Feeds:
Posts
Comments

Archive for April, 2011

A busy week for political polls. We now have another new poll released on Sunday, 18 April by TV3-Reid Research. The updated polling averages now have National on 54.4% +/- 1.4% and Labour on 30.7% +/- 1.3%, both of which appear to be statistically significant changes when compared with the previous update five days ago. Changes for other parties are not statistically significant, which means in the last couple of weeks National has most likely won about 3% +/- 2% points of support at the expense of the Labour Party.

As usual, the two graphs below summarise the polling averages for the party vote after the new poll. The horizontal axes represent the date, starting 60 days before the 2008 NZ General Election, and finishing on the present day. The solid lines with grey error bands show the moving averages of the party vote for each party, and circles show individual polls with the vertical lines representing the total errors.

Party vote support for the eight major and minor NZ political parties

Party vote support for the eight major and minor NZ political parties as determined by moving averages of political polls. Colours correspond to National (blue), Labour (red), Green Party (green), New Zealand First (black), Maori Party (pink), ACT (yellow), United Future (purple), and Progressive (light blue) respectively.

Party vote support for the six minor NZ political parties

Party vote support for the six minor NZ political parties as determined by moving averages of political polls. Colours correspond to Green Party (green), New Zealand First (black), Maori Party (pink), ACT (yellow), United Future (purple), and Progressive (light blue) respectively.

As always, please check the Graphs page for further simulation results.

I should probably point out here that when calculating the polling averages I correct for “pollster bias” – the tendency for each pollster to systematically overestimate support for some parties and underestimate support for others – by comparing each pollster’s results to the trendline and subtracting out any differences. The problem here is that TV3 have moved from using TNS as their pollster prior to the 2008 Election to using Reid Research since, but the calculation treats TNS and Reid Research as the same pollster. More specifically, the calculation assumes that Reid Research and TNS have the same systematic polling biases. TNS used to consistently overestimate support for Labour relative to the trendline, which means that polling numbers for Labour are now generally forced down a little, even though Reid Research don’t tend to overestimate Labour’s support relative to the trendline in the same manner. For this reason, the polling averages given above are probably a little worse for Labour than they realistically should be, and the opposite applies for National. In the future I may have to fix this problem by treating TNS and Reid Research as separate pollsters; it’s something I’m looking into at the moment and I’ll make a decision before the countdown to the 2011 Election.

In other news, it appears National has officially decided to not campaign in Epsom and Ohariu for the electorate seats held by the leaders of their coalition partners ACT and United Future, respectively. This could be a game changer come election time. I have previously blogged on the strategic implications of the Ohariu seat for Labour, which haven’t really changed, although their best short term tactic may now be to contest the seat aggressively instead of throwing it to National, as suggested in the previous analysis.

Read Full Post »

Since the last update we have seen the release of two polls: the fortnightly Roy Morgan poll and the April One News-Colmar Brunton poll. The updated polling averages now have National on 51.3% +/- 1.5% and Labour on 33.7% +/- 1.4%.

As usual, the two graphs below summarise the polling averages for the party vote after the new poll. The horizontal axes represent the date, starting 60 days before the 2008 NZ General Election, and finishing on the present day. The solid lines with grey error bands show the moving averages of the party vote for each party, and circles show individual polls with the vertical lines representing the total errors.

Party vote support for the eight major and minor NZ political parties

Party vote support for the eight major and minor NZ political parties as determined by moving averages of political polls. Colours correspond to National (blue), Labour (red), Green Party (green), New Zealand First (black), Maori Party (pink), ACT (yellow), United Future (purple), and Progressive (light blue) respectively. Party vote support for the six minor NZ political parties

Party vote support for the six minor NZ political parties

Party vote support for the six minor NZ political parties as determined by moving averages of political polls. Colours correspond to Green Party (green), New Zealand First (black), Maori Party (pink), ACT (yellow), United Future (purple), and Progressive (light blue) respectively.

As always, please check the Graphs page for further simulation results.

Read Full Post »

Missed the boat. Appears that while I was drafting the last post on party lists the Labour Party announced their party list for the 2011 Election.

