New Nanos poll: NDP voters switching to the Liberals?

We didn't have a lot of polls so far during this campaign, but it seems things are about to change with Nanos rolling their daily updates. For those of you who don't know, Nanos surveys 400 persons every evening and add them to its 3-days rolling average, dropping the 400 oldest interviews. This is a nice way to provide daily updates without ocurring the cost of a full size, 1200 respondents poll every day. What it also means is that numbers in those daily updates move slower, since by definition it is an average and 2/3 of it doesn't change over one day.

This is why today's update is surprising. We see the Conservatives still very high, at 39.1%. But the surprise is to see the Liberals jumping 4 points, at 32.7%. On the other hand, the NDP's drop is the exact opposite, minus 4 points to 15.9%. To see such big changes, it means that the support for the NDP among the newly added respondent is really low. Or that it was a lot higher in the dropped off ones, or a combinantion of both of course. At the end, this is a bad trend for the NDP. Seems like the candidate in Elgin-Middlesex-London wasn't alone after all lol

Using this single poll, we get the displayed projections. Here is the pdf with the detailed projections. By the way, Nanos is merging Alberta and the Prairies together. In order to use this poll, I had to do some math in order to recover the vote intentions. Specifically, during the past 3 elections, the Conservatives' votes in Alberta was 1.13 times the percentages in the Alberta+Prairies. Doing the same calculations for all parties, I can then use this poll in the model. It's not perfect but it provides a good enough approximation.

You may be surprised to see that the Tories would still get as many as 147 seats despite the Liberals' surge, but if you look at the poll, you see an increased lead in Ontario, so that explains a lot of things. Of course, as always when using only one poll, be carefull not to be carried away. In particular, if you have to participate in an election pool, I still suggest to use the projections based on the average of all polls (which have been updated today by the way).

At the end, it's partially good news for the Liberals, even though they might want to still votes from the Conservatives and not only from the NDP. After all, a big part of getting a majority is the lead the first party ahs over the second. In this poll, this lead is still very comfortable for the Tories.

Atlantic Canada: the source of the majority for the Conservatives?

Atlantic Canada: the source of the majority for the Conservatives?
Everyone pretty much assumes that the Conservatives have to make big gains in Ontario if they want their majority. Indeed, Quebec seems like a dead end for this party and Harper will likely settle for 9-11 MPs there. As for the Prairies, Alberta and BC, the Tories have pretty much maxed out over there. BC has a couple of seats around Vancouver where Harper and its party wish to make some gains, but that's all.

One region I personally think we don't give enough attention is the Atlantic. I'll try to explain here why I think the Conservatives can actually make some important gains over there. Recent polls consistently show that the CPC is higher in this region than it has ever been since 2004 and, more importantly, usually enjoy a decent lead over the Liberals.

This region has traditionally been very liberal. Even Stéphane Dion finished first over there in 2008, with 17 seats, while the Tories got 9 and the NDP 4. But of course, the key for this analysis is to determine if the ABC (Anything But Conservatives) effect of Danny Williams is still there in NF-L. The former Premier was very successfull in building a movement against the Conservatives and effectively managed to wipe them out of his province. So before looking at some scenario, depending on how much of the effect is still present, we need to estimate this ABC effect.

I admit that I worked very hard to estimate my model using the data for the Atlantic. It took me many tries before reaching a model that made sense. The real problem is of course that I have to design my model at the regional-level since polls are not provided at the province-level. One issue with that is that the overall swing a party experiences in the region can be the result of many different swings provincially. It was the case in 2008 where the Conservatives dropped from 32.6% in 2006 to 29.6%. But if you exclude NF-L, the Tories actually experienced a positive swing of 2.2%, with a particularly big increase in NB. Therefore, the huge drop in NFL where the Tories collapsed from 42.7% to 16.6% skewed the regional swing. On the other hand, the boost the Liberals got in NF-L allowed them to partially hide a lackluster performance in the Atlantic.

After some fancy tricks, I manged to estimate the coefficients to translate the regional swing AND estimate the NF-L specific effects. Namely, I found that the ABC effect was -32.7-points for the Tories, + 16.5-points for the Liberals and +8.3-points for the NDP (and virtually 0 for the Gren). If you are surprised to see the boost for the NDP to be so low, you need to remember that I also included a riding-specific effect for St-John-East where the NDP got a giant boost of like 46-points. So this effect alone accounted for a lot of the provincial swing of the NDP.

So whether those effects are still there is crucial in analyzing the current regional swing. Let's look at some scenarios to see that more in details.

1. The effects are still 100% there.
If this is the case, then it means the regional swing is really big. The current projections have the Tories standing at 39.2% in the Atlantic. That would mean a swing of 10-points, applied in all four provinces. With this swing and the effects still there, the current projections would give the Conservatives as many as 21 seats. Don't forget the Liberals are currently projected around 34%, thus slightly below 2008.

2. The effects are completely gone.
If this is true, then it means a large part of the observed regional swing is coming only from the increase in NF-L. Indeed, by jumping 32-points in NF-L and staying stable everywhere else, the Conservatives would now stand at 35% in the Atlantic. Therefore, the provincial swing would be only around 4-points, and not 10-points. With this scenario, the Tories would win 15 seats. Some might be surprised that the party of Stephen Harper is better off having the ABC still there, but it makes sense if you think about it. With the effect still there, it means the CPC is enjoying a big swing everywhere. Thus, big potential gains in NS, IPE and NB.

