The Election that might have been

10 May 2010

Britain’s new political map is simple. Just take a look. We have a blue south, a red north, and a few spots of yellow in between.


But it’s not a map most voters would recognise as legitimate after last week's broken election. Whole counties and cities are now the personal property of one party. Once again First-Past-the-Post has generated results that stretch the idea of ‘representative’ democracy to breaking point.


So we’ve had a look – modelling the systems currently on the table in coalition talks. There are whispers of Tory pledges on AV. There are pretty loud shouts from Labour that Clegg might get AV without a public vote, and possibly STV at a future referendum.


So let’s take a look at the numbers. (Full data is available here)

2010 Election

Nationally the picture is much as we’d expect. AV would have made a negligible difference to the parliament produced by last Thursday’s election, short of giving every MP a real mandate. STV on the other hand would be a major step to restore credibility back to representative government, by ensuring that our parliament that actually looks like Britain.


But regionally the picture is starker. Lets’ take a look at two regions, and the varied fortunes of the 2 big parties.


North East

  Con Lab LD
FPTP 2 25 2
STV 8 13 8
AV 1 26 2


The South East

  Con Lab LD other
FPTP 75 4 4 1
STV 50 11 23  
AV 74 4 5 1

South East England’s political map is bright Blue. But that’s not quite how the all those voters who backed other parties see it.  AV does a little something for them, but a single seat changing hands, but STV levels the playing field. The Tories get the bulk of seats, but Labour have a presence, and Lib Dems could more than quadruple their presence.


Just like the song they don't seem to sing anymore, the North East is deepest Red.  AV again does little to address the balance, while  STV actually hands Lib Dems and the Tories more seats between them then the party that currently 'owns' the region.


STV is a historical commitment for the Lib Dems. We’re hoping their memory serves them as they stand on the cusp of power.


PS: At the onset of the campaign, we published the names of the winners in nearly 400 safe seats. We called these 'victories' an affront to democracy. Even with the most fluid election in generations 380 seats came in as called.


Well we’re sorry that we got 3 wrong. But we’re more concerned for the tens of millions of voters that, despite the most unpredictable contest in generations, didn’t see an election. That’s safe seats for you!


PPS: Since we began on this post the PM has resigned. For those who think reformers deal in fantasy politics, think again.


41 Responses to The Election that might have been

grahams 10 May 2010

Three points:
1) The AV system, far from making a "negligible" difference, would (on your questionable assumptions) have changed the outcome fundamentally. Either Labour or Conservatives would have commanded a majority in the HoC in coalition with the Liberal Democrats, who would face a straight choice of preferred partner.
2) I note that the STV system (again on your second preference assumptions)would have stopped the Green Party from entering the HoC. Not exactly a gain in improving representation of voters.
2) You continue to ignore/suppress the two round voting system, where the run-off vote would allow all voters to assess the situation (unlike AV) and positively choose their MP. This would give more power to voters and also eliminate the random (but manipulable) chance element of the STV system, where the outcome depends on the order in which our votes are counted.

admin 10 May 2010

Our modelling is informed by ComRes polling taken a few days before the election. AV can either exagerate majorities or constrain them. In this case, with no landslides on the cards, its effects are muted.

STV wouldn't have helped Caroline Lucas in Brighton last week, based on this yes. But that's because the Green's were forced to focus their efforts on one seat. They are playing the game all politicians play, to get their candidate in place. It's pretty clear that any change of voting system changes the way people vote, and the way politicians campaign. We would expect Greens would apply more energy in Kemptown and Hove to get in.

