Hi, my name is Bernard and I am a political junkie. I get my fix from election days and polls. As much as a I know these polls are not real proper data, it is all I have to go on. It has been six months since I last had a BC poll hit.
Here are the results from Ipsos and Insights West
Party All of BC Metro Van Van Isle Rest of BC
Ipsos Insit Ipsos Insit Ipsos Insit Ipsos Insit
Liberal 42 34 32 37 32 21 44 35
NDP 36 40 39 39 39 52 28 35
Green 10 14 6 13 25 22 9 13
Cons 11 10 10 9 4 5 18 18
Other 1 2 1 2 0 0 2 1
So what can we make of these numbers? First, there is no clear leader, there is a race. Second, the Greens have a measurable strength on Vancouver Island, high enough that more seats are a realistic target. Third, the moribund BC Conservatives have reasonable strength in the interior and north.
The 10% or so support for the BC Conservatives says to me there is a desire by a part of the BC population for an actual right wing party. This number is in my opinion people parking their support. The party has no leader, no media presence and no obvious grass roots organization but it is polling at 10%.
The NDP has good enough numbers that it could win the next election but the polls are not showing the sort of monster lead like in the run up to the 2013 election. I have to wonder if it is seen as a race if more people will come out and vote for the NDP?
Showing posts with label Election Prediction. Show all posts
Showing posts with label Election Prediction. Show all posts
Thursday, May 12, 2016
Wednesday, March 11, 2015
2015 Election Stock Market for the October Federal Election
Werner Antweiler of the Sauder School of Business at UBC runs an election stock market during federal elections in Canada. On March 1st the market for the fall federal election opened
You can invest up to $1,000 in the market and buy and sell shares in how well the political parties will do in the election. The market is an overall zero sum game.
The concept behind the election stock market is that if people have to put their money where their mouth is they are likely to make their best estimates of the outcome of the election. Can markets be used to predict outcomes of events? With enough people taking part the wisdom of crowds comes into play.
One of the problems with the Sauder School of Business election stock markets is that often there have not been enough people taking part in them making the market not very liquid. It also meant that one or two people with a few hundred dollars could significantly impact the results of the market.
In 2013 the election stock market did not do any better than most pundits in predicting the results of the BC election The market predicted an NDP majority and had the popularity of the parties close to what the polls were showing. In this case I do not think having a lot of traders would have made a dramatic difference to the outcome.
If predicting the election is of interest to you, I highly recommend taking part in the election stock market. I have averaged a 30% return over 7 elections - when I factor in 2013 it falls to 20%.
One useful thing that Werner provides is his Seat Distribution Forecaster. You can enter in your assumptions on what the voters from the 2011 election will do in 2015 You can do the table for the whole nation, parts of the country or individual provinces. The table reflects the new electoral boundaries in use for 2015
This tool has been available for numerous elections in Canada since the late 1990s. Currently you can use it for not only the next federal election, but the next election in Alberta and Saskatchewan
 |
| My holdings as of March 11th |
The concept behind the election stock market is that if people have to put their money where their mouth is they are likely to make their best estimates of the outcome of the election. Can markets be used to predict outcomes of events? With enough people taking part the wisdom of crowds comes into play.
One of the problems with the Sauder School of Business election stock markets is that often there have not been enough people taking part in them making the market not very liquid. It also meant that one or two people with a few hundred dollars could significantly impact the results of the market.
In 2013 the election stock market did not do any better than most pundits in predicting the results of the BC election The market predicted an NDP majority and had the popularity of the parties close to what the polls were showing. In this case I do not think having a lot of traders would have made a dramatic difference to the outcome.
If predicting the election is of interest to you, I highly recommend taking part in the election stock market. I have averaged a 30% return over 7 elections - when I factor in 2013 it falls to 20%.
One useful thing that Werner provides is his Seat Distribution Forecaster. You can enter in your assumptions on what the voters from the 2011 election will do in 2015 You can do the table for the whole nation, parts of the country or individual provinces. The table reflects the new electoral boundaries in use for 2015
 |
| This is the table you enter you assumptions into about how the voters from 2011 will act in 2015 |
 |
| This is one scenario I quickly input for BC. |
Sunday, February 8, 2015
An interesting trend in the EKOS polling numbers from July 23rd to February 3rd
Here is what I find interesting about the trend over the last six months.
In the late summer of 2014 the Liberals had a large lead over both the NDP and Conservatives. Since then there has been a clear upward trend for Conservatives since the end of October. What is interesting is where the Conservative support has come from. The Conservatives rose close to 10 percentage points in the last four months of which the support seems to have come equally from the Liberals and the NDP.
What this highlights is that politics is not a simple left/right spectrum. The public is just as willing to move from supporting Mulcair to Harper as from Trudeau to Harper. The hardcore NDP support would never move to the Conservatives. The people in question are the less politically active. A lot more of them are open considering any of the three major political parties.
When we look at the low point of all three major parties in the last six months and this comes up to 74.6% of the vote. Another 4% seem to fairly securely Bloc and 7% for the Greens. This leaves about 14% that are very fluid.
So where could this 14% go? Any of the three major parties which means any one of them could win the next election. One wildcard to this is if these people decide that Elizabeth May is a breath of fresh air. There is a realistic chance the Greens could finish the election with as much as 20% of the vote.
I think the Conservatives have a upper limit of around 40%, the Liberals of around 38% and the NDP at only 32%.
I honestly do not know how the 2015 election will turn out. I think anyone that projects what the result will be on October 19th is doing little more than make a wishful guess. I think that close to half the seats in the country will be in play in this election.
Monday, July 8, 2013
Westside-Kelowna - what would it take for Christy Clark to lose? Also my prediction of the vote
Is it possible for the NDP to win in Westside-Kelowna, but a whole host of factors have to come into play:
The 2012 Chilliwack-Hope by-election had all the factors come into play and lead to the NDP the winning in what really is one of the most conservative parts of BC. One very important factor that lead to Gwen O'Mahony winning in the by-election was that even though the voter turnout went down the NDP vote not only held but rose. Meanwhile there was some degree of division among the Liberals and the Conservatives.
Chilliwack-Hope was a little bit easier of a target than Westside-Kelowna is.
Assuming the NDP hold their vote, the Conservatives gain a bit and Dayleen Van Rysyk has the impact we get the following:
That is 11,500 votes. If the voter turnout is 80% of what it was in May that would mean 18,000 voters leaving Christy Clark with 6,500 votes, in the danger zone of losing. This is not any sort of prediction but just showing that it is possible for the NDP to win. I honestly do not think that all the factors will come into play. I do think Christy Clark will be elected.
Now to trying to predict this race. First off, we need to get some reasonable estimate of the voter turnout.
Here is how turnout numbers have compared in the by-election compared to the previous general election going back to 1994 - the percentages are not turnout percentage, but percent of the last general election's turnout
- Voter turnout has to be low overall
- Voter turnout in downtown Kelowna has to do better than that on the Westside
- The NDP voters from May have to show up
- The BC Conservatives have to capture some disaffected Liberals
- Where will Dayleen Van Ryswyk get votes from?
- Are people motivated to vote for the premier or against her?
The 2012 Chilliwack-Hope by-election had all the factors come into play and lead to the NDP the winning in what really is one of the most conservative parts of BC. One very important factor that lead to Gwen O'Mahony winning in the by-election was that even though the voter turnout went down the NDP vote not only held but rose. Meanwhile there was some degree of division among the Liberals and the Conservatives.
Chilliwack-Hope was a little bit easier of a target than Westside-Kelowna is.
Assuming the NDP hold their vote, the Conservatives gain a bit and Dayleen Van Rysyk has the impact we get the following:
- NDP Carol Gordon 7,000
- Cons Sean Upshaw 3,000
- Ind Dayleen Van Ryswyk 1,000
- Others 500
That is 11,500 votes. If the voter turnout is 80% of what it was in May that would mean 18,000 voters leaving Christy Clark with 6,500 votes, in the danger zone of losing. This is not any sort of prediction but just showing that it is possible for the NDP to win. I honestly do not think that all the factors will come into play. I do think Christy Clark will be elected.
Now to trying to predict this race. First off, we need to get some reasonable estimate of the voter turnout.
Here is how turnout numbers have compared in the by-election compared to the previous general election going back to 1994 - the percentages are not turnout percentage, but percent of the last general election's turnout
- Chilliwack-Hope 2012 85.2%
- Parksville-Qualicum 1998 78.3%
- Delta South 1999 77.7%
- Surrey-Panorama 2004 77.3%
- Abbotsford 1995 70.9%
- Vancouver Point Grey 2011 69.5%
- Matsqui 1994 69.5%
- Surrey-White Rock 1997 65.3%
- Port Moody-Coquitlam 2012 60.1%
- Vancouver-Burrard 2008 48.5%
- Vancouver-Quilchena 1994 47.5%
- Vancouver-Fairview 2008 43.9%
This is an average turnout of 66.1% of the previous general election - this is higher than what happens federally in by-elections. I calculated out the average change in turnout over 50 federal by-elections and the result was 59.4% of the previous general election. If we factor out the three by-elections in Vancouver with the lowest turnout we get to an average turnout of 72.6% of the previous general election.