3. The effects are halfway gone.
This is the scenario I'm currently using. I see it as a compromise. I understand the Conservatives are saying the ABC is gone, but it doesn't mean it's true as some have commented in the past on this site. With this assumption, the CPC is projected to win 18 seats, right in the middle of the two other scenarios. I really like this compromis since it allows me to spread the observed increase of the Conservatives equally between the ABC effect being gone and a regional swing.

In any case, as I said before, Harper can still expect to make as many as 9 gains in the Atlantic. That alone would put the Tories almost in majority territories. Do not think this is negligeable. Yes Ontario is the real battleground, but we might well know whether Harper got his majority after seeing the results of the Atlantic alone. This is also a key region for Michael Ignatieff. It seems he can simply take the Atlantic as granted and a sure source of seats.

The useless move

The useless move
The ex-NDP candidate in Elgin-Middlesex-London really didn't help his party. He decided to pull out of the race because he's affraid of a Conservatives majority and would thus prefer not having the NDP/LPC vote split. He backed up the Liberals.

It's kind of funny because this is exactly the kind of pre-election coalition scenario that I presented before. Of course, the difference being that there isn't an actual coalition and this move only put some heat back on the NDP and especially the LPC. It reopened the door for Harper to attack his opponents on this issue.

But more importantly, was the NDP candidate even right in doing that? I mean, could he actually prevent one CPC seat by giving his support to the Liberals? The current projections for this riding are:

CPC: 52%
LPC: 23%
NDP: 19%
Green: 6%

So even by adding all the votes from the NDP to the Liberals, we would still have a Conservatives win. If you ask me, this NDP candidate really made a useless move (and I'm not even talking about the fact that the party will probably replace him). He reopened the debate about the coalition, something the NDP and Liberals would rather have behind them at the moment, and he did that in a riding where his move can't change the outcome, even in the best scenario.

If any other NDP candidate is considering doing that, please read before ;-)

New domain name

New domain name
I've decided to make this website look a little bit more professional by buying an actual domain name, instead of using the This one will naturally still works, but from now on, the official address of this website is:

Thanks for sharing the good news ;-)

Can the Conservatives lose one more seat in Alberta?

Can the Conservatives lose one more seat in Alberta?
The Conservatives are looking for a majority obviously, but they are also looking to sweep Alberta, like in 2006. We talk a lot about the currently only non-CPC riding: Edmonton-Strathcona, hold by the NDP. This party won this riding by a slim 463 votes in 2008. How did this happen?

Well, beyond the fact that obviously, the electoral machine of the NDP worked a little bit harder than the Conservatives one, we can also look at the provincial swings. Between 2004 to 2008, the NDP increased from 9.5%, to 11.7% and finally a "high" of 12.7% in 2008. During the same period, the CPC moved from 61.7% to 65% to slighty decrease to 64.7% in 2008. As for the Liberals, they went from a decent 22% in 2004, to a low of 15% in 06 and a really low 11.4% in 2004. So all in all, it seems the small swing of the NDP and the Tories were enough to give this riding to the party of Jack Layton. This is actually a good example of how the model can work: it can identifies where the provincial swing will be amplified or decreased.

So what about the current situation? Well, believe it or not, but the recent polls show a drop of the Tories, to a new "low" of 57.5%. Don't laugh, I know any party would do anything to stand that high in a province, but the fact remains: the current provincial swing is negative for the Tories. On the other hand, the NDP is stable at 12.7% and the Liberals seem to have found a new life, currently standing at 18.9%. We also have some polls that place the Green 2nd or 3rd in this province, but I won't talk about that right now.

So with those swings, the NDP would naturally keep its riding. It would be a close race again of course. But I think we are wrong to focus only on this riding. We should also watch Edmonton-Centre. This one was actually Liberal until 2006. And with the current swing for this party and the slight drop of the Conservatives, we could well have another riding to watch in Alberta. The current projection place the CPC at 41.4% and the Liberals at 36.1%. So almost a close race!

Also, there could be another race in Edmonton-Sherwood-Park, where the former Conservative candidate is running again, as independent. He did that in 2008 already and almost beat the official CPC candidate. This effect was so strong that I had to include a riding-specific effect in my model to not bias my other coefficients! My guess is that the Tories will win this riding again though, with an even larger majority. The model currently assumes no independent candidate, soI'll have to adjust for that.

Ok, here you go, I did my best to find some interesting ridings in Alberta. It's a tough job, believe me! I welcome your feedback. If you want more details about the ridings in Edmonton, this text from Daveberta is quite good.

Can Larry Smith win his riding?

Can Larry Smith win his riding?
It seems to be a question a lot of people are asking, whether on this site or elsewhere. The Conservatives were very happy to recruit this high profile candidate in order to get more votes in Montreal West, a region that is Liberal to death. But can Larry Smith really win the riding of Lac-Saint-Louis? Let's look into it.

First of all, here are the results of the 2008 elections.