We've not tackled the Two Round system as it's not on the table. But also because it suffers very much from the limitations of FPTP. The Chirac- Le Pen run off in 2002 is quite illumintating. Le Pen won 16.9 % of the votes in the first round of voting, qualifying him for the second round thanks to a poor showing by the Socialist Jospin and the presence of 15 other candidates. Socialists voted for Chirac with nose pegs in 2002. Preference voting would have meant they wouldn't have needed to.

grahams 10 May 2010

Thank you for your reply. I think that the Chirac/Le Pen run-off illustrates one of the virtues of the two round system: extremists cannot sneak in by the back door. Under STV, unless I have misunderstood, a "Le Pen" standing in a five member constituency would have sailed into Parliament with 16.9 per cent.

Anthony Smith 10 May 2010

You can't take voters' behaviour under one system and project it onto another system - under AV people would have voted differently.

For example, any Lib Dem would-be-voters who tactically voted Conservative/Labour in order to prevent the Labour/Conservative candidate from winning - instead of voting Lib Dem - would under AV vote Lib Dem, according to their true preference. So, without further information, it is not possible to say based on the election results what would have happened under AV or STV.

Please tell me you knew this??

admin 10 May 2010

Anthony. You make a good point. Modelling is part art, part science, and it can only give us a glimpse of what might happen. We've used it (and there's a big note on methodology here to illustrate what could have happened, informed by 2nd preference polling from ComRes.

When the Jenkins Commission sat down to look at systems, and particularly when the Citizens Assembly in British Columbia convened to pick a system, they both looked at modelling to work out the way forward. Our research tells us that STV works. In Scottish local government especially it's broken Labour's monopoly on council seats, brought in scrutiny and delivered voter choice. But it's hard to make the argument for changing British politics by talking council wards in Aberdeen. So we have to look at possible worlds, and compare what we've got with what's possible.

We've looked at the mechanics of STV and AV in practice, and hope to do it some justice at least.

Anthony Smith 10 May 2010

Sorry - I hadn't appreciated the other data that were used, but I suppose it's obvious, giving it a bit more thought. I'm fully in favour of electoral reform and STV, but get a bit frustrated with the "under PR we would have nn Conservative ... and nn BNP MPs" kind of statistics that get thrown around elsewhere - based on voting behaviour under FPTP.

Calvin Smith 10 May 2010

I think that what all here seem to be missing is that an election as part of a democracy is to give the people a fair vote. PR is the only system that offers each member of the democracy an equally valid vote and is therefore the only fair (democratic) system!
Let's not be misled about the problems of outcome, that is a job for politics not election.
Once again I see an attempt to make the voters responsible for political bargaining which is the job of the politicians not the electorate.
One day those voted into office will take responsibility for their role and stop trying to make the voters do their jobs for them!

cna training 11 May 2010

Terrific work! This is the type of information that should be shared around the web. Shame on the search engines for not positioning this post higher!

Robin 11 May 2010

I despise the BNP, believe me, but what I find severely worrying is that people want to "fix" the voting system to prevent parties like the BNP from entering Parliament. If we do indeed live in a democracy, then parties like the BNP need to be able to be voted into Parliament, if the voter wishes to do so.

The solution is not to oppose proportional representation by sticking to AV. The solution is to campaign and convince people that racism is not the answer.

But trying to "fix" the system to keep a party out sets a very dangerous precedent indeed, and it worries me that I frequently hear such undemocratic drivel espoused by senior Conservatives and also some senior figures in the Labour party.

If that is their understanding of how democracy works, the BNP is the least of our worries.

Karlos 11 May 2010

Mixed Member Proportional (you know it as AMS - Additional Member System) is what we're used to in New Zealand now.

It's less complicated that STV & AV (no Donkey Votes). Just have a 4% cut-off to keep the loonies out -lets call that a "margin-of-error" - and institute an "overhang" (if needed) to keep the proportions honest.

It's really not that complicated, no 23 choices of candidates to rank (after the first couple...randomly?), no second round (oh great...not again), plus you can vote for that 'nice' MP from the 'other' party if you want him representing your electorate, and can vote for your party too!

I'd love to see what the results would have been in the UK election using MMP.

Electoral Refor... 11 May 2010

We're working on it Karlos!