Using 72.6% of the turnout of the recent general election we end up with roughly 16,000 people likely to vote. Based on a decent advance turnout, this seems more likely than not now.
The existence of five more candidates than in the recent general election will take some of the votes, my estimate is 1,000 votes total. This leaves 15,000 votes for the Liberals, NDP and Conservatives.
I think the BC Conservatives will see vote drop to 1,500 total because I see little or no evidence of any sort of strong campaign from them on the ground in the riding.
My best guess if that the May NDP voters are motivated to show up again, I think they can hold 6,000 to 6,500 votes which means we are at the following result:
Christy Clark Lib 7,250 45.31%
Carol Gordon NDP 6,250 39.06%
Sean Upshaw BCCP 1,500 9.38%
The Rest 1,000 6.25%
So my prediction is that the race will look close at the end of the day but not as close as the 2011 Vancouver-Point Grey by-election was.
Riding Polling
There have been two polls by Forum of the riding, though the sample size was very small in each case I do do not have the details on the poll.
June 6th - 354 surveyed, no idea how many said they vote or what the sample looked like
Liberal 57%
NDP 31%
BCCP 8%
Others 4% - I am assuming this because I did not see it reported
July 3rd - 308 surveyed, once again no idea of the details and this time I do not have numbers for the
Liberal 55%
NDP 36%
Riding Polling
There have been two polls by Forum of the riding, though the sample size was very small in each case I do do not have the details on the poll.
June 6th - 354 surveyed, no idea how many said they vote or what the sample looked like
Liberal 57%
NDP 31%
BCCP 8%
Others 4% - I am assuming this because I did not see it reported
July 3rd - 308 surveyed, once again no idea of the details and this time I do not have numbers for the
Liberal 55%
NDP 36%
Wednesday, March 27, 2013
Someone is trying to boost the BC Conservatives in the Sauder Prediction Markets
Over the last few days someone has been taking part in the Sauder School of Business Prediction Market for the 2013 BC Election and they have been buying up BC Conservative popular vote and seat shares for very high prices. What this highlights is that as long as there is not a lot of money and few active traders in the market it is not that hard to manipulate the market.
As of yesterday the market only had $7,879 invested and the trading volumes have been less than 100 shares in each of the three markets till about three days ago. Today we saw 450 Conservative seat shares change hands and over 200 in the popular vote market. This is much higher than on any other day before but still only amounts to a total of $80 to $85 in value.
Since the Sauder School Prediction Markets have partnered with the Global and Mail, the small trading volumes are getting press everyday. As I writing this the market has the BC Conservatives at 17.6% and predicted to win 8 seats, both of these results are not even close to what all the other evidence is saying about the election. The fact that a small amount of trades can influence the results of the market is not good.
Until there a lot more active traders in the market, it will be possible to significantly manipulate the market without a lot of money.
The bright side of this for people is that there are people out there willing to part with their money for the benefit of their party. There is good money to be made in NDP shares in all three markets, a 20% return between now and election day is reasonable. If you think the Liberals might get over 30% of the vote and take 25 seats, there is money you can make.
If you were to only invest in buying NDP majority government shares, something that everyone that knows anything about BC politics knows is going to happen, you will make a 6.1% return on election day.
The more people that take part in the market, the more the results will reflect the wisdom of the crowd. Ideally there would be 1000 to 2000 investors putting in a total of $100,000.
As of yesterday the market only had $7,879 invested and the trading volumes have been less than 100 shares in each of the three markets till about three days ago. Today we saw 450 Conservative seat shares change hands and over 200 in the popular vote market. This is much higher than on any other day before but still only amounts to a total of $80 to $85 in value.
Since the Sauder School Prediction Markets have partnered with the Global and Mail, the small trading volumes are getting press everyday. As I writing this the market has the BC Conservatives at 17.6% and predicted to win 8 seats, both of these results are not even close to what all the other evidence is saying about the election. The fact that a small amount of trades can influence the results of the market is not good.
Until there a lot more active traders in the market, it will be possible to significantly manipulate the market without a lot of money.
The bright side of this for people is that there are people out there willing to part with their money for the benefit of their party. There is good money to be made in NDP shares in all three markets, a 20% return between now and election day is reasonable. If you think the Liberals might get over 30% of the vote and take 25 seats, there is money you can make.
If you were to only invest in buying NDP majority government shares, something that everyone that knows anything about BC politics knows is going to happen, you will make a 6.1% return on election day.
The more people that take part in the market, the more the results will reflect the wisdom of the crowd. Ideally there would be 1000 to 2000 investors putting in a total of $100,000.
Monday, March 4, 2013
What is rock bottom for the BC Liberals?
I think this question is one worth asking because of a couple of weeks of bad news for the government
It is not only these event but an ongoing set of problems and general sense of malaise about the government. This should all realistically have an impact on the popularity of the Liberals. The danger is that as the party stumbles more people with issues are going to pile on and things are going to spiral downwards. So my question becomes how low can support for the Liberals go?
The last poll to come out was on February 24th from Angus Reid and they were polling February 21st and 22nd so we have not seen if the recent events are having an impact on support for the Liberals. I would be very surprised if the problems did not reduce the party support to 25% or less.
It is hard to know where the bottom is but when you look at some examples from BC political history it gives you some parameters:
Election Party Pct of the vote
1972 Social Credit 31.2% they lost 33% of their 1969 vote
1991 Social Credit 24.1% they lost 51% of their 1986 vote
1993 Federal PCs in BC 13.5% they lost 62% of their 1988 vote
1993 Federal NDP in BC 15.5% they lost 58% of their 1988 vote
2001 NDP 21.6% they lost 46% of their 1996 vote
The average drop in a wipe out election is 50% of the percentage the party achieved in the previous election. This means it is entirely reasonable to contemplate the BC Liberals being at 23% on election day. The range of the drops in past elections translate into 17.4% to 30.7% for the BC Liberals. Finishing the election with less than 20% of the vote is now a possibility for the Liberals. Some polling in 2012 indicated that the Liberals were in the 20% to 24% range, I personally was not convinced that those numbers were an accurate reflection of the public mood at the time and I still remain skeptical of them. Given the recent events I now would not be surprised to see the Liberals poll at 20% to 25% in the next polls to come out.
At 23% the Liberals are only realistically safe in two ridings and would be lucky to win five in total.
At 20% the Liberals have a shot at two seats and could be shut out.
At 17% the Liberals would likely be shut in the election.
If the Liberals do end up at 23% or less on election day it is not even certain they would be the official opposition. At that low a support level it becomes possible for the Greens to potentially win more seats than the Liberals and be the official opposition. Possibly even the BC Conservatives could win more seats than the BC Liberals and form the official opposition.
If the BC Liberals are reduced to 375,000 votes, what are the other 375,000 2009 Liberal voters going to do in this election? It is those people that will determine what sort of an opposition we will have in BC and they could hand us something entirely new and very different.
- A weak throne speech
- A unbelievable budget - only 12% of people in an Ipsos poll believed it was balanced
- Blurring of the lines between work for the province and the party when trying organize to get the "ethnic" vote
- Pat Bell and the Wood Innovation Centre
- Resignation of riding executive members
- The hamfisted government of BC advertising - ask the under 40s you know what they think of them, the government is driving them away with the ads.
It is not only these event but an ongoing set of problems and general sense of malaise about the government. This should all realistically have an impact on the popularity of the Liberals. The danger is that as the party stumbles more people with issues are going to pile on and things are going to spiral downwards. So my question becomes how low can support for the Liberals go?
The last poll to come out was on February 24th from Angus Reid and they were polling February 21st and 22nd so we have not seen if the recent events are having an impact on support for the Liberals. I would be very surprised if the problems did not reduce the party support to 25% or less.
It is hard to know where the bottom is but when you look at some examples from BC political history it gives you some parameters:
Election Party Pct of the vote
1972 Social Credit 31.2% they lost 33% of their 1969 vote
1991 Social Credit 24.1% they lost 51% of their 1986 vote
1993 Federal PCs in BC 13.5% they lost 62% of their 1988 vote
1993 Federal NDP in BC 15.5% they lost 58% of their 1988 vote
2001 NDP 21.6% they lost 46% of their 1996 vote
The average drop in a wipe out election is 50% of the percentage the party achieved in the previous election. This means it is entirely reasonable to contemplate the BC Liberals being at 23% on election day. The range of the drops in past elections translate into 17.4% to 30.7% for the BC Liberals. Finishing the election with less than 20% of the vote is now a possibility for the Liberals. Some polling in 2012 indicated that the Liberals were in the 20% to 24% range, I personally was not convinced that those numbers were an accurate reflection of the public mood at the time and I still remain skeptical of them. Given the recent events I now would not be surprised to see the Liberals poll at 20% to 25% in the next polls to come out.
At 23% the Liberals are only realistically safe in two ridings and would be lucky to win five in total.
At 20% the Liberals have a shot at two seats and could be shut out.
At 17% the Liberals would likely be shut in the election.
If the Liberals do end up at 23% or less on election day it is not even certain they would be the official opposition. At that low a support level it becomes possible for the Greens to potentially win more seats than the Liberals and be the official opposition. Possibly even the BC Conservatives could win more seats than the BC Liberals and form the official opposition.