CPC: 23.5%
LPC: 46.4%
NDP: 15.8%
Green: 8.6%
Bloc: 5.8%

By the way, you won't find a lot of ridings in Quebec where the Bloc is 5th and last! So this riding seems like a safe liberal seat. But let's look at the current projections. Since the last election, the polls indicate some important swings. In particular, the CPC had decreased from 21.7% provincially to 20.6%. The drop for the Liberals is bigger, moving from 23.8% to 18.7%. As for the other parties, the NDP is up, Green as well and the Bloc slightly up. But it's not really relevant for this riding since the game is played between the Liberals and the Tories.

So with the current votes intentions, we have the following projections:

CPC: 23%
LPC: 42.7%
NDP: 17.9%
Green: 9.3%
Bloc: 7.1%

It requires a little bit more explanations. My model is there to translate the provincial swing into riding-level swings, taking into account the region, incumbency effect and other variables. So, maybe a better way to see if the Conservatives can indeed win this seat is to look at what happens when the CPC and LPC experience a 1-point swing provincially. Here how a 1-point swing translates in this riding.

CPC: 0.18
LPC: 0.62

So in both cases, the swing is decreased in this riding. It means the Liberals can lose some votes provincially and still be safe there. For the CPC' it's a bad news as it indicates that, if past results are of any indications, even a big provincial swing wouldn't help. So what chances are left? Three solutions:

1) The historical trend about the link between the provincial swing and the swing in this riding doesn't apply any more. I don't like this solution cause it goes completely against the model.

2) A Within-province swing, i.e: even though the Conservatives remain constant around 20%, there is a shift within Quebec. In other words, this party loses some voters in some ridings but is gaining voters in others. We couldn't see such a trend with province-level polls of course. This explanation is possible, in particular we could imagine that Harper was successfull if getting more votes from the anglophone community in Quebec. But to what extent? I don't know. In 2008, in the region of Montreal West, the Tories got 18.18% (thus below the provincial average) and the Liberals got 45.3%. Thus you need a big within-province shift.

3) A star-candidate effect. Larry Smith is well known and he will likely beneficiate from some kind of boost, independently of what happens to his party province-wide. However, the boost would need to be pretty big for him to win. We are talking a boost of 10-points, taken from the Liberals. For a non-cabinet minister, non-leader of a party, that is pretty big.

So at the end, the short answer to the question of this post is no. If he wins, that would be a big surprise. My guess is that Harper is more playing long-term in Montreal West: he wants to be the choice of the English-speaking community in Quebec. While it will likely not be the case yet this time, this strategy might pay off in a couple of years.

New Abacus poll: finally some good news for the Liberals

New Abacus poll: finally some good news for the Liberals
The Liberals are having a somewhat good start of the campaign, with the 1000-persons rally yesterday in Montreal. Of course, you could argue that the Conservatives are currently winning because they managed to focus everyone on the word coalition instead of talking about the wording of the non-confidence motion. But in any ways, polls have been harsh for the Liberals. On the other hand, the NDP is consistently projected high (around 19-20%). This constitutes a big change as compared to before the election where this party woud stand around 14-15%.

Yet today a little ray of hope with the release of this new Abacus poll. There is also a new HD poll, but we don't have the regional breakdown yet, so I won't use it for my daily-projections (but the HD poll is much more in line with the other recent polls though). Oh, and I saw this poll from Forum Research. I don't really remember this compagny. This one is actually saying a very different story than Abacus, with the Conservatives being at 41%! But since there isn't a regional breakdown provided (well, there is, online, if you subscribe and pay, so no thanks), I can't use it. It's too bad, I could have used the two polls to illustrate how to introduce some uncertainty in the model, something I'll talk about again very soon. If anyone has access to the regional details of the Forum poll, please let me know.

So back to the Abacus poll. That would give the projections displayed here (sorry no fancy picture for single-poll projections).

CPC: 128
LPC: 89
NDP: 42
Green: 0
Bloc: 49

Yes, the Tories would still win, but would go back to a smaller minority. I know, I know, I said good news for the Liberals. But at that stage of the campaign, limiting Harper to 128 seats would already be an accomplishment for Ignatieff and his party. I'm not saying he'll not be able to increase his level of support substantially during the next 34 days, but for now, let's be realistic and use what we currently know.

But before any liberal fan gets high expectations, remember this is only one poll, going against the general trend and with a sample size of 1000 only. It means that some of the weird results in some provinces (the NDP at 33% in the Atlantic, while the CPC stands at only 24% there; the high Prairies score for the Liberals; etc) are probably due to a ridiculously small sample size.

Not that I want to talk about a coalition anymore (as a former sutdent in political science, I find it fascinating, but as a voter, I'm getting tired of it), but for the first time in a long time, the number of seats for LPC+NDP is actually higher than the CPC. Just saying...

Update on the motivations of May to run in BC

Update on the motivations of May to run in BC
When I estimated the "May effect", i.e: the boost that the leader of the Green party can expect when running in a given riding, I said that I didn't understand why she choose the riding of Saanich-Gulf-Island in BC instead of Guelph in Ontario.

Well when reading this article (in french), it seems that May and the Green party did do some research before choosing. She says that (my translation) "The party did a lot of research, in collaboration with two poll firms. That allowed us to determine that, out of all the ridings in the country, this one was where the voters were the closest to the values of the Green party".

