster TV sites are split. A poll on 23 November had Chuka Umunna ahead of Labour. However although B4B are suggesting voting for the LibDems, their polling shows Labour ahead. Jon Worth suggests LibDems. So close to call. Recommendation Labour but watch out for any new information. Postscript - new poll has LibDems slightly ahead, so my Recommendation is now LibDem.

Colchester All the TV sites are recommending Labour, So Recommendation Labour.

Finchley and Golders Green I think the large Jewish vote here makes this impossible for Labour, and the TV sites and local polls agree. So Recommendation LibDem

Kensington Narrowly won by Labour last time. A poll in mid November had LibDems a bit ahead of Labour, but the national polls have moved since then. TV sites split, and B4B are on the fence. Too close to call, so I would go for sitting MP. Recommendation Labour but watch out for any new information. Postscript - new local poll has Labour in the lead. 

St Austell and Newquay Both Labour and LibDems have a strong presence here. B4B’s MRP poll puts Labour ahead here in a Leave voting constituency. Recommendation Labour

These two Cambridgeshire seats are similar.

South Cambridgeshire
South East Cambridgeshire

They are Tory seats where Labour were second, but LibDems had a significant vote as well. B4B’s MRP puts LibDems in front, and the only positive recommendations from TV sites are LibDem. Recommendation LibDem


Polls for Wimbledon show LibDems ahead, but there has been a movement away from LibDems towards Labour nationally since then. TV sites split. B4B MRP poll has LibDems ahead. Recommendation LibDem


Polls for Wokingham show LibDems ahead, but there has been a movement away from LibDems towards Labour nationally since then. TV sites that make a positive call all say LibDem. B4B MRP poll has LibDems ahead. Recommendation LibDems.

Again happy to be corrected on any factual errors. If you want to swap your vote use

Other useful sites

***Postscript: Not a complete list. I should have added "that is happy to see climate change increase unchecked".


  1. I'd add as a useful site - easy to search, allows one to compare different tactical sites' recommendations.

  2. Re Totnes recommendation for the LDs: in 2017 the results were: Tory 26k, Lab 13K, LD 6K, Green 3K. Greens not standing this time. For the LDs to win this time, at least 40% of those who voted Tory last time would have to switch, like the LD candidate, from Tories to the LDs. Is that credible?

  3. I think I can fill in for Scotland.

    The SNP and Labour have been competively copying each others policies. So much so that, apart from Scottish independence, it's the same policy programme. The SNP have filched the Green New Deal from Labour, Labour have "borrowed" the SNP policy of free prescription charges. (There may be minor differences).

    So essentially a vote for either is a vote for the same policy programme. Even on the issue of independence the outcome is the same because Labour will agree to indyref2 if it gets a chance to form a government with the SNP.

    So in most seats simply vote for whichever one is polling higher or if you can't easily find that out go by the 2017 result.

    The only exceptions are a few LibDem/SNP marginals where the Tories almost no chance of winning.

    For me, here in Dumfries we have an exciting game of can we depose Boris's Secretary of State for Scotland by voting SNP. It's going to be tight I think. (Although it wouldn't be if everyone followed this blog post.)

  4. I think the repeated game argument is stronger than you suggest. If you vote for the good party, you encourage the not-so-good party to become better in order to gain your vote next time - including changing the voting system. Over a repeated game, future outcomes are thus improved, perhaps significantly. And it becomes more important for the terrible party to improve under such as system. After all, for some people, the terrible party is the not-so-good party and vice-versa.

    The alternative is that the not-so-good party gets worse, turning into the almost-terrible party. You would argue that we should still vote for it, so there is no disincentive for them not to become worse.

    Regardless, in non-marginal seats at least there seem to be no downsides to voting for my preferred good party.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.