If the BC Liberals are reduced to 375,000 votes, what are the other 375,000 2009 Liberal voters going to do in this election? It is those people that will determine what sort of an opposition we will have in BC and they could hand us something entirely new and very different.
Monday, July 30, 2012
Possible May 2013 election results - some thoughts
I made a projection last year which looked better for the BC Liberals and BC Conservatives, my current estimate has the NDP doing better.
I am starting with 1,900,000 voters in 2013. This number is based on the sort of turn out that seems reasonable based on past federal and provincial elections in BC for the last 12 elections.
My best estimate of what happened in the 2009 BC election is that 200,000 or so right wing voters did not vote in that election and are unlikely to vote for the BC Liberals after reluctantly voting for them in the past.
In my opinion it is important to approach the election from total votes and not from percentages. Based on what I think are the current best estimates of support for the political parties, I get to the following:
I do not see any realistic way for the BC Liberals to get any significant amount of the BC Conservative vote. Where I think the growth is possible for the BC Liberals is among the reluctant or occasional voters. I suspect voter turn out will be 1,900,000, but I also think that another 600,000 to 800,000 people who will be considering voting. This is a large enough block of voters to decide the election.
For those of you that point to the polls contradicting my numbers, if you read this blog regularly you will know of my skepticism over polling methodology which means I take them with a grain of salt. Data on past voting patterns and how often and when people vote in elections is at least as important to understanding probable election outcomes. Feel free to disagree, just do you mind doing it politely?
Greens, BC Conservatives, and Independents
These levels of support do not indicate to me that the Greens have any hope of winning a seat. For them to win a seat it would take focusing all their resources on a handful of seats and a lucky break of the votes on election day.
The BC Conservatives could win a couple of seats, but nothing is certain. John Van Dongen is an incumbent and is running in a conservative area, he could be re-elected. John Cummins as leader may win in Richmond East. Beyond that I have trouble seeing any other serious chances of BC Conservative wins.
I lean towards Vicki Huntingdon and Bob Simpson being re-elected as independents.
The one other person that might win a seat is Arthur Hadland in Peace River North. He came a competitive second in 2009 and as far as I understand plans on running again in 2013.
BC NDP
With the sort of scenario as I suggest above, realistically I see the NDP taking 60 to 70 seats. I have the NDP on about 50 safe seats.
BC Liberals
Realistically I see the BC Liberals taking 12 to 23 seats. I am working on a list in order of how safe the Liberal seats are, but it it is not complete and a bit messy. Currently the list has about eight safe Liberals seats and 20 more that are in play.
Safe Seats - Seats in Play
Assuming the scenario as I see it happens next May, 60 of the seats in BC are safe and only 25 are realistically in play.
I am starting with 1,900,000 voters in 2013. This number is based on the sort of turn out that seems reasonable based on past federal and provincial elections in BC for the last 12 elections.
My best estimate of what happened in the 2009 BC election is that 200,000 or so right wing voters did not vote in that election and are unlikely to vote for the BC Liberals after reluctantly voting for them in the past.
In my opinion it is important to approach the election from total votes and not from percentages. Based on what I think are the current best estimates of support for the political parties, I get to the following:
- Party Total Vote Pct
- BC NDP 850,000 44.7%
- BC Liberals 600,000 31.6%
- BC Conservatives 300,000 15.8%
- Greens 150,000 7.9%
I do not see any realistic way for the BC Liberals to get any significant amount of the BC Conservative vote. Where I think the growth is possible for the BC Liberals is among the reluctant or occasional voters. I suspect voter turn out will be 1,900,000, but I also think that another 600,000 to 800,000 people who will be considering voting. This is a large enough block of voters to decide the election.
For those of you that point to the polls contradicting my numbers, if you read this blog regularly you will know of my skepticism over polling methodology which means I take them with a grain of salt. Data on past voting patterns and how often and when people vote in elections is at least as important to understanding probable election outcomes. Feel free to disagree, just do you mind doing it politely?
Greens, BC Conservatives, and Independents
These levels of support do not indicate to me that the Greens have any hope of winning a seat. For them to win a seat it would take focusing all their resources on a handful of seats and a lucky break of the votes on election day.
The BC Conservatives could win a couple of seats, but nothing is certain. John Van Dongen is an incumbent and is running in a conservative area, he could be re-elected. John Cummins as leader may win in Richmond East. Beyond that I have trouble seeing any other serious chances of BC Conservative wins.
I lean towards Vicki Huntingdon and Bob Simpson being re-elected as independents.
The one other person that might win a seat is Arthur Hadland in Peace River North. He came a competitive second in 2009 and as far as I understand plans on running again in 2013.
BC NDP
With the sort of scenario as I suggest above, realistically I see the NDP taking 60 to 70 seats. I have the NDP on about 50 safe seats.
BC Liberals
Realistically I see the BC Liberals taking 12 to 23 seats. I am working on a list in order of how safe the Liberal seats are, but it it is not complete and a bit messy. Currently the list has about eight safe Liberals seats and 20 more that are in play.
Safe Seats - Seats in Play
Assuming the scenario as I see it happens next May, 60 of the seats in BC are safe and only 25 are realistically in play.
Monday, May 14, 2012
Looking at the difference in party support in BC on election day
I thought it would be interesting to see how closely the difference in support between the two main parties correlates to the number seats the governments wins.
Election Major party vote share diff govt seats
2001 - Liberal 57.62% NDP 21.56% +36.06 97.47%
1949 - Coalition 61.35% CCF 35.10% +26.25 81.25%
1945 - Coalition 55.83% CCF 37.62% +18.21 77.08%
1956 - SoCred 45.84% CCF 28.32% +17.52 75.00%
1963 - SoCred 40.83% NDP 27.80% +13.03 63.46%
1969 - SoCred 46.79% NDP 33.92% +12.87 69.09%
1966 - SoCred 45.59% NDP 33.62% +11.97 60.00%
1933 - Liberal 41.74% CCF 31.53% +10.21 72.34%
1975 - SoCred 49.25% NDP 39.16% +10.09 63.64%
1937 - Liberal 37.34% Cons 28.60% +8.74 64.58%
1972 - NDP 39.59% SoCred 31.15% +8.44 69.09%
1991 - NDP 40.71% Liberal 33.25% +7.46 68.00%
1986 - SoCred 49.32% NDP 42.60% +6.72 68.12%
1960 - SoCred 38.83% CCF 32.73% +6.10 75.00%
1983 - SoCred 49.76% NDP 44.94% +4.82 61.40%
2005 - Liberal 45.80% NDP 41.52% +4.28 58.23%
2009 - Liberal 46.02% NDP 42.06% +3.96 57.65%
1979 - SoCred 48.32% NDP 45.99% +2.24 54.39%
1941 - Liberal 32.94% CCF 33.36% -0.42 43.75%
1996 - NDP 39.45% Liberal 41.82% -2.37 52.00%
I left out the 1952 and 1953 elections because a different voting system was used
After the 1941 election the Liberals and Conservatives formed a coalition government
Highlighted dates are elections that saw a change in government
What patterns can I see here? There is a board connection between how much the government won over the second place party by though there are exceptions to this such as in 1963 and 1966 Social Credit underperformed and in 1960 Social Credit and 1996 NDP both overperformed.
What the pattern seems to indicate is that if the NDP manages to finish 15 percentage points ahead of the Liberals on election day, the NDP will likely take about 62 to 65 seats. If they finish 7-8 percentage points ahead they are likely to take 58 to 60 seats. If the NDP is 20 percentage points ahead, this would point to 68 to 69 seats.
Wednesday, March 21, 2012
2013 BC election - some analysis trying to figure out what might happen
I know I am doing this without knowing what sort of campaign any of the parties will run, but based on what I have seen, no one is going to anything out of the ordinary.
My estimate is that 1,700,000 to 1,800,000 people will vote in the election.
My estimates of the vote totals of the parties:
The NDP gains some green and 2009 Liberal voters
The Liberals lose a few to the NDP, 50,000 to 75,000 to the Conservatives and 75,000 to 125,000 to non voting
The BC Conservatives gain about 150,000 2009 non-voters and 50,000 to 75,000 BC Liberal voters.
Here is my current estimate of the the 2013 election result
What would it take for the Liberals to win? They need to find 300,000 centre to centre right voters and get them to vote. They need to give them a reason to vote for the Liberals. There will be something like 3,000,000 people considering voting in the next election but 1,200,000 to 1,300,000 will end up not voting. Of these people, a large portion are politically from the centre or centre right. There will likely be 400,000 to 500,000 that have voted in one of the past five elections for the BC Liberals that will not voting in 2013 because there is no party they want to vote.
There are another 400,000 to 500,000 that are more right wing than anything but respond to a strong populist message. Without that populist message they are unlikely to come out and vote. Should John Cummins become a serious populist threat, I would say all bets are off on what will happen in the next election. To date I do not see him being able to engage that populist streak.
Adding significantly the vote turn out is not unrealistic. The 2011 Federal election saw 1,872,636 people voted in BC versus only 1,640,542 in the 2009 BC election. A voter turn out of 1,900,000 to 2,000,000 can be achieved with the right positive campaign that gives people a reason to vote.