G Cox 11 May 2010

Just as I am fed up with certain papers for misinformation against better voting systems, I am annoyed /dismayed that the BBC is carrying your estimates of AV etc for this election. There are no assumptions listed and they are the devil to find on your site. Quite scandalous really. Propaganda from both sides, but the BBC should not promulgate yours without the details. Also on the BBC site your bias for one form is not stated: ie your are not independent like the IFS.
You models are probably as good as those that got us into the banking crisis. For example, under AV, parties may even choose different candidates to try to get the second preferences: eg by eschewing extreme candidates.

Electoral Refor... 11 May 2010

Please take a look at methodology. We used ComRes polling asking people how they'd use 2nd preferences to inform this modelling.

Peter Berrow 11 May 2010

Why no information on the Majority vote system that they use in France, Czech and Hungarian elections. A far better system as you only ever have one vote(which stopped the Front National winning many seats back in 88). Does this mean that you are in fact a Liberal Democrat front or will you show this option?

Electoral Refor... 11 May 2010

Please see our reply on an earlier post on how that system impacted on the Chiract/Le Pen election. We've modelled all the key systems on the table in these negotiations. The French system lacks advocates in the UK, our research hasn't seen too much to recommend to it, but we're happy to hear the arguments.

We've been around for a lot longer than the Lib Dems (100 plus years), so we're happy that STV has picked up advocates in all parties. Douglas Carswell and newly elected Cardiff MP Jonathan Evans are Tories who really deserve a nod on speaking out for it.

Alasdair Scott 11 May 2010

I am curious to know whether or not a particular mixture of First-Past-the-Post and PR has been proposed or tried.

In this system:

a) the constituency system would be retained as is;

b) Westminster votes would become fractional and weighted by a PR factor;

So, hypothetically, for last week's election:

305 Conservative MPs were elected, but their 36.1% proportional share of the vote should have gained 234.65 seats. Thus each Conservative MP's voting weight would be 0.77

Similarly for a Labour MP, the PR factor would be 0.73 and the Liberal Democrat MPs 2.62

The PR factor would be applied to all lobby votes - I haven't thought about select committees etc.

Any thoughts?

Martin H 11 May 2010

Accountability is the main requirement for a democratic voting system to be fair and just. If the voters can not determine the outcome of an election it is by definition unfair. Voters need to be able to kick their politicians out, and this is what we get from FTP.

AV is an appalling alternative - its absurd to suggest that the 3rd, 4th, 5th placed candidates 2nd preference should have any influence on the outcome of an election.

I am with Grahams - if we need a constituency MP to have 50% of the vote cast - a run-off election is the only reasonable option.

Jonathan 11 May 2010

As an ex-Liberal Alliance councillor but now having lived in Ireland for the last 23 years, I have seen both sides of FPTP and STV.

As a ex-Liberal (now Irish Labour Party) member, I am passionate about fair voting and the inclusion of every voter by ensuring that every vote counts equally.

Now I have experienced it in Ireland, I'm not sure that STV resolves the problem of "safe seats". The two largest parties - Fianna Fail and Fine Gael - both deriving from a split in Sinn Fein following the Irish Civil War in the 1920s and both being centre right parties, have, realistically, stitched up the system

The strategy here is for the candidates to split up the multi-member constituencies into areas that only they campaign in. This leaves smaller parties fielding only one, or occasionally two, candidates in constituencies to attempt to attract all their party's votes in the constituency.

The effect of this is that the two largest parties, whilst on paper attaining the largest number of votes and seats, are in fact elected on the basis of negative voting - anyone but Fianna Fail/Fine Gael but Labour/Greens/Sinn Fein/Socialsts, etc. can't win so I'll vote for the least worst option.

It seems to me that the German system may be more appropriate for the UK. AV used for single member constituencies, providing personal, local, accountability, and a national closed party list system to ensure proportional representation in the House of Commons, based on each parties' percentage of the popular vote.