It’s too early to figure out who the winners and losers are (except Ashraf Choudhary, who is a loser); I can’t run any simulations yet as I need a corresponding list showing in which electorate each party list candidate will stand. Having said that, if I was Phil Twyford (#33), Stephanie (Steve) Chadwick (#34) or Iain Lees-Galloway (#37) I would be a bit disappointed.

Read Full Post »

In a post a few days ago on effective party lists I took a few swipes at those drawing up the party lists, but didn’t really offer any constructive criticism. I thought I should probably try to remedy that, so I’ve drawn up a few tips on party list ordering for those drawing them up. (I should warn that what follows is 100% opinion and 0% science.)

There’s two reasons that those drawing up the party list should give thought to the resulting effective lists. Firstly, some voters will be looking at them, one way or another, when deciding for which party they should cast their vote. Not many though; realistically most will just glance at the top of the list, if they bother to check at all. Secondly, there is chance of causing a “Tizard Problem” (NB scare quotes to distance myself from the term), or a “Ward Problem” if you prefer to look at the precedent set by the Green Party in 2008.

In both cases, however, the work-around is the same: you want to structure the party list such that you get appropriate candidates sitting at the top of the effective party list immediately prior to the election. I think that ideally you would want the effective party list to look a bit like this:

  1. Popular candidates (excluding party leadership and senior MPs) who lost toss-up (battleground) electorates.
  2. Popular list-only candidates standing for the first time.

The reason I think this is the ideal structure for an effective party list is that it doesn’t give any of the other parties any ammunition to criticize those next in line if they are to come into Parliament post-election due to a retirement or by-election, and therefore gives a party more flexibility with their strategy during the upcoming term of Parliament. I think there is also a bit of dissatisfaction with MMP allowing rejected candidates back in to Parliament “through the back door,” especially with post-election retirements and so forth, something that would be less of an issue with list-only candidates or candidates that were unlucky enough to lose a close electorate battle. I think there is also more sympathy in these situations for young, first-time, or otherwise up-and-coming politicians than there would be for a long-term parliamentarian who is seen as having already had a fair crack at the job.

Given the above conditions, and assuming you know roughly how well a party will be polling, you can figure out where the sweet spot is on the party list and then reverse-engineer an optimal list. I think the party lists should ideally look something like this:

  1. Popular senior MPs and party leadership.
  2. Other popular candidates in safe electorates.
  3. Popular first-time list-only candidates being choppered in to Parliament.
  4. Unpopular list-only candidates and candidates predicted to lose their electorate seats, who, for one reason or another, must be returned to parliament.

    === Party list sweet spot ===

  5. Popular candidates (excluding party leadership and senior MPs) who are standing in toss-up electorates.
  6. Popular list-only candidates standing for the first time.
  7. Unpopular candidates in toss-up electorates.
  8. Unpopular candidates in safe electorates.
  9. Other token list-only candidates.

There are several advantages to this template for a party list in addition to it resulting in an attractive effective party list as mentioned above. In particular, it gets the more popular candidates and senior leadership to the top of the list where they are most visible, and manages to bury the unpopular candidates who must be returned to parliament by the list, whilst still giving them a safe enough position. Unpopular candidates in safe electorate seats are also buried out of sight at the bottom of the list, without affecting their chances of being returned to Parliament. The low placings for those in toss-up seats may also serve as extra encouragement for them to get out and campaign, for the party vote as well as the electorate vote, obviously.

Of course, it’s not quite this simple in practice, because the sweet spot will move around a bit depending on the order of the list, so the whole process has to be done iteratively. Nevertheless, I think the above serves as a good guide, for the two major parties at least.

I have no idea how the parties actually go about ordering their lists in practice, but would hope they have somebody going through them and running through a few potential scenarios. If anybody wants me to run through a hypothetical party list for them before it is finalised then I would be more than happy to. I suspect, though, that we are instead in for a lesson in poor planning.

Read Full Post »

When I first published a post on the probabilities of being elected on a candidate-by-candidate basis last year I mentioned that the main motivation was that I hoped that people will be able to see which individual list candidate their party vote is likely to be counted for.