In the back rooms of politics in Canada the big goal in the elections for the last generation has normally been finding ways to get people not come out and vote. The primary role of all negative campaigning is to get people not to vote. The parties work hard to suppress the vote of the competition and this is the fundamental reason voter turn out has been falling. Just so everyone is clear, negative campaigning causes almost no one to change their mind who they vote for, it just makes them lose hope and not vote at all.
There is no indication that there is any significant space for the NDP to grow their total vote to over 800,000. It is in the interests of the NDP to suppress potential Liberal voters but at the same there is little hope that the Liberals can significantly suppress the NDP vote in 2013. There is also little hope the Liberals can suppress the Conservative vote. Going after Adrian Dix or John Cummins is not adding any votes for the Liberals, they need a positive vision for the province to get more people to vote Liberal in 2013.
When Elections BC surveyed the people that did not vote in 2009, a significant portion of the non-voters decided not to vote on election and decided not to vote because there was no one they wanted to vote for. The non-voters are the people looking for the positive vision from the parties and they are the low hanging fruit that the parties are leaving on the branch to rot.
My estimate is that 1,700,000 to 1,800,000 people will vote in the election.
My estimates of the vote totals of the parties:
- NDP 750,000 to 775,000 - abt 44% of the vote
- Liberals 525,000 to 550,000 - abt 31% of the vote
- Conservatives 250,000 to 300,000 abt 16% of the vote
- Greens 125,000 to 175,000 abt 8.5% of the vote
- the rest 25,000 abt 1.5% of the vote
The NDP gains some green and 2009 Liberal voters
The Liberals lose a few to the NDP, 50,000 to 75,000 to the Conservatives and 75,000 to 125,000 to non voting
The BC Conservatives gain about 150,000 2009 non-voters and 50,000 to 75,000 BC Liberal voters.
- I am assuming that Bob Simpson and Vicki Huntingdon will be re-elected.
- The 34 sitting NDP MLAs should all be re-elected.
- For the Liberals much will depend on how many incumbents run, in many places the lack of an incumbent will harm the Liberal re-election chances.
- The Conservatives suffer from not having any concentration of their vote.
Here is my current estimate of the the 2013 election result
- NDP 58 (54-63)
- Liberal 23 (20-27)
- Conservatives 2 (0-5)
- Independents 2
- Greens 0
What would it take for the Liberals to win? They need to find 300,000 centre to centre right voters and get them to vote. They need to give them a reason to vote for the Liberals. There will be something like 3,000,000 people considering voting in the next election but 1,200,000 to 1,300,000 will end up not voting. Of these people, a large portion are politically from the centre or centre right. There will likely be 400,000 to 500,000 that have voted in one of the past five elections for the BC Liberals that will not voting in 2013 because there is no party they want to vote.
There are another 400,000 to 500,000 that are more right wing than anything but respond to a strong populist message. Without that populist message they are unlikely to come out and vote. Should John Cummins become a serious populist threat, I would say all bets are off on what will happen in the next election. To date I do not see him being able to engage that populist streak.
Adding significantly the vote turn out is not unrealistic. The 2011 Federal election saw 1,872,636 people voted in BC versus only 1,640,542 in the 2009 BC election. A voter turn out of 1,900,000 to 2,000,000 can be achieved with the right positive campaign that gives people a reason to vote.
In the back rooms of politics in Canada the big goal in the elections for the last generation has normally been finding ways to get people not come out and vote. The primary role of all negative campaigning is to get people not to vote. The parties work hard to suppress the vote of the competition and this is the fundamental reason voter turn out has been falling. Just so everyone is clear, negative campaigning causes almost no one to change their mind who they vote for, it just makes them lose hope and not vote at all.
There is no indication that there is any significant space for the NDP to grow their total vote to over 800,000. It is in the interests of the NDP to suppress potential Liberal voters but at the same there is little hope that the Liberals can significantly suppress the NDP vote in 2013. There is also little hope the Liberals can suppress the Conservative vote. Going after Adrian Dix or John Cummins is not adding any votes for the Liberals, they need a positive vision for the province to get more people to vote Liberal in 2013.
When Elections BC surveyed the people that did not vote in 2009, a significant portion of the non-voters decided not to vote on election and decided not to vote because there was no one they wanted to vote for. The non-voters are the people looking for the positive vision from the parties and they are the low hanging fruit that the parties are leaving on the branch to rot.
Friday, March 16, 2012
The pending federal election - what might happen?
The odds of seeing a federal election this year are rising. The whole robocalls things is getting to a point where there may be no choice other than an election. Let us say we end up with an election this fall because of it, how would it look?
First, the new ridings will not be in place yet, we will still have the same 308 that we have now.
Second, I am ignoring the polling done of late because I am not certain it is a realistic reflection of what might happen in an election caused by the robocalls scandal. I am not making a projection on what the polls are saying, but on how I think an election caused by the scandal would play out.
Third, how ready at the parties for an election?
The Conservatives are financially best prepared for an election. They can run full campaigns in 308 ridings across the country. They are also the incumbents in the most ridings in Canada. Overall the Conservatives were a significant factor in 239 ridings in 2011.
The Liberals are out of money and have no organization in most of the country. In 2011 the Liberals were a factor in only 104 seats. If the Liberals were to win every seat one of those seats, it would not bring them close enough to anything. The Liberals are more likely to lose more seats if there were an election in 2012.
The Bloc, I have a hard time judging them because I am not really in touch with Quebec politics well enough. I think they will be making gains if there were an election this year.
The NDP, they traditionally have not been a factor in more than 70-80 ridings in the country. In 2011 they were a factor in 232 ridings. With the leadership race the party has built itself up in many parts of Canada were there was not much in place in the past.
Fourth, how would this election look?
I am going to assume Thomas Mulcair will be the NDP leader.
Conservatives lose seats in BC, Ontario and Quebec. Possibly on the prairies as well.
The Liberals lose their seats west of Ontario in a Conservative NDP squeeze. In Ontario they lose seats to the NDP. Quebec, they may lose some more to the NDP.
I am going to assume the Bloc manages to regain some seats, all of them from the NDP.
The NDP will be the the overwhelming choice of people pissed off with the Conservatives. I see them gaining seats across the country from the Conservatives and Liberals but lose a few Quebec ones to the Bloc.
Popular Vote Estimate
Seat Estimate
First, the new ridings will not be in place yet, we will still have the same 308 that we have now.
Second, I am ignoring the polling done of late because I am not certain it is a realistic reflection of what might happen in an election caused by the robocalls scandal. I am not making a projection on what the polls are saying, but on how I think an election caused by the scandal would play out.
Third, how ready at the parties for an election?
The Conservatives are financially best prepared for an election. They can run full campaigns in 308 ridings across the country. They are also the incumbents in the most ridings in Canada. Overall the Conservatives were a significant factor in 239 ridings in 2011.
The Liberals are out of money and have no organization in most of the country. In 2011 the Liberals were a factor in only 104 seats. If the Liberals were to win every seat one of those seats, it would not bring them close enough to anything. The Liberals are more likely to lose more seats if there were an election in 2012.
The Bloc, I have a hard time judging them because I am not really in touch with Quebec politics well enough. I think they will be making gains if there were an election this year.
The NDP, they traditionally have not been a factor in more than 70-80 ridings in the country. In 2011 they were a factor in 232 ridings. With the leadership race the party has built itself up in many parts of Canada were there was not much in place in the past.
Fourth, how would this election look?
I am going to assume Thomas Mulcair will be the NDP leader.
Conservatives lose seats in BC, Ontario and Quebec. Possibly on the prairies as well.
The Liberals lose their seats west of Ontario in a Conservative NDP squeeze. In Ontario they lose seats to the NDP. Quebec, they may lose some more to the NDP.
I am going to assume the Bloc manages to regain some seats, all of them from the NDP.
The NDP will be the the overwhelming choice of people pissed off with the Conservatives. I see them gaining seats across the country from the Conservatives and Liberals but lose a few Quebec ones to the Bloc.
Popular Vote Estimate
- NDP 37% (32%-40%)
- Conservatives 34% (30-36%)
- Liberals 17% (12-20%)
- Bloc 7% (6-8%)
- Greens 4% (4%)
Seat Estimate
- NDP 140 (120-150)
- Conservatives 130 (125-140)
- Liberals 20 (5-30)
- Bloc 17 (5-25)
- Greens 1 (1)
I think the most likely out come of a 2012 robocall caused election would be an NDP minority government if Thomas Mulcair is the NDP leader.
The core messaging of the election will be from the NDP about the scandal and the Conservatives about the socialist hordes at the gates. These two parties will crowd out anyone else in the election. A 2012 election would be the most ideological in Canadian history. It would not impossible for the election to play our with the Bloc and Liberals doing worse than I am projecting. If that were to happen, there is a potential for an NDP majority government. For minority governments to occur you need either a very close election where a small party holds the balance or four or more parties winning enough seats to get party status.
Given the scale of the Conservative party resources and demographics of who is most likely to vote, they have a certain degree of certainty on a large batch of seats. The federal Liberals are too far to the left and too disorganized to become a home for disgruntled Conservatives.
If someone else were to lead the NDP, I suspect we would see the Bloc come back with a lot of seats and a parliament that looks something like the 2006 one expect the NDP would in there with 100 seats and the Liberals with less than 30. A weak Conservative minority is what I think would happen.