Clearly, this system has the main drawback of the list candidates not being locally accountable to identifiable voters. It would also require a variable number of MPs to guarantee proportional representation. However, it does appear to overcome the issues of local representation, proportionality of representation and negative voting.

Wilfred Day 11 May 2010

You are presenting STV as a single model. But what is the average District Magnitude of your model? Northern Ireland considered STV with all five-seaters, decided it was not proportional enough, and chose all six-seaters. Conversely, if you use smaller DMs in rural areas -- where the Conservatives would expect to elect 2 MPs in most three-seaters -- you build in a bias towards the Conservatives, just as Fianna Fail gets a large-party bonus in Ireland. Who do you propose should decide the District Magnitude?

grahams 11 May 2010

May I respectfully suggest that the run-off or two-round system should not be bracketed with First Past the Post, even though it would probably be more acceptable than other reforms to those happy with the present system. Rather, as others point out, the run-off vote should be seen as a superior form of the Alternative Vote. It is superior because: 1) it is open, easy to understand and transparent, making it almost proof against conspiracy theories; 2) it gives voters more information about the national trend and the local picture after the first round, empowering those of us who back losing candidates to use our second preference to best effect and 3) the winner will have to be positively backed by a majority of voters.
For these reasons, a run-off system is more likely to be acceptable to those who are sceptical about AV, AV+, STV, PR etc. I realise that you have concluded after much study that STV is best but this does not seem to be on the political agenda for the foreseeable future. A referendum on some form of AV is, so it is surely worthwhile to try to settle on the best form of AV and the one most likely to win general support. I do not think it is a foregone conclusion that a referendum on voting reform would be passed, especially if memories of recent events are still vivid.

Nick Rowe 11 May 2010

I'm intrigued by the ideas of AV/AV+/AM but there are issues of the complexity of the forms, etc etc. I've come up with a system which, although not as representative as STV, is far easier than AV+:

Use a First Past the Post system with a small number of additional top-up of MPs for each Region. But instead of getting people to vote twice, or specify preferences, you compare the FPTP results with proportion of vote. Any party with more seats than proportion of vote can't have a top-up seat. Those that have less than the proportion of vote then have the top-up seats distributed amongst them on the basis of their share of the vote - a kind of mini PR. The seats are allocated to those candidates who achieved the highest proportion of votes in their constituencies and they then become "Additional Members" for those constituencies and the Region as a whole. So if a candidate runs a good campaign, comes in second but their party fails to take many (or any) seats in a Region (LibDems in East Midlands, Tories in Scotland, etc.) they are likely to be awarded a top-up seat.

- Some representative allocation of seats
- No complicated forms
- Linked to constituency
- Results almost as quick as FPTP
- Voters make 1 choice, which should correspond to how they want to vote rather than tactical voting

- Some constituencies will have 2 MPs (although there are Pros to this too)
- May give extremist parties an MP (although technically this is merely democracy at work)
- Likelihood of arguing over number of top-ups required (linked to constituency reform)


Christopher Polis 12 May 2010

In Tasmania we have had a STV system (called Hare Clark here, but same thing) for quite some time.

Parties here seems to have settled largely on running the same number of candidates as there are seats available, although there are exceptions to this.

Parties are not allowed to preference individual candidates or provide 'how to vote' cards that show this, which works well (and leads to much less mess around the booths than at federal elections where this is not the case.)

I think one of the big advantages of the system that is really not covered here is the ability to separate out the candidate from the party. Because there are a surplus of candidates for each party, you have the ability to be selective while still voting your party preference. What this means is that when a poor performing candidate does get in, the don't last more than a single term. This is, I believe, the single best feature of the system.

As to numbers, I'd suggest sets of 7. We've had that before (currently at 5) and will probably return to 7 shortly. 7 still sets a reasonable minimum bar, keeps the regions relatively localised, and gives good proportionality.