Is that true? If this is the case, then how come the Green did better in Guelph than in this riding? I understand that the Green party gets its best score in BC, but the fact remains that Guelph seemed better, with potentially more votes to take from the NDP.

But we'll see. I modified the model to take into account of this May effect and so far, the projections don't agree with ehr when she says "the race is between the Conservatives and us. The Liberals and NDP are far behind". Yes the Green party seems to be in the race, but the Liberals are still projected 2nd, despite losing some votes to the May.

I hope we'll get a poll specifically for this riding throughout this campaign so that we could compare and maybe adjust the projections. But in the mean time, I'm still not sold on her choice of riding.

Day 1 (March 26th): Conservatives start with a majority

The election is officially started and we've got a couple of new polls (AR, Léger, Ipsos). They all show a big lead for the Conservatives and all the medias are talking about the Tories being "on the edge of a majority". According to the latest projections, the Tories are indeed in majority territory. So if you are a Liberal, you better hope that campaigns DO matter.

Some of you might say that 155 seats is a very narrow majority and you'd right. But if you look in details, you'll see that the Conservatives are projected to win 143 safe seats (i.e: where their margin of victory is more than 5-points). And then, they have a short lead or are within 5-points of the leaders in 33 other close races. So currently, the model projects the Tories to win 12/33 close races. This is a very low percentages, especially given the record of the CPC in the past election. As I explained, this party is really good at winning close races (probably by getting the vote out). Therefore, I would see the current projections as a Conservatives majority, period. They could well win as many as 176 seats in the perfect world! This party seems to be really high in the Atlantic (the last good region for the Liberals) and in Ontario.

For the Liberals, the best scenario (i.e: winning all close races) would be... 77! Yes, Micheal Ignatieff starts the campaign with having a best scenario equal to what Dion did 2 years ago. I really wonder why the Liberals wanted to bring the government down so bad Of course, we'll see on May 2nd if Ignatieff was right to gamble. But he sure has a steep hill in front of him.

When you look at those projections, you can see that the NDP didn't have so much so lose for voting against the government. There can't expect to make big gains, but they are not really at risk of losing a lot. So that might explain their decision.

So here you go folks, enjoy the elections. It is Harper's majority to lose and it's a new game in (recent) Canadian politics.

Estimating the Elizabeth May effect

Estimating the Elizabeth May effect
Elizabeth May is a puzzle for me. In 2008, she had an unique deal with Stephane Dion in that he agreed not to run any liberal candidate versus her. Yet, where did she decide to run? In the riding of Peter McKay! While she did get a pretty decent score of 32%, it wasn't enough to win and quite honestly, there was no way she would win this riding. In 2006, the Green party only got 1.59% in this riding! So why would she choose to run there? I understand this is where she lives, but still, the Green party is at the point where having an elected MP matters more than that.

So you would think May learned from her mistake and would run in a riding that she can win. But no, instead of running in Ontario where the Green did pretty good in two ridings (Guelph at 21% and Bruce-Grey at 27%), she decided to go to BC and run on Vancouver island in the riding of Saanich-Gulf-Island, a riding where the Green party only got 10.46% of the votes, well behind the Tories and the Liberals.

But can she win this riding? After all, it is clear there is a May effect. But how big is this effect? Let's estimate it using the results of 2008 and the by-election of London North/Centrer where she ran in 2006 and got a surprising 26%.

According to my estimates, in 2008, the Green party was in Central-Nova 29-points above what it should have been! Of course, this is the combine effect of Elizabeth May running AND having no Liberal candidate. Surprisingly, using the same estimation, I find that the NDP was down 15-points in this riding, while the Tories were up 8-points. Again, see those numbers as what would have happened if May wasn't the candidate and the Liberals had a candidate. It's interesting to see that some NDP voters decided to switch in order to give the Green party a chance. It isn't completely surprising given that a lot of NDP voters have the Green as second choice. It's also worth noting the little boost that CPC got. I'm guessing some Liberals voters were upset with the deal and switch to the Tories. As for the Liberals, if they had ran a candidate, I estimate they would have got 21%. So without May and with a liberal candidate, the results would have been:

CPC: 38%
LPC: 21%
NDP: 35%
Green: 3%

Instead of respectively 47%, 0%, 20% and 32%. Note how the maths sum up: the 21% of the Liberals were split into +8 for the CPC and +13 for the Green. Therefore, May got 13+15=28 points more than if it was a regular Green candidate.

So what about the May effect in the by-election in 2006? In the 2006 general election, the Green party got 5.49%. Then, during the by-election, May got 26%! So, at the end, it seems Elizabeth May does benefit of a huge boost in term of votes whenever she runs in a riding. Quantitatively, I would say she can get a jump as high as 20-points. Mostly at the expense of the NDP. Indeed, the Liberals were barely affected in the by-election. In both cases, I would say the May effect was: 5-8-points taken from the Liberals, 10-15-points taken from the NDP. (yes she got around 13-points from the Liberals in 2008, but there was no candidate. It's not the same as having liberal voters deciding to switch simply because it was Elizabeth May).