Saturday, February 11, 2012
My Current Estimate of a BC Election
We are still more than a year away from an election but here is my current estimate of what I think the election will look like:
There is a broad range in these numbers and I have no idea what is coming in the world of BC politics that could change this. For the moment this is what I am going with.
I think the NDP will win, but I do not think a landslide is on the cards. Nothing like 1972 or 1991 when the NDP won with a landslide. I do think this election will mark the highest percentage of the vote the NDP has achieved when it has won the election. The NDP achieved 39.59% in 1972, 40.71% in 1991 and 39.45% in 1996.
The quirks of our electoral system has meant that the NDP has been more popular when it lost - 45.99% in 1979, 44.94% in 1983, 42.60% in 1986, 41.52% in 2005 and 42.06% in 2009.
I can not find anything that shows me the party is any more popular that it has been in the last couple of elections and what data I can glean from the polls shows the party seems to be gaining support in the areas it has already been winning in. Specifically the NDP seems to be gaining on Vancouver Island, a place they really do not need a lot more support.
I see the NDP's 90,000 vote increase coming as follows - 50,000 2009 non-voters, 25,000 2009 BC Liberal voters, and 15,000 2009 BC Green voters.
The BC Liberals have lost supporters to the BC Conservatives, but I think on election day the bigger impact will be the Liberals that decide to stay home and not vote. They are 150,000 votes lower than in 2009. 50,000 I assigned to the Conservatives, 25,000 to the NDP and 75,000 to non-voters.
Can the BC Liberals recover? I think it is not impossible if the premier switches to a positive campaign to convince centrist voters in BC to vote. There are a lot of people that did not vote in 2009 that are at heart Liberals. If only one in three of the people that intended to vote in 2009 but changed their mind on election day did actually vote in the next election for the Liberals, that is enough to make the election a dead heat. I am assuming 1,825,000 people will vote, but there will be something like 2,800,000 people that will be intending to vote in 2013. There are more than enough votes there for a well done positive campaign to allow for a serious increase in the vote. I see Christy Clark as likely the politician with the most likability in a long time to get people to vote if she gives them a reason to come out and vote.
Though, given the approach by the Liberals to date to focus on convincing NDP and Conservative voters not to vote in the next election, I currently do not think that Christy Clark will be able to get any large number of non-voters in 2009 to come out and vote for her party.
Most of the support for the BC Conservatives is coming from non-voters in 2009, a total of 215,000. The polling data I can glean is not very positive for the BC Conservatives winning a lot of seats. They do not seem to be building a core of enough support in any region that I would expect them to start winning a lot of seats. If I was seeing data that showed the BC Conservatives running at well over 20% in the interior, then I would expect them to start winning seats. The three seats I see them winning are Peace River North, Chilliwack, and Chilliwack Hope.
I see nothing that indicates to me that either Vicki Huntingdon or Bob Simpson will be defeated in the general election. Delta South is simply not on the radar for the NDP and the Liberals are unlikely to do as well as in 2009. Meanwhile Bob Simpson has done a very good job of building a following in his neck of the woods. I can not see the NDP, Liberals or Conservatives taking much of the vote.
As to the others, the host of minor parties like BC Refederation, BC First and the 7-12 parties, 10,000 votes is all I see. Since none of them seem to actually be doing any active prep for the election, I am assuming their combined vote will not change from 2009. If they were smart, they would be spending every penny they have on the two coming by-elections and building campaign teams of 250 to 500 people, but they are not doing this.
- Party total votes percentage Seats
- NDP 780,000 42.7% 51
- Liberals 600,000 32.9% 29
- Cons 300,000 16.4% 3
- Greens 115,000 6.3% 0
- Ind 20,000 1.1% 2
- Others 10,000 0.5% 0
- Total 1,825,000
There is a broad range in these numbers and I have no idea what is coming in the world of BC politics that could change this. For the moment this is what I am going with.
I think the NDP will win, but I do not think a landslide is on the cards. Nothing like 1972 or 1991 when the NDP won with a landslide. I do think this election will mark the highest percentage of the vote the NDP has achieved when it has won the election. The NDP achieved 39.59% in 1972, 40.71% in 1991 and 39.45% in 1996.
The quirks of our electoral system has meant that the NDP has been more popular when it lost - 45.99% in 1979, 44.94% in 1983, 42.60% in 1986, 41.52% in 2005 and 42.06% in 2009.
I can not find anything that shows me the party is any more popular that it has been in the last couple of elections and what data I can glean from the polls shows the party seems to be gaining support in the areas it has already been winning in. Specifically the NDP seems to be gaining on Vancouver Island, a place they really do not need a lot more support.
I see the NDP's 90,000 vote increase coming as follows - 50,000 2009 non-voters, 25,000 2009 BC Liberal voters, and 15,000 2009 BC Green voters.
The BC Liberals have lost supporters to the BC Conservatives, but I think on election day the bigger impact will be the Liberals that decide to stay home and not vote. They are 150,000 votes lower than in 2009. 50,000 I assigned to the Conservatives, 25,000 to the NDP and 75,000 to non-voters.
Can the BC Liberals recover? I think it is not impossible if the premier switches to a positive campaign to convince centrist voters in BC to vote. There are a lot of people that did not vote in 2009 that are at heart Liberals. If only one in three of the people that intended to vote in 2009 but changed their mind on election day did actually vote in the next election for the Liberals, that is enough to make the election a dead heat. I am assuming 1,825,000 people will vote, but there will be something like 2,800,000 people that will be intending to vote in 2013. There are more than enough votes there for a well done positive campaign to allow for a serious increase in the vote. I see Christy Clark as likely the politician with the most likability in a long time to get people to vote if she gives them a reason to come out and vote.
Though, given the approach by the Liberals to date to focus on convincing NDP and Conservative voters not to vote in the next election, I currently do not think that Christy Clark will be able to get any large number of non-voters in 2009 to come out and vote for her party.
Most of the support for the BC Conservatives is coming from non-voters in 2009, a total of 215,000. The polling data I can glean is not very positive for the BC Conservatives winning a lot of seats. They do not seem to be building a core of enough support in any region that I would expect them to start winning a lot of seats. If I was seeing data that showed the BC Conservatives running at well over 20% in the interior, then I would expect them to start winning seats. The three seats I see them winning are Peace River North, Chilliwack, and Chilliwack Hope.
I see nothing that indicates to me that either Vicki Huntingdon or Bob Simpson will be defeated in the general election. Delta South is simply not on the radar for the NDP and the Liberals are unlikely to do as well as in 2009. Meanwhile Bob Simpson has done a very good job of building a following in his neck of the woods. I can not see the NDP, Liberals or Conservatives taking much of the vote.
As to the others, the host of minor parties like BC Refederation, BC First and the 7-12 parties, 10,000 votes is all I see. Since none of them seem to actually be doing any active prep for the election, I am assuming their combined vote will not change from 2009. If they were smart, they would be spending every penny they have on the two coming by-elections and building campaign teams of 250 to 500 people, but they are not doing this.
Wednesday, January 4, 2012
580,000 people in BC decided on election day, May 12th 2009, not to vote
If you are a political junkie and have not read "Who Heads to the Polls?: Exploring the Demographics of Voters in British Columbia" I highly recommend it. It was published in March of 2010 by Elections BC. I also suggest you read "Motivations and Barriers: Exploring Voting Behaviour in British Columbia" published January 2010. Also interesting is the "2009 Elections BC Post-Election/Non-Voter Satisfaction Survey Executive Summary Report."
The three reports uncover some interesting information about who votes. The assumptions that political junkies have made about how elections work is not really true at all as these reports show that a lot of people vote some of the time and very few people never vote.
711,158 registered voters that voted in 2005 election did not vote in 2009 election. Meanwhile 393,073 people that voted in 2009 did not vote in 2005. There were 1,257,402 who voted in both elections and 633,829 voted in neither election. This means the almost 80% of the BC population voted in at least one of the 2005 and 2009 elections.
1,104,231 people voted in only one election versus 1,257,402 people voted in both elections. The inconsistent voters are close to the same as amount as the consistent voters.
When we add in 2001 to the mix, of people able to vote in all three elections, 955,212 voted in all three elections, 708,180 voted in one or two elections, and 183,147 did not vote at all. Only 10% of people that were able to vote in all three elections chose not to vote at all. 90% of the public did vote in at least one of the three provincial elections.
1,148,923 people were not registered as voters in 2001 that were registered in 2009.
When looking only at 2005 and 2009 elections, 96% of people that voted in both the elections planned on voting in 2013. 88% of the people that voted only 2009 were planning on voting in 2013, 68% of the people that skipped 2009 are planning on voting in 2013. And finally 1/3 of the people that did not vote in 2005 or 2009 plan on voting in 2013. Putting this all together and close to 80% of registered voters intent to vote in 2013.
Of all non-voters in 2009, 63% had voted in the 2008 federal election and 62% had voted in the 2005 provincial election. This means the majority of the non-voters in 2009 had recently voted.
Most interesting is when people decided not to vote. 31% decided before the election or did not even consider voting. 24% decided during the campaign not to vote and finally, this is the stunning number, 45% of the non-voters in 2009 decided not to vote on election day. This means about 580,000 people in BC decided on election day not to vote.