Best of luck moving forward with things over there.

mbt shoes 12 May 2010

great information you write it very clean. I am very lucky to get this tips from you.

James Gilmour 12 May 2010

Karlos and All
We have UK experience of MMP, here called the Additional Member System (AMS). We've used it to elect the members of the Scottish Parliament, with regional lists rather than one national list. AMS (MMP) is certainly better than FPTP, as just used to elect the UK Parliament at Westminster. But AMS (MMP) falls far short of STV-PR.

Anyone who is flirting with AMS (or its very poor relation, AV+) would do well to read the evidence submitted by Fairshare (Scotland's Campaign for a Better Democracy) to the Arbuthnott Commission and to the Calman Commission:

Fairvote 13 May 2010

Surely the only honest democratic method is 'ONE PERSON ONE VOTE' . The Parties are displayed on the voting paper and the voter chooses the party they prefer. Every individual vote is treated with respect, not dishonestly as under the present method. Each constituency candidate is elected to sit in parliament as a result of their winning percentage. So simple, but with an honest and clear conclusion to the nation's voting results. That's proper democracy.

Beth Bromley 13 May 2010

I am looking for data that indicates the fraction of the voters that voted 'tactically'. In other words what fraction of the voters are sufficiently disenfranchised that they voted for a person/party that was not the person/party they actually wanted in power? This seems to me to be a clean statistic on how badly this system is failing people. Is that data available?

Electoral Refor... 13 May 2010

We're reliant on polling for this. For our modelling of other systems we used ComRes polling ( addressing a range of these sort of questions. A hard figure is very hard to pull off, as polling itself isn't an exact science.

There are lots of hard indicators you might use in place. The last Labour government had a 1/3 of the votes, but was only supported by 20% of the whole electorate. The new coalition has nearly 60% of voters behind it, - 38% of the total electorate. A step in the right direction if we're talking mandates.

Martin H 13 May 2010

Perhaps compulsory voting is the only change we need to make. Seeing the Electoral Reform Society (and others) trotting out figures for % of total electorate etc as an argument for electoral reform is simply mixing orange and lemons.

I suspect that if everyone voted, issues of legitimacy would disappear.

Compulsory voting and the checks that would need to be in place to achieve this would also address the electoral fraud issue which has now become so prevalent.

A fair voting system gives the electorate the opportunity to determine and understand the process and likely outcome. None of the PR systems I have seen described are anything except opaque. And have the systems such as AV is all its forms simply gives that casting vote to people who voted off-piste (often for extreme single issue parties such as BNP and UKIP). Lets not forget the Labour Party, the Conservative Party and the Lib-dems are coalitions and between them have room for almost everybody.

Alan Walton 13 May 2010

AV+ is said not to be not fully proportional. I think it can be, certainly as far as the main parties are concerned. You just need to use the "+" for a single national constituency. That could leave existing constituency boundaries unchanged, though it would be better to reduce their number. Of course with any PR system the precise boundaries don't change the overall result. With a 4 or 5% threshold needed to qualify for PR enhancement, and a total number of seats to be won, it would be simple arithmetic to subtract the seats won by non-qualifying parties, and ensure that the remainder were shared pro-rata their respective "+" votes by the qualifying parties. I don't know how many added seats there would need to be to ensure this was always possible, but in this latest election it wouldn't be many. I would add another ingredient. If the "+" vote is also conducted by AV, still allocating seats pro-rata the first preferences, the full AV count could be used to determine the coalition leader (if needed). This would be the one most acceptable to most voters, and therefore most likely to be able to build a coalition of principle rather than convenience. This could significantly reduce the post-election bargaining.
I appreciate it would create two kinds of member, but that is a good thing. There would still be people with local knowledge in the Commons when needed, but others would be free to specialise in areas of policy. Surely front bench spokesmen cannot give constituencies the time they deserve. I would further make it a qualification for inclusion on party lists to have served previously for a constituency. This is to address concerns that unelectable people could get in by that route. To me a significant secondary benefit of the AV+ process is the ability to elect an individual of particular merit without affecting the overall power balance, since that is determined by the "+" element.