Closer to the 2011 General Election I will begin publishing “effective party lists” on a regular basis. These effective lists will show only those candidates on the cusp of winning a list seat for each party, and will hopefully give NZ voters a better idea of where their vote is going, and a more meaningful alternative to a quick glance at the top of the lists for each party which some seem to use now when deciding how to cast their votes.

I think this is still quite important information, and I want to make sure it is available publicly immediately before the 2011 General Election so that people can make a more-informed party vote. But recently another benefit has become obvious: helping parties to avoid the so-called “Tizard Problem.” Since Darren Hughes announced his resignation last week there has been speculation that Labour will ask the next highest placed list candidate, Judith Tizard, to stand aside in order to allow a lower ranked list candidate to filter through to Parliament. There were also suggestions on some blogs that the party didn’t want to stand list MPs in the by-elections for safe Labour seats in Mana and Mt Albert because that would have allowed Tizard back into Parliament, thus limiting their options for candidate selection for the by-elections.

I’m not sure how accurate these theories are of what the Labour Party strategists are actually thinking, and I’m not going to speculate, but if they are true, then they represent a massive failure on the part of those who determined the Labour Party List for the 2008 Election.

After a bit of thought I’ve decided that preventing a hypothetical “Tizard Problem” basically just involves figuring out the effective party lists for each of the parties. Specifically, figuring out who your party vote is going to help get elected is simply a case of figuring out who the lowest ranked list candidate to win a list seat for the party that you voted for is. And this is basically the same problem as figuring out who the highest ranked list candidate to not be awarded a seat is. This can be calculated from the results of the election simulation we use here to figure out which candidates get in to Parliament, and which miss out. The methodology is the same as that for the other simulations on this site.

By simulating a large number of elections based on current opinion polling, and counting up the number of times each party list candidate is the lowest-ranked list candidate to be elected for their party, we can see who the party votes for each party are going towards electing. The results are shown below for each of the eight major and minor NZ political parties, with the probability of a candidate being the lowest-ranked list candidate to be elected for their party shown under the heading “Prob. Last” in the right-most column. The effective party lists are ordered by decreasing values of “Prob. Last.” The party list rankings of the candidates shown at the tops of these lists correspond to the “sweet spots” on each respective parties list, in other words, the list rankings that could potentially result in a “Tizard Problem” for those parties during the next term of Parliament if they were to be filled by electorally unpopular candidates.

The first example is for the Green Party, one of the easier results to understand.

Effective party list for the Green Party

Effective party list for the Green Party.

First place on the Green Party effective party list is Steffan Browning, with list ranking #9 and standing in the Kaikoura electorate. He has a 0% chance of winning his electorate seat, a 79% chance of getting in on the list, and a 32% chance of being the lowest ranked Green party candidate to be elected. In other words, if you vote for the Green party with your party vote there is a 32% chance that your vote will go towards electing Browning. The effective party list then skips up and down the actual party list, with Mathers (list ranking #10), Hughes (#8) and Ward (#11) following. The reason Hughes (#8) is below Browning (#9) is because whilst he is slightly more likely to be elected, he is less likely to be the last person elected. For the same reason, highly-ranked list candidates such as co-leaders Metiria Turei and Russel Norman are almost guaranteed to be elected whether you vote for the Greens or not, and thus have low effective party list rankings of #10 and #11, respectively.

If you add up the numbers for “Prob. Last” in the far right column you will get 100% (give or take a bit of rounding error), which means there is a 100% chance that your party vote for the Green Party will go towards getting somebody elected from their list.

Next up is the Maori party.

Effective party list for the Maori Party

Effective party list for the Maori Party.

Nothing interesting is happening here, with the effective party list exactly matching the order of the actual party list, and candidates either winning an electorate seat or not being returned to Parliament. This is because of the overhang caused by the Maori Party, with the party winning more electorate seats than the total number of seats they would be entitled to based on their party vote.

If you add up the numbers for “Prob. Last” in the far right column you will get 0%, which means there is a 0% chance that your party vote for the Maori party will go towards getting somebody elected from their list. In other words, a party vote for the Maori Party is a guaranteed “wasted” vote.