So let's be generous and give May a 20-points boost, taken at 1/3 from the Liberals and 2/3 from the NDP. Does it allow her to win her seat in BC? Using the latest Ekos poll who puts the Green as high as 19.7% in BC, we have the following projections for Saanich-Gulf-Island:

CPC: 36%
LPC: 42%
NDP: 0.1%
Green: 21.5%

Let's add the May effect. One problem is that this riding has already a low percentage for the NDP, so not so much votes to take from them! At the end, May would need to steal as much as 11-points from the Liberals. It's more than what we have observed in the past. And remember, I'm using a poll very favourable to the Green!

So back to my question: why running in a riding where the Gren party was low in 2008 and where the NDP is low as well? That doesn't make any sense. On the other hand, the current projections place the Green at 36% in Guelph and 28% in Bruce-Grey. In both case, if you add the May effect, you would get a win for her and her party (especially in Guelph). So why did she decide not to run in one of those ridings? I don't know. Let me know if you have a theory.

I'll modify the model to take into account of this May effect of course.

Can we predict individual ridings using the provincial swing?

Can we predict individual ridings using the provincial swing?
Ok so here we are, in election! I know most people are not happy about that, but I'm sure that most of my readers are, as any political junkie. Today, I'll not post new projections based on one or two polls. Instead, I would like to justify a little bit my model and why I think it is valid. Of course, you can get more details if you read the methodology under the tab FAQ.

The point I want to address today is the following: can we really predict riding-level swings using the provincial swings only? After all, I'm sure some of you are skeptical. Simply because one party increases by 4 points in a province doesn't mean this party will increase by 4-points in every ridings. The 4-points provincial swing is an average, and an average could hide a lot of variation. For instance, it is technically possible that this 4-points increase came from an increase of 12-points in 50% of the ridings, and a decrease of -4-points in the other 50%.

I showed in this post what the swing looks like. It isn't perfectly uniform, nor is it perfectly proportional. Actually there is a lot of variations. Nevertheless, it turns out that the provincial swing is still a good predictor, especially if we interact it with other variables. But before showing you numbers, let's see what kind of information we have about each riding. I'm talking here about information publicly available during an election, so the GDP per capita is not one of them.

-The past results of every party.
-The current incumbent.
-The province and region where the riding is located (ex: Montreal West or Quebec city).

So by using the provincial swing and these other variables, we can get a lot more predictive power. For instance, it is obvious that when the Conservatives increase in the province of Quebec, they increase much more in the region of Quebec city than on the island of Montreal. We know that because we can observe the past elections' swings. We can also look if a party did better or worse in the ridings it won the last election (an incumbency effect). This is really where the use of statistical tools such as a regression (or econometrics in general) helps a lot. Other websites can't take into account so many variables, not at once. I can even differentiate the swing in function of the level of support of the other parties (for instance, when the Tories go up in Ontario, do they take votes from the Liberals or the NDP?).

At the end, the predictive power varies across provinces and political parties. It works really well for Quebec and Ontario, for the two main parties. For instance, I can explain as much as 89% of the riding-level swing for the CPC in Quebec! This number is 86% for the Liberals. The remaining 12-15% could be inputed to riding-specific events, or on an efficient organization (to get the vote out).

On the other hand, it works usually less well for the NDP. The reason being that this party has experienced less variaiton over the last couple of elections (in most provinces, the NDP stayed really flat), so I have less variation to identify (pin down if you prefer) the coefficients. Nevertheless, I can explain around 76% of the swing for the NDP in Quebec, but only 22% in Ontario. When this number is that low, I actually switch to a simple linear-uniform model for this party. This is unfortunate but there isn't much that can be done. And anyway, as shown in my methodology, the linear-uniform model already provides a good approximation.

So, here you go. It seems that we can indeed project riding-level swings using only the provincial ones as the source of new information. In average, we could explain over 70% of the swing. But it works better in big provinces and for political parties that have experience changes over the year. I would say the best scenario for the model would be to have the CPC down a little bit (going closer to its 2006 results) and the Liberals up a bit. On the other hand, if the Tories get closer to 40% and the Liberals keep falling, we could face a problem of extrapolation. Indeed, since we've never observed the Liberals so low, we don't really know what would happen. But as long as the changes are not too big (and let's face it, they usually aren't), it works just fine. In 2006, given the correct percentages, the model would have predicted correctly 286 ridings. In 2008, it was 283. And most of the mistakes were in close races (where the margin of victory was less than 5-points). Better than that, in average, the projections were accurate within around 2-points (so for instance, the model predicted the CPC at 46% and the actual result was 44%.

Scenario of a coalition PRE-election

Scenario of a coalition PRE-election
The election isn't officially started yet, but we're already talking about a possible coalition between the LPC and NDP (and, to a lesser extent, the Bloc). This post is not intended to promote such coalition. Nor do I want to argue whether such coalition is legitimate or a good idea. My objective here is simply to show how many seats such a coalition could grab. And I'm well aware that the scenario studied here is highly unlikely to happen in Canada. So don't fall on me for that lol See that as an academic exercise, nothing more.

I'm talking here about a coalition PRE-election, where the NDP and LPC officially announced that they would do one. If such a situation should occur, then I believe the two parties should actually "exchange" candidates in some key ridings. Indeed, it is the case that in some ridings, the CPC can win, but the total sum of votes for LPC and NDP is actually greater. When it is the case AND those two parties want to do a coalition, then they should be logical and leave only one candidate. That would avoid a split of the "coalition votes". But of course, to really forecast what would happen in such circumstances, we need to take into account the second choices of the voters. Indeed, it is obvious that in the event that the NDP removes its candidate, not all voters will follow the order and give their vote to the Liberals. Some will simply not vote, while other will vote for another party, including the Conservatives.