These are 580,000 people that up until the day before the election intended to vote. Those people who were planning on voting but at the last minute decided not to would have been enough votes to let almost every second place candidate in 2009 win the seat.
So why are these people not voting?
The two top issues are engagement and person circumstances.
Of personal circumstances, which was 44% of the people that voted in 2005 but did not vote in 2009, 43% that set said they were too busy, 21% were ill or physically unable to vote and 28% were away from home.
Of people that did not vote in 2005 or 2009, 41% of the did not vote because they were not engaged in the 2009 election. Of that set 30% were not interested in the election, 60% had some sort of dislike for politics and politicians.
What this all says to me is that you win and lose elections based on getting occasional voters to go to the polls and not about trying to get consistent voters to change their minds. The 580,000 people that decided on May 12th 2009 not to vote are all people that are the low hanging fruit. These are people that had planned on voting.
What I take from this is that 90% of the population will consider voting but they did not get enough positive reasons to go to the effort of voting in every election.
How do you motivate the people to go through with voting? Telling them that Christy Clark, or John Cummins or Adrian Dix and their parties are horrible is not going to motivated people to go to the polls.
A well thought and reasonable centre of the road campaign that focuses on the positives and ignores the other parties is the one and only thing I can see that would engage these people to vote in more elections. People a reason to go the polls and negative campaigning simply makes people stay home.
The three reports uncover some interesting information about who votes. The assumptions that political junkies have made about how elections work is not really true at all as these reports show that a lot of people vote some of the time and very few people never vote.
711,158 registered voters that voted in 2005 election did not vote in 2009 election. Meanwhile 393,073 people that voted in 2009 did not vote in 2005. There were 1,257,402 who voted in both elections and 633,829 voted in neither election. This means the almost 80% of the BC population voted in at least one of the 2005 and 2009 elections.
1,104,231 people voted in only one election versus 1,257,402 people voted in both elections. The inconsistent voters are close to the same as amount as the consistent voters.
When we add in 2001 to the mix, of people able to vote in all three elections, 955,212 voted in all three elections, 708,180 voted in one or two elections, and 183,147 did not vote at all. Only 10% of people that were able to vote in all three elections chose not to vote at all. 90% of the public did vote in at least one of the three provincial elections.
1,148,923 people were not registered as voters in 2001 that were registered in 2009.
When looking only at 2005 and 2009 elections, 96% of people that voted in both the elections planned on voting in 2013. 88% of the people that voted only 2009 were planning on voting in 2013, 68% of the people that skipped 2009 are planning on voting in 2013. And finally 1/3 of the people that did not vote in 2005 or 2009 plan on voting in 2013. Putting this all together and close to 80% of registered voters intent to vote in 2013.
Of all non-voters in 2009, 63% had voted in the 2008 federal election and 62% had voted in the 2005 provincial election. This means the majority of the non-voters in 2009 had recently voted.
Most interesting is when people decided not to vote. 31% decided before the election or did not even consider voting. 24% decided during the campaign not to vote and finally, this is the stunning number, 45% of the non-voters in 2009 decided not to vote on election day. This means about 580,000 people in BC decided on election day not to vote.
These are 580,000 people that up until the day before the election intended to vote. Those people who were planning on voting but at the last minute decided not to would have been enough votes to let almost every second place candidate in 2009 win the seat.
So why are these people not voting?
The two top issues are engagement and person circumstances.
Of personal circumstances, which was 44% of the people that voted in 2005 but did not vote in 2009, 43% that set said they were too busy, 21% were ill or physically unable to vote and 28% were away from home.
Of people that did not vote in 2005 or 2009, 41% of the did not vote because they were not engaged in the 2009 election. Of that set 30% were not interested in the election, 60% had some sort of dislike for politics and politicians.
What this all says to me is that you win and lose elections based on getting occasional voters to go to the polls and not about trying to get consistent voters to change their minds. The 580,000 people that decided on May 12th 2009 not to vote are all people that are the low hanging fruit. These are people that had planned on voting.
What I take from this is that 90% of the population will consider voting but they did not get enough positive reasons to go to the effort of voting in every election.
How do you motivate the people to go through with voting? Telling them that Christy Clark, or John Cummins or Adrian Dix and their parties are horrible is not going to motivated people to go to the polls.
A well thought and reasonable centre of the road campaign that focuses on the positives and ignores the other parties is the one and only thing I can see that would engage these people to vote in more elections. People a reason to go the polls and negative campaigning simply makes people stay home.
Friday, October 7, 2011
Newfoundland Election
I am primarily interested in this election because Steve Kent is running for re-election in Mount Pearl North. He should do well as he took over 85% of the vote in 2007 election. Good Luck Steve!
So there have been two polls for Newfoundland released this week and one last week.
So there have been two polls for Newfoundland released this week and one last week.
- Pollster PCs NDP Libs Date MOE
- Environics 54 33 13 Oct 4 3.7%
- CRA 59 25 16 Oct 3 3.5%
- MarketQuest Omnifacts 54 33 13 Sept 30 4.6%
Looking at all three bell curves of the polls, the numbers for all the parties that fit within them are:
- PCs 56.5%
- NDP 29.5%
- Libs 14%
With only three data points, these numbers have to be seen as rough estimates of support.
So what does this mean for the election? The numbers are close to what I looked at for possible outcomes in the election on September 20th. The biggest change is that the Liberals are slightly lower in support and the PCs slightly higher.
The Liberals seem to be uniformly in trouble through out the province, all that could save them at this point is having their two incumbent MHAs running for re-election win again, though one of them was elected in a by-election. You can also never count out a leader of a party, especially one that has won his seat in past elections.
The few data entrails I can glean from the polls seems to say to my that the NDP has picked up most of the Liberal support in St John's and the Avalon. Of the 23 ridings in the region, the NDP could be a threat in about eight seats. They are also a threat in Labrador West.
My prediction at the moment of this election, which is close to last week's, is as follows:
- PCs 41 (38-43)
- NDP 6 (5-9)
- Liberals 1 (0-3)
The only change is that I think the NDP range has risen from 3-8 seats to 5-9 seats. The PCs will do well in the suburbs of St John's, but looking at the polling the NDP seems to be a strong second in St John's itself and in parts of the Avalon and Burin peninsulas.
Wednesday, October 5, 2011
Ontario Election and polling
I look at the polling numbers and I am not certain which ones are giving the best data. Between September 29th and October 4th there were 10 polls in the field.
Here are the ranges from the 10 polls
- Nanos - 4
- EKOS - 3
- Ipsos Reid - 1
- Abacus Data - 1
- Angus Reid - 1
Here are the ranges from the 10 polls
- Liberals - 33% to 41% (8 percentage points)
- PCs - 29.1% to 36.4% (7.3 percentage points)
- NDP - 22.7% to 26.8% (4.1 percentage points)
- Greens - 1.2% to 7% (5.8 percentage points)
The results should all be much closer together. Total range should not be more than about 5 percentage points for the Liberals and should really be more like 3 percentage points if the methodologies in use are working well. Different companies and different polling mechanisms should be fairly close together.
Could the variation of the polling numbers be due to the campaign? In my opinion this is not the case. The lowest and highest results for the PCs come from polls that were in the field on more or less the same days. The Liberal results from the Angus Reid and Abacus Data polls were 4 percentage points apart and released on the same day. If you assume the real support is half between the two the odds of the results from the two polls happening is less than 1 in 400. The only answer is that some of the pollsters methodologies are flawed. Though who would that be?
There are more sources of error going on than just the statistical measure of confidence.
One could average of all the polls, but this means we build in a larger error as the accurate modelling is weakened by the inaccurate modelling.
So, in my parsing of all the bits and pieces from the various polls and factoring for likely and unlikely voters and some more factors, I come to roughly the following support levels:
Liberals - 38.5%
PCs - 34.0%
NDP - 22.5%
Greens - 4.0%
My numbers are higher for the PCs and the Liberals than Eric at ThreeHundredEight. I think the main reason there is a difference is because I am assuming the NDP will get less of the popular vote than the polls because more non-voters are saying they want to support the NDP.
My prediction at the moment of the Ontario election:
- Liberals 52
- PCs 38
- NDP 17
Much will depend on the close races fall out and the Liberals could win a majority.
If it is a minority government, this would be sixth minority government. Past ones were in:
- 1919 when the United Farmers of Ontario came out of nowhere to win the most seats
- 1943 which was the first win by the PCs in their 42 run of government. The CCF came out of nowhere to win official opposition
- 1975 first Davis minority government
- 1977 second Davis minority government
- 1985 Conservatives won the most seats but an agreement between the NDP and Liberals allowed David Peterson to become premier.
Thursday, September 15, 2011
The BC Conservatives have arrived - it really was inevitable
With the new Can't Trust Cummins campaign from the BC Liberals, the BC Conservatives seem to have arrived on the political scene in BC. By all measures the BC Conservatives have to be seen as the #3 party in BC now.
I have a habit of paying attention to the fringe parties in politics provincially and federally, so I have been watching the rise of the BC Conservatives over the last year. It was only last year that the party was little more than a joke and was suffering yet another internal division. But here we are and the party seems to be solid and gaining real support.
It was not that long ago that few commentators were not taking the BC Conservatives seriously but I have thought was an error for a long time. There is a significant conservative constituency in BC that simply can not vote BC Liberal - they are not voting. I see the evidence of this through the differing turnouts in Provincial and Federal elections in BC in the last several elections.