Alan Walton 13 May 2010

Alasdair Scott proposed an idea to keep everything the same, but give each member a "voting power" scaled to make the total for each party proportional to the popular vote. This is ingenious, perfectly fair, and (to me at least) original. The detail of how to handle independents and other very small groupings is not obvious but could be worked out. I'm pretty sure though that it won't fly. A rather more complex election every 5 years is one thing; needing electronic voting for every division in the house, and the very idea that a Liberal is worth 2 or 3 Conservatives - would be rather harder pills to swallow.

Alan Walton 13 May 2010

The ERS reply referring to coalition mandates is debatable. Where a coalition of similar parties puts together a programme that fits both manifestos, it can reasonably claim that everyone who voted for either supports their programme. Where the agenda involves implementing things that one or other of them opposed in the election, their votes cannot be deemed a mandate for that.

Paul Renshaw 13 May 2010

Ignoring straight-forward closed-list PR is a miserable way to behave. It is the only system which genuinely represents every vote cast (I love that when the Conservatives and Labour say that FPTP represents 'all significant points of view' you laugh them down, but when STV mainly benefits parties about the size of (coincidentally) the Lib Dems, suddenly it's a reasonable thing to say).

We have two Houses of Parliament: if contact with a local representative is desirable, which I agree that it is, why not have the House of Lords elected by FPTP/run-off/AV, and continue to perform their current function of scrutinising legislation passed to them by an entirely proportional Commons.

Electoral Refor... 13 May 2010

Paul. Party lists suffer from several problems, evidenced in our Euro elections. For one, the parties dictate which candidates make the lists and in which order, preventing voters exercising real choice. We support STV because it allows voters to choose between parties and individuals, and delivers proportionality while retaining a constitency link - which is pretty much non-negotiable for the House of Commons.

Beth Bromley 13 May 2010

Electoral Reform Society

I understand the point that you don't have your own data. However I think your interpretation of existing data is underplaying the role of tactical voting. You suggest it can be estimated from the difference between which party people said they supported and the results of the actual vote. However this assumes that tactical voting only goes one way ie from lib to lab, or lib to con, but in reality there are two horse race seats in all combinations so in some places there will be con to lib tactical voting that will cancel out lib to con tactical voting elsewhere, thus reducing the visible level.

I think you could really only get a decent number for this by polling people and asking directly if they voted for someone who was not their first choice but as a better first approximation you could take data on party support versus vote on a seat by seat basis. I will see if I can find that data.

Paul Renshaw 13 May 2010

I agree that contact is desirable, hence my suggestion of a less powerful house which does elect representatives on an individual basis.

With respect to the disadvantages of any house elected by closed list, it is not necessary to fill all seats in a closed-list system; given that the individual MPs are unelected, the idea of them rebelling, or worse switching parties, is unacceptable. Frontbenchers could be selected from a closed list, and would have total control of, AND ACCOUNTABILITY FOR, casting their proportion of the popular vote.

I agree that in theory individual accountability to a section of the public (let's not forget that it's a very small, partisan section of the public) for behaviour in parliament and in public life would be preferable. However, in practice our current system does not provide this; strong whips, safe seats and weak local parties have made a joke of the idea of selecting individuals rather than parties: I see no reason why STV would be much different in these respects (although I concede that the ability to split your vote provides an alternative in cases of the most outrageous abuses).

For me, far more dangerous than accepting that anyone who seeks power is highly likely to be unpleasant and prone to dishonesty is the idea of refusing to give representation to views of small minorities, often the most progressive in society who have aspirations to make genuine, revolutionary changes to society (such as Green Parties), or those who are suffering the worst under the status quo (notably the BNP).