The examples for both the Green and Maori parties are easy enough to understand intuitively, but things get a bit more complicated when you look at parties that may win both electorate and list seats, such as the ACT party.

Effective party list for the ACT Party

Effective party list for the ACT Party.

First up is John Boscawen, with a 31% chance of being the lowest ranked list candidate elected, followed by Roger Douglas (5%) and Heather Roy (0%, rounded down). In fourth place on the effective list is party leader Rodney Hide, with 0% chance of being the lowest ranked candidate. The reason for this is that ACT Party Leader Hide, like Maori Party leader Pita Sharples, will either win his electorate seat or not return to Parliament. Unlike #2 – #4 on the ACT party list he has no chance of winning a list spot on current polling (where ACT are well below the 5% threshold).

If you add up the numbers for “Prob. Last” in the far right column you will get 36%, which means there is a 36% chance that your party vote for the ACT party will go towards getting somebody elected from their list. In other words, there is a 74% chance that a party vote for the ACT party will be a wasted vote. More specifically, there is a 4% chance that a party vote for the ACT party will be a wasted vote because Hide will lose Epsom and nobody from ACT will be elected, and there is a 70% chance that a party vote for the ACT party will be a wasted vote because although Hide wins Epsom, the party does not get enough party votes for somebody else to join him as a list MP.

The three patterns above explain most of the possible outcomes as far as effective party lists go, and for the other parties the effective party lists just behave as a combination of the patterns seen above. Next is National:

Effective party list for the National Party

Effective party list for the National Party.

The sweet spot on the list here belongs to Ravi Musuku (Mt. Albert, list ranking #63), Mita Harris (Mangere, #62) and Marc Alexander (Wigram, #60). These are the people your party vote will be going towards getting elected if you party-vote National.

What happened to those on the list ranked #61 (Malcolm Plimmer) and #64 (Jonathan Young)? They’re on the effective list as well, just a bit further down. The reason is that for them to be the lowest ranked National candidates to win a list seat they would have to first lose their electorate seats, New Plymouth and Palmerston North, which are National-leaning seats. (Yes, I’m aware that New Plymouth and Palmerston North are not traditionally considered “National-leaning seats.” The individual electorate results are modeled on the principle of Uniform National Swing). This is why you have to take the results in individual electorate seats into account when trying to figure out where your party vote is going!

Next is the Labour Party:

Effective party list for the Labour Party

Effective party list for the Labour Party.

A vote for the Labour Party is most likely going to go to Damien O’Conner (West Coast-Tasman, #32), Grant Robertson (Wellington Central, #34), or Iain Lees-Galloway (Palmerston North, #36). All three face a difficult battle to win their electorate seats on current polling.

Judith Tizard is still on the effective party list in #9 position, with a 5% chance of being the lowest ranked Labour list candidate to win a list seat in Parliament.

Next up, the Progressive Party and the United Future Party:

Effective party list for the Progressive Party

Effective party list for the Progressive Party.

Effective party list for the United Future Party.

Effective party list for the United Future Party.

These two show a pattern very similar to that shown by the ACT Party. In all three cases you have a party leader with a non-100% chance of winning an electorate seat, and a party that is not guaranteed to break the 5% threshold to get seats without the electorate waiver, which results in a non-zero possibility of a wasted vote. In all three cases the wasted vote can be caused by the leader failing to win their electorate seat, or by winning their electorate seat but the party not having a high enough party vote to elect other list MPs along with them.

Finally, the New Zealand First Party.

Effective party list for the New Zealand First party

Effective party list for the New Zealand First party.

Another fairly intuitive result. With NZF polling about 4% at the moment, they are within a margin-of-error or so from breaking the 5% threshold, which, like the Green Party, would see them win list seats even though they are not predicted to win an electorate seat. The sweet spot on the NZF party list corresponds to candidates with list rankings of #6 and #7.

Obviously the results above are purely hypothetical, given that they are based on old party lists from the 2008 General Election that have been modified arbitrarily to take account of retirements and so on during the current term of Parliament. I look forward to whipping up new lists when the finalised party lists for the 2011 General Election are made public.

Read Full Post »