Federal election turnouts are underlined
I have a habit of paying attention to the fringe parties in politics provincially and federally, so I have been watching the rise of the BC Conservatives over the last year. It was only last year that the party was little more than a joke and was suffering yet another internal division. But here we are and the party seems to be solid and gaining real support.
It was not that long ago that few commentators were not taking the BC Conservatives seriously but I have thought was an error for a long time. There is a significant conservative constituency in BC that simply can not vote BC Liberal - they are not voting. I see the evidence of this through the differing turnouts in Provincial and Federal elections in BC in the last several elections.
Federal election turnouts are underlined
- Year Total Votes Turnout
- 2011 1,879,304 60.4% - this is 240,000 more than 2009
- 2009 1,640,542 50.99% - this is lower than previous 3 fed elections
- 2008 1,799,892 60.1%
- 2006 1,827,183 63.7%
- 2005, 1,762,450 59.19%
- 2004 1,739,999 63.3%
- 2001 1,591,306 55.44%
- 2000 1,621,101 63.0%
- 1997 1,529,139 65.6%
- 1996 1,582,704 59.11%
Turnout is measured slightly differently provincially and federally because they do not use the same voter list.
As you can see voter turn out in the three federal and three provincial elections from 1996 to 2005 are roughly the same. when one compares the matching federal and provincial elections. It is in 2009 that BC took a dive in the provincial election total vote.
So who is missing? I believe it is mainly federal Conservative voters. Honestly, the socially conservative right wing was never going to vote Liberal long term. The BC Liberal party is however you want to measure it a Liberal party in the Canadian tradition, the policies of the Gordon Campbell government was more or less in political sync with the federal Liberal party under Jean Chretien and Paul Martin.
If the 2009 election had a similar turn as the 2011 federal election, the BC Liberals should have attracted about 861,000 votes, not the 751,661 they did manage to achieve. This is close to 110,000 missing voters.
A further dynamic I believe going was on in 2009 was a number of 2005 BC NDP voters that vote Federal Liberal voting for the BC Liberals as well a significant number of 2005 Green voters moved over as well. 2005 supports of the two parties seem to have contributed about 100,000 votes to the BC Liberals in 2009. Before you say it, I do not have rigorous data to prove this, just a lot of back of the envelope calculations using results in federal and provincial ridings in the six elections from 2004 to 2011.
Something like 210,000 2005 BC Liberal voters did not vote for the party in 2009. So where did they go? I believe about 40,000 either voted for the BC Conservatives or one of three independents, Vicki Huntingdon who won in Delta South, Arthur Hadland in Peace River North, and David Marley in West Vancouver Capilano. The other 170,000 stayed home.
I also see some 50,000 2005 NDP supporters from 2005 stayed home in 2009 because they could not vote for the 2009 version of the party.
What all this says to me is that if a BC Conservative party runs a full slate, bringing 100,000 to 150,000 back to the polling booth is realistic. The BC Conservatives could collect 200,000 votes in the next election, about 11%, and not take a single voter away from the BC Liberals. The BC Conservatives need to be polling over 15% to be pulling any new support from the BC Liberals.
The addition of an extra 150,000 BC Conservative voters in the next election and no change in NDP or Liberal support means their share of the vote falls. For the Liberals the drop is to 41.7% and the NDP to 38.4%. What this means is that the NDP can win a majority was as little as 39% and for the BC Liberals with 41%.
When I look at all this information and crunch these numbers, the BC Liberal attack campaign on John Cummins does not strike me that useful for the party. It only makes sense if the Conservatives are polling around 20% and 2009 BC Liberal voters are being lost to the BC Conservatives. I have not seen any BC provincial polls in a long time so I have no idea what the trends are, but I know the NDP and Liberals have been doing polling.
Friday, June 17, 2011
Frank Graves of Ekos analyses what happened
Frank Graves of Ekos close enough to the 2008 election to not have worried about the gap between the result of the Conservatives and where he had them. In this election that error, and that of other parties, was large enough that he has taken a look at what is going on.
I think all the pollsters should take a hard look at what is going and try to understand why none of them came particularly close to the election result.
He has looked at what is going on and why his forecast of the election was wrong. He highlights three reasons that may have caused the error
Frank goes through and shows his sampling methodology is accurate.
He also shows that there is no evidence in the data before or after the election to indicate a large change in the voters on election day.
In the end what he finds is that the people that vote are not the same as the people that do not vote. The support among the voters is higher for the Conservatives than the non-voters. He manages to figure this out using the data that only Ekos was tracking - how likely people were to actually vote. When he applies those numbers his results come close to the national result.
I used his numbers to make my final estimate of party support on May 1st and I was closer than any pollster was in the nation. I still pulled back from going has high on the Conservatives that my data showed or as low for the Liberals, it simply was too far off from what everyone else was saying.
In the last ten days of the election Frank and I were messaging back and forth on Twitter about polling methodology issues, I raised issues about the small sample sizes in BC and the issue of who will actually be voting versus the polling numbers.
In closing, I highly recommend all of you polling number crunchers read the article.
I think all the pollsters should take a hard look at what is going and try to understand why none of them came particularly close to the election result.
He has looked at what is going on and why his forecast of the election was wrong. He highlights three reasons that may have caused the error
- Sampling error on the part of Ekos
- There was a lot of movement of voters on election day
- The people that vote and do not vote are not statistically the same.
Frank goes through and shows his sampling methodology is accurate.
He also shows that there is no evidence in the data before or after the election to indicate a large change in the voters on election day.
In the end what he finds is that the people that vote are not the same as the people that do not vote. The support among the voters is higher for the Conservatives than the non-voters. He manages to figure this out using the data that only Ekos was tracking - how likely people were to actually vote. When he applies those numbers his results come close to the national result.
I used his numbers to make my final estimate of party support on May 1st and I was closer than any pollster was in the nation. I still pulled back from going has high on the Conservatives that my data showed or as low for the Liberals, it simply was too far off from what everyone else was saying.
In the last ten days of the election Frank and I were messaging back and forth on Twitter about polling methodology issues, I raised issues about the small sample sizes in BC and the issue of who will actually be voting versus the polling numbers.
In closing, I highly recommend all of you polling number crunchers read the article.
Thursday, April 28, 2011
A Reality Check on the NDP
The NDP is going to win a lot more seats on Monday than anyone though was possible before this election. The NDP is going to elect people that were only there as paper candidates. The NDP was far from a serious national party before this election. Here is one measure of far they were from a serious threat in the last five elections.
Rebates are the number of candidates that managed to get more than 10% of the vote and qualify for rebate of some of their expenses. >75% is the number of candidates that spent more 75% of the spending limit, >50% is the same for 50%
Election Rebates >75% >50%
2008 243 35 66
2006 214 34 61
2004 195 23 54
2000 57 28 43
1997 87 43 75
What you can see here is the huge boost in the number of New Democrats that break 10% after Jack Layton but the number of candidates that are spending serious money to get elected is still limited and in fact not much up from the bleak years.
The spending numbers show me how many serious campaigns a party is running. If you do not even spend half of what you are allowed, you do not really expect to win or even do well.
This data says to me that the NDP only thinks that there were about 60-70 ridings in which they had any serious chance. In fact the number of Bloc candidates spending more than 75% or more than 50% is not much smaller than the NDP.
In 2008 and 2006 only the Conservatives had more than 155 candidates spend more than 75% of the limit. In 2004, 2000 and 1997 only the Liberals had achieved the same. You will see that the party won each election was the only one to have more than half their candidates spend more than 75% of the limit.
Looking at the NDP in Quebec
Election Rebates >75% >50%
2008 50 3 5
2006 8 0 0
2004 5 0 0
2000 0 0 0
1997 0 0 0
You can see that 2008 is the first election in which the NDP had any presence in Quebec and then only a handful of serious campaigns. I suspect the NDP went into this election only expecting to run the same number of serious campaigns in Quebec but are now stuck scrambling trying to figure out how to suddenly run many more campaigns. You really need a three month lead time to have any decently well organized campaign, two weeks is almost not worth the effort.
One impact of this data is that the NDP is only set up to get the vote out in about 70 ridings in the whole country. The Bloc will have E-Day campaigns in 50 or ridings in Quebec. The Conservatives will have 250 or so E-Day campaigns nationwide and 30 of them in Quebec. The Conservative E-Day campaigns go far enough that I was asked to work on one in my riding in which the Conservative candidate has no real hope at all, it would only go Harper in a 1984 Mulroney style landslide, which is the last time the right won Victoria.
What these numbers mean is that the NDP will have lower than polled results in 230 to 240 ridings in Canada because there is no push to get the people out to vote for the candidate. This will likely cost them roughly 1.5% on election day nationwide and potentially 3% in most of Quebec.
Meanwhile the Bloc should be able to boost their vote by about 1.5% province wide and 2% in Bloc ridings.
The combined differential in Bloc ridings of less than expected NDP vote and better Bloc turnout moves the gap between the two parties 5% towards the Bloc. This should be enough for them the win a number of seats that the various projections are giving to the NDP.