It's obvious that many people who vote for the BNP feel disenfranchised in this society and completely alienated from the legislature and the government. As long as their votes go unrepresented and can be entirely ignored, they will be, as will their problems, and so resentment will continue to fester.

STV is just another way of turning one vote into no representation, it's another brick in the wall, just a slightly yellower brick than FPTP.

Paul Renshaw 13 May 2010

"pretty much non-negotiable for the House of Commons" - disappointing, could have come straight from the Tory backbench regarding any electoral reform whatsoever.

Martin H 14 May 2010

None of the voting systems on offer (STV, AV, AV +) seem fair - all require re-distributive recounting - this means they are not transparent, easily understood and a voter would have no idea of the consequence of their vote.

However, I believe what I call Regional FPTP + addresses all the issue of fairness, keeps the MP/Constituency link, is simple to understand and administer, and a vote for a persons 1st preference will count a regional level - so eliminating wasted vote syndrome. A further advantage of Regional FPTP + is that the Regional MP's would retain the link with the community they serve.

Regional FPTP + in practice:

Constituency MP's elected as now (75% of parliament)
Regional MP's elected in proportion to votes cast (25% of Parliament)

In addition I would introduce the following at the same time:
Parliament reduced to 600 MP's (450 Constituency and 150 Regional)
Fixed term parliaments
Compulsory voting
Limit access to postal voting (cut down fraud)
Limit regional MP's to two terms (No limit on Constituency MP's service)
Threshold for Regional MPs to be 5-10% of vote


The larger constituencies under Regional FPTP + would also reduce some of the boundary anomalies that come into play under the current system, and result in the outcome being more closely related to the vote.

Further at the moment it takes fewer votes to elect an Labour MP (average 33,000 votes) than a Tory MP (average 35,000 votes) and as for a Liberal MP it takes 120,000 votes for each MP they have. Regional FTP + would smooth out these differences.

I don't have a full regional breakdown of votes, so can't work out all the implications.

However, if one looks at Scotland under Regional FPTP + there would probably be a total of about 50 MP's (Down from 59) - 35 Constituency and 15 Regional. An obvious winner in this would be the Scottish Tory voter (17% of electorate), as they would get 3 or 4 Regional MP's. At present they only have 1 MP for the whole of Scotland and thus a Conservative vote in Scotland is currently perceived to be 'wasted'.

One thing is certain the Lib-Dems would have benefited substantially under a Regional FPTP + system. Their share of the vote (26%) would have given them around 37 Regional MP's.

Christopher Polis 15 May 2010

A few more notes on STV in practice (Tasmania, 5 regions with 5 elected candidates):

In Tasmania, for a vote to be legal, at least 5 clear preferences must be shown (i.e 1, 2, 3, 4 and 5 must clearly and unambiguously be marked on the voting slip).

In practice, this means parties stand at least 5 candidates as otherwise there would be significant risk of voters who only fill in one party column not recording a valid vote.

Here, the most any party wins is 3 candidates in 5. So even for the dominant party of the day, the electorate has ample opportunity to throw out poor candidates - which happens regularly. This is, to me, the best feature of the system. It simply results in better politicians being elected.

I would say from my experience counting votes, that approximately one third of voters fill out only their preferred party.

People who care about how things go beyond their 'own party' vote more, and that makes a difference the further a count goes. If you don't vote further, it is effectively saying that 'while I'd like to see party A in government, i'd rather let someone else choose between the other parties'.

For reference on how flows work in practice see for example:

polo ralph lauren 28 May 2010

You continue to ignore/suppress the two round voting system, where the run-off vote would allow all voters to assess the situation (unlike AV) and positively choose their MP. This would give more power to voters and also eliminate the random (but manipulable) chance element of the STV system, where the outcome depends on the order in which our votes are counted.

avery 26 Apr 2011

here is a simple diagram that I think demonstrates why "First the past the post" isn't exactly perfect.

see what you think...

Electoral Reform Society's picture
Electoral Refor... 30 Mar 2012

This is a test.