So here are the details of this experiment:

- I'm using the latest projections, based on the average of most recent polls.
- I'm using the second choices as given in one of the latest Ekos poll.
- I'm assuming that the coalition would do such a candidate swap only in the ridings where they can potentially beat the Conservatives. Specifically, it has to be the case that CPC is first, and LPC+NDP is greater than CPC. As you'll see, the number of ridings currently in this situation is not that big.
- The "coalition" would remove the candidate from the party that is currently third. In other words, if the Tories are first, the Liberals second and the NDP third, then the NDP candidate would have to go.
- When one candidate is removed, I'm using the second choices of the voters of this party and redistribute the votes.

This redistribution requires some explanations and assumptions. First of all, I'm well aware that some voters would switch even though their first choice is still technically available. For instance, in our previous example, we could expect some Liberal voters to actually change their votes simply because they don't want a coalition with the NDP. That could happen to some voters that are exactly between the Tories and the Liberals. Here, I'm assuming that this isn't the case. This introduces a bias, but hopefully it will be canceled out by the next assumption. The second choices are biased as well. Indeed, it is one thing to not name the Liberals as your second choices when you are a NDP voters (possibly, you don't have a second choice, which happens in around 30% of the cases), but it's another thing to be told by the NDP to vote LPC. We can imagine that some voters that didn't have the Liberals as second choice would still switch their votes, simply because they understand the whole principle of the coalition and the candidates exchange. I don't model that. So at the end, I have two assumptions who both introduce a bias, but in the opposite direction. In the future, I could try to "play" with those assumption and see what it would change (or even better: maybe we'll have one pollster asking specifically how many people would switch or not, in the event of a coalition).

So, enough about the technicalities, let's see if such a deal could actually benefit the Liberals and NDP. Based on the current projections (they are a little bit different from the ones displayed on the right because I added the latest Ipsos poll), here is what such a deal would change:

Number of ridings where the LPC+NDP>CPC: 34

seats actual coalition
CPC 153 148
LPC 72 74
NDP 32 35
Green 0 0
Bloc 51 51

So, out of the total of 34 ridings where the coalition votes is higher, such a deal would only take 5 seats away from the Conservatives. The main reason? Well it is the retention rate. Specifically, only 32% of NDP voters have the Liberals as second choice (and 33% of LPC have the NDP). Those numbers are not high enough to warrant such a deal pre-election. However, let me say two things.

1) The current projections are very favourable to the Tories. We'll see what happens during the campaign. Such an exercisee would give different results if the Liberals go up and the Tories go down. For instance, if we were to go back to the levels of support of 2006, you would see that such a coalition would be much more effective.

2) As I've said, maybe one of my assumptions is to harsh. Do you really think that only 32% of the NDP voters would switch to the Liberals, while 23% would prefer to vote for the Green and 25% would simply not vote? I have a hard time believing that. Polls usually showed that the NDP voters were quite pro-coalition, back in 2008 (but in the same time, as I've said, some Liberals voters could be scared off as well). So I might adjust those assumptions. In my simulations, it seems that the coalition becomes effective when the retention rate is at 45% or more.

In any case, as I've said, this is simply meant as an exercise. I will repeat it or modify it during the campaign. In the mean time, let me know what you think. By the way, you might think that "stealing" 5 seats away from the Tories is not that big of a deal, but if those 5 seats are the difference between a majority and minority, it could well be important!

(Almost) first day day of the campaign: game over already?

I know we are not officially in election right now, but everybody knows we actually are. So we have our first poll, conducted between March 22nd and 23rd, thus at a time where the government had already fallen or was about to. It is from Ipsos-Reid. If this poll is true, it's game over already! Just remember that you are supposed to apply the margins of error to these results. Still, it's an interesting experiment to use this single poll for projections.

Remember when I posted this? Showing the huge gap that happened between the CPC and LPC right at the beginning of the last election? And how I thought that could well happen again. Well, it's too early to tell, but the Conservatives sure are happy to fall tomorrow.

This poll is just crazy and, let's face it, some of the numbers are more than likely due to a small sample size (and/or at the extreme of the margins of error). But still, I believe they reflect a general trend going on. The Tories have a 19-points lead nationwide. In Ontario, they stand at 46%, 16-points ahead. In Atlantic, they even stand at 54%! (I don't believe this number if you ask me though). They are even up in Quebec (25%) and let's not mention everything west of Ontario.

With this poll only, you get the projections above. Yes it isn't a typo, the Conservatives could win as many as 183 seats. If you look at the detailed projections here, you see that the new seats would mostly come from the Atlantic and Ontario. Again, while I can believe the big lead in Ontario, I trully think the Tories are not at 54% in Atlantic Canada and this is only due to a sample size of less than 200. But even if we use an average of other polls for this region, we would still get the Tories with 168 seats.

The Liberals would not even finish as the official opposition. Again, if we use an average for the Atlantic, this party would get 59 seats. Miacheal Ignatieff would still go back to Harvard if that was the case. You have to admit that it would be a weird situation, with the Conservatives getting a huge majority without any breakthrough in Quebec, and the Bloc as the official opposition!