-----------------------------------------
All my data comes from PunditsGuide - check it out
Rebates are the number of candidates that managed to get more than 10% of the vote and qualify for rebate of some of their expenses. >75% is the number of candidates that spent more 75% of the spending limit, >50% is the same for 50%
Election Rebates >75% >50%
2008 243 35 66
2006 214 34 61
2004 195 23 54
2000 57 28 43
1997 87 43 75
What you can see here is the huge boost in the number of New Democrats that break 10% after Jack Layton but the number of candidates that are spending serious money to get elected is still limited and in fact not much up from the bleak years.
The spending numbers show me how many serious campaigns a party is running. If you do not even spend half of what you are allowed, you do not really expect to win or even do well.
This data says to me that the NDP only thinks that there were about 60-70 ridings in which they had any serious chance. In fact the number of Bloc candidates spending more than 75% or more than 50% is not much smaller than the NDP.
In 2008 and 2006 only the Conservatives had more than 155 candidates spend more than 75% of the limit. In 2004, 2000 and 1997 only the Liberals had achieved the same. You will see that the party won each election was the only one to have more than half their candidates spend more than 75% of the limit.
Looking at the NDP in Quebec
Election Rebates >75% >50%
2008 50 3 5
2006 8 0 0
2004 5 0 0
2000 0 0 0
1997 0 0 0
You can see that 2008 is the first election in which the NDP had any presence in Quebec and then only a handful of serious campaigns. I suspect the NDP went into this election only expecting to run the same number of serious campaigns in Quebec but are now stuck scrambling trying to figure out how to suddenly run many more campaigns. You really need a three month lead time to have any decently well organized campaign, two weeks is almost not worth the effort.
One impact of this data is that the NDP is only set up to get the vote out in about 70 ridings in the whole country. The Bloc will have E-Day campaigns in 50 or ridings in Quebec. The Conservatives will have 250 or so E-Day campaigns nationwide and 30 of them in Quebec. The Conservative E-Day campaigns go far enough that I was asked to work on one in my riding in which the Conservative candidate has no real hope at all, it would only go Harper in a 1984 Mulroney style landslide, which is the last time the right won Victoria.
What these numbers mean is that the NDP will have lower than polled results in 230 to 240 ridings in Canada because there is no push to get the people out to vote for the candidate. This will likely cost them roughly 1.5% on election day nationwide and potentially 3% in most of Quebec.
Meanwhile the Bloc should be able to boost their vote by about 1.5% province wide and 2% in Bloc ridings.
The combined differential in Bloc ridings of less than expected NDP vote and better Bloc turnout moves the gap between the two parties 5% towards the Bloc. This should be enough for them the win a number of seats that the various projections are giving to the NDP.
-----------------------------------------
All my data comes from PunditsGuide - check it out
Some thoughts on the polling in BC in this election
I am a political junkie and even worse a poll junkie. It is truly like a drug and polls being the crack of political junkies. I try not to pay attention to the polls, but I can not help myself. Anyway, I am back with some thoughts on the polls in BC.
I talked with Frank Graves of Ekos on Twitter and pointed out their sample sizes from BC were small and often smaller than those from Alberta even BC has many more people. I pointed out that BC has a lot of seats in play and Alberta has maybe two in play. He took my point and they have increased their sampling in BC.
The latest Ekos poll has a sample size of 322 in BC. This is means a margin of error of +-5.5%
The biggest difference between the two polling trends are with the Conservatives and the Greens.
Nanos is in a third trend, they have very, very small BC samples and they move wildly from day to day, but broadly the Conservatives are over 40% with the Liberals and NDP close to tied at near 25%, the Greens are very low in BC according to them.
When I take account of some of the new questions that Ekos is asking about depth of support, the shift in the numbers gives me something like this:
Keep in mind this is a very rough parsing of the numbers that has no real statistical legitimacy, it is simply an attempt to quantify how the depth of support data may impact party support levels.
Looking over the history of federal elections in BC, any party that has managed to win more than 40% of the vote has won the majority of the seats. Several times, 2004, 1993, and 1988, the majority of seats were won by a party with around 37% of the vote. In 2004, with 36.3% of the vote, the Conservatives won as many seats as in 2008 when they had 44.4% of the vote.
In the last 13 elections only twice, 2006 and 1972, did no party win a majority of the seats in BC.
So what the polling and electoral history is saying to me is that the current prediction I have for BC, Conservatives 22, NDP 12 and Liberals 2, is broadly correct.
I talked with Frank Graves of Ekos on Twitter and pointed out their sample sizes from BC were small and often smaller than those from Alberta even BC has many more people. I pointed out that BC has a lot of seats in play and Alberta has maybe two in play. He took my point and they have increased their sampling in BC.
The latest Ekos poll has a sample size of 322 in BC. This is means a margin of error of +-5.5%
- Conservatives 35.7%
- NDP 28.6%
- Liberals 19.5%
- Greens 13.4%
The NDP and Greens are both up, the Liberals are about even and the Conservatives are down a fair bit. Broadly these numbers were the same in the recent Forum poll, but I do not know the exact size of their BC sample.
Meanwhile Angus Reid has some rather different results. A week ago they did the poll with the biggest sample of anyone in BC in this election. The results were broadly similar in their very small sample from their national poll this week.
- Conservatives 42%
- NDP 32%
- Liberal 18%
- Green 6.0%
The biggest difference between the two polling trends are with the Conservatives and the Greens.
Nanos is in a third trend, they have very, very small BC samples and they move wildly from day to day, but broadly the Conservatives are over 40% with the Liberals and NDP close to tied at near 25%, the Greens are very low in BC according to them.
When I take account of some of the new questions that Ekos is asking about depth of support, the shift in the numbers gives me something like this:
- Conservatives 41%
- NDP 30%
- Liberals 19%
- Greens 9%
Keep in mind this is a very rough parsing of the numbers that has no real statistical legitimacy, it is simply an attempt to quantify how the depth of support data may impact party support levels.
Looking over the history of federal elections in BC, any party that has managed to win more than 40% of the vote has won the majority of the seats. Several times, 2004, 1993, and 1988, the majority of seats were won by a party with around 37% of the vote. In 2004, with 36.3% of the vote, the Conservatives won as many seats as in 2008 when they had 44.4% of the vote.
In the last 13 elections only twice, 2006 and 1972, did no party win a majority of the seats in BC.
So what the polling and electoral history is saying to me is that the current prediction I have for BC, Conservatives 22, NDP 12 and Liberals 2, is broadly correct.
Wednesday, April 27, 2011
The polls lately
There have been a number of polls out today and yesterday and they all show the Liberals now in third place with the NDP comfortably in second place. The gap seems to be about five percentage points between the two.
More interesting is how close the NDP is getting the Conservatives. The Forum poll has then within three percentage points. I think that result will ultimately benefit Stephen Harper because it will energize wavering Conservative voters to go to the polls.
I remain confident that my prediction seats won by each party in this election will be close to the mark. Based on the latest polls, here is where I think the parties will finish in election day:
I think there will be a drop in election from the polling numbers for the NDP because they do not have the machinery in place to get their new vote out. I also think their new support is weaker and more likely to not bother going to the polls.
I see some 'shy tory factor' going on. I am surprised at the number of people who quietly are saying they will support the Conservatives but do not want to say to publicly. Canadians like Harper as much as Brits liked Thatcher. I think the Conservatives will out preform their current numbers by a bit.
More interesting is how close the NDP is getting the Conservatives. The Forum poll has then within three percentage points. I think that result will ultimately benefit Stephen Harper because it will energize wavering Conservative voters to go to the polls.
I remain confident that my prediction seats won by each party in this election will be close to the mark. Based on the latest polls, here is where I think the parties will finish in election day:
- Conservatives 38%
- NDP 27%
- Liberals 23%
- Bloc 7%
- Greens 5%
I think there will be a drop in election from the polling numbers for the NDP because they do not have the machinery in place to get their new vote out. I also think their new support is weaker and more likely to not bother going to the polls.
I see some 'shy tory factor' going on. I am surprised at the number of people who quietly are saying they will support the Conservatives but do not want to say to publicly. Canadians like Harper as much as Brits liked Thatcher. I think the Conservatives will out preform their current numbers by a bit.
My prediction province by province
I am know the provincial seat totals will not be as accurate as the overall national numbers, the closest races across the nation tend to balance themselves out. I meant to get these numbers out yesterday, but life intervened.
Prov Cons Libs NDP Bloc Ind
BC 22 2 12
AB 27 0 1
SK 11 1 2
MN 9 1 4
ON 59 30 16 1
PQ 8 4 31 31 1
NB 7 2 1
NS 4 5 2
NF 2 4 1
PE 2 2
Nth 1 1 1
If I could do 1/2 seats, I would have some on the list, especially it would 1/2 an independent in Alberta and 1/2 an independent in Ontario.
Prov Cons Libs NDP Bloc Ind
BC 22 2 12
AB 27 0 1
SK 11 1 2
MN 9 1 4
ON 59 30 16 1
PQ 8 4 31 31 1
NB 7 2 1
NS 4 5 2
NF 2 4 1
PE 2 2
Nth 1 1 1
If I could do 1/2 seats, I would have some on the list, especially it would 1/2 an independent in Alberta and 1/2 an independent in Ontario.
Subscribe to:
Posts (Atom)