Time will tell us if this poll was simply way off or not. But as John Ivison from the NP writes, it would be almost impossible for Ignatieff to avoid a big loss on May 2nd. And you should really wonder how on earth did the opposition parties decide to take the government down!

20 ridings to watch according to the National Post: comparisons with the projections

20 ridings to watch according to the National Post: comparisons with the projections
The NP lists 20 ridings where the action will likely take place during this election. This is very interesting but it is based on the results of the last election, and not the CURRENT situation. Using the latest federal proejctions, let's see if those 20 ridings are really that interesting.

Vancouver South: Ujjal Dosanjh won by only 20 votes last time. But thanks to an increased in this province since Ignatieff is the leader, this riding seems safer for the Liberals.

Burnaby-Douglas: Here the NDP would lose this seat won by only 798 votes last time. This is due to the fact that the Tories are pretty stable in BC since 2008, as opposed to the NDP who dropped a couple of points. With that being said, the polls in BC seem to have a high debree of volatitly (most likely due to the small sample size) and sometimes the NDP is pictured as being very high (around 30%), while some other times, this party would finish 3rd, barely ahead of the Green.

Esquimalt-Juan De Fuca: Another seat won by a handful of votes by the Liberals last time. As for Vancouver-South, the margin of victory seems to be safer, but still under 5-points. The question for those ridings is how effective could the Conservatives be in targeting such specific ridings.

Edmonton-Strathcona: The only seat not won by the Tories last election. The projections still give this riding to the NDP, but it's gonna be a close race. If I had to guess today, I would say the NDP will eb able to keep it simply because this party in this province will most liekly invest a lot of money and effort and this simple riding. And the NDP seems to have become quite efficient in that (like Outremont in Quebec).

Saskatoon-Rosetown-Biggar: close win for the CPC ovet the NDP back in 2008. The model currently projects this riding to stay within the CPC, with a slightly increased majority.

Winipeg North: Weird case for this riding. Easily won by the NDP in 2008, this party lost it to the Liberal in a surprising by-election in 2010. Since the model does an average of the last general election's results and the by-election's ones, this riding is projected as being a safe NDP seat again, despite the incumbency effect now working for the Liberals. It's a very tricky riding to predict when you see such a big difference between a general an a by-election. My feeling is that the notoriety of the Liberal candidate will be less of a factor during a general election.

Kitchener-Center: tough loss for the LPC in 2008, to the CPC. Since then, the Tories have increased their leads in Ontario and are thus projected to not only keep the seat, but have an "easy" win.

Brampton-Springdale: The Conservatives lost this riding by 773 in 2008 and are now projecting to win it by a margin of... 0.7%! So yes, it is definitely a riding you want to watch.

Vaughan: Another tricky case since this riding changed colour after a by-election. After years of Liberal domination, the Conservatives won it back in 2010. The current projections give the riding back to the LPC for... 0.1%!

Eglinton-Lawrence: a close win for the LPC in 2008 and a riding currently projected to be even closer, although the Liberals should be able to keep it.

Simcoy-Grey: A very easy win in 2008 for the Tories, but the MP has since then been ejected from the caucus and is running as an independant. The model currently proejcts this riding to remain a safe Conservatives seat, but I haven't made any adjustment to take into account of this major independant candidate. Will she be able to win? Or will she split the votes enough to aloow a third party to win? I'll have to think more about this one.

Outremont: The surprised win by the NDP during a by-election and the even bigger surprise that this party was able to keep it during a general election. With the NDP being so high in Quebec in almost every poll, it seems Mulcair will keep this seat. But it will largely depend on the star-factor of the new liberal candidat, Martin Cauchon.

Lac-Saint-Louis: easy win for the Liberals in 2008. So why is it on this list? Cause the Tories are running a new star: a former CFL player. It still seems like a really, really long shot to win this riding. Especially with the Tories being usually polled low in Quebec.

Moncton-Riverview-Dieppe: A close win for the LPC last time. Given the new strength of the Tories in the Atlantic, this riding is now projected to switch to the Tories in a very close fight. However, if the rumours of Bernard Lord running are true, I think ths riding will then become a safe win for the CPC. [update: it seems Lord has ruled himself out]

Halifax: honestly, I'm not sure why this riding is on this list. The NDP won it easily last time and are doing quite well provincially in NS. I think the NP is wrong on this one.

Egmont: a win by only 55 votes back in 2008, for the Tories. The projections seem to indicate that the race will indeed be very close again, with a slight edge currently given to the Conservatives.

St-John's South-Mount-Pearl: Not much change there, a close win for the Liberals (over the NDP) in 2008, and currently projected as an unsafe LPC seat. Except that the race is now between the Liberals and the Tories. How is that possible? Well, as I explained in another post, I'm assuming that 50% of the effects of the ABC (anything but Conservatives) campaign are gone. Therefore the Liberals are down, the Tories up and the NDP down. But as it is the case for almost the entire province of NF-L, this is tricky to predict.

The three territories: No change according to the model, but given the lack of polls for those ridings, it's hard to tell.

So here you go, the NP is right in that most (if not all) of those ridings are interesting to watch. As opposed to 2008, we currently project 4 changes as compared to the results of the 2008 election. But with so many close races, it's really hard to tell.

Let's the fun begin!

Let's the fun begin!
So it's almost official: we are in election! The only question remaining seems to be whether the 41st federal election will start tomorrow (with a vote on the budget, well an amendment to it) or on Friday (with the vote on the non-confidence motion of the Liberals). I'm pretty sure the Tories would rather be defeated on the budget and avoid the vote of Friday about the ethic of this government. But in any case, by Saturday at the latest, the campaign will be started. [update: it seems there will not be any vote on Thursday after all. So the government will fall on Friday].

So what can you expect from this blog? Well first of all, I've updated the latest federal projections and you can see them on the right. As the election begins, the Conservatives are 5 seats shy of a majority (or even 4, if you choose the speaker of the House from the opposition). Moreover, if you add the safe seats plus the potential ones, the Tories could well have a majority secured.
The Liberals and the NDP are pretty much where they were in 2008, with a couple of seats lost for both parties. Whether we'll see a quick shift in vote intentions like in 2008 or not is unknown, but the fact that we'll start an election with the Conservatives having a virtual majority is completely new. The Bloc could go back to over 50 MPs, despite not increasing in term of percentages. As for the Green party, they will have to work really hard to finally get their first elected MP.

During the campaign, this blog will have daily updates (with some exceptions). We'll likely have 2 to 3 daily polls, so each day I'll use those new polls (and only them) to make some seats projections. As for the "general proejctions" (the more reliable ones, based on an average of all recent polls), I'll update them every three days. There is no point in updating them every day as the average will prevent big changes to occur on a daily basis.

Regulars may have noticed that I added the three territories to the model. Since we usually don't have polls just for those three ridings, I based the model on the overall results, nationwide. It's not perfect but at least I'm now projecting 308 ridings out of 308.

The simulator is there and I invite everyone to use it. Being able to make your own projections should be fun for a lot of you. And as far as I'm concerned, I'm the only website offering this, as both DemocraticSpace and 308 currently don't do that.

Besides the usual projections posts, I will also write some other posts. For instance, I'll show you how many seats a coalition LPC-NDP could get. I'll also show you historical results or additional details about the model (for instance, I'll explain the predictive power of using the provincial swings to make projections at the ridings-level). Or why I think Elizabeth May would have been better to run in Ontario.

If you want to follow all the updates, the easiest thing to do is to subscribe both to the RSS feeds and my Twitter. I also hope to get a lot of comments during this campaign.

So let's start having fun for the next months (and a couple of days more)!

The situation before the budget: latest Nanos poll [Update]

The federal budget will be presented tomorrow and almost evryone expects the government to be taken down by the opposition by the end of the week (whether it would be for the budget or the recent scandals is yet to be seen). So where do the political parties stand just before this budget and likely snap election?

The only poll released this week (so far) is the latest Nanos poll, conducted between March 12th and 15th. If we use only this poll, it gives the projections of this post (note: because Nanos groups Alberta, Sasktachewan and Manitoba together, and my model splits those three provinces in two, I used an average of the other recent polls to make the seat projections in those provinces. In any case, it shouldn't really matter as those three provinces are not the ones where a lot of changes are expected, especially Alberta). The Conservatives would make big gains in Atlantic and Ontario.

So, basically, if an election was to start today, the Conservatives would start this campaign with a virtual majority. Even when using the average of other recent polls, the projections always put the Tories around 150 seats or higher. So I think it's pretty safe to assume that Harper and his party want an election to happen now. Of course, a campaign is long and a lot of things can happen (ask Harper in 2006...), but the fact is that for the first time in a long time, we have a party that has a clear potential to get a majority. Even more impressive is the fact that this majority would be possible without any gains in Quebec. For the Tories, as long as they keep 8-10 seats in Quebec (and most of their seats are pretty safe), they can simply focus on Ontario, Atlantic and BC. For Ontario, polls have consistently shown a big lead for this party, in the range of 13-15 points, over the Liberals. In my opinion, such a big lead in a province with 106 seats would be enough to secure a majority. Then there is Atlantic Canada where the CPC did good last time, except in NFL. If the recent polls are to be trusted, it seems most of the ABC effect is gone and the CPC can expect to make some gains there. Finally, BC has become the source of many seats for this party over the year. I don't think this is a province where the CPC can expect to make a lot of gains, except maybe in the suburbs of Vancouver. But we are taking 2-3 seats, nothing compared to Ontario.

The Liberals are in bad shape. Not only would they start this election far behind the Tories, but there isn't much indication that Michael Ignatieff can expect to make a big come back. The leadership index of Iggy is usually always low, no matter which pollster asked the question. In many cases, Jack Layton is seen as being a better potential PM' which is really bad for the Liberals. Of course, we could have a surprise. Maybe Ignatieff could have a perfect campaign, win the debates and increase his level of supports substantially. But at the end, for the Liberal leader, I don't think getting back to power is really possible. The best bet for the Liberals would be to have a "loss" like the PQ in Quebec in 2008: yes you lose the election, but it's almost a moral win. In other terms, the Liberals should focus on gaining seats back and make sure Harper doesn't get his majority. From then, everything is possible. Harper would probably quit (or be forced to quit) and the Liberals could start getting ready for next time, where they could get the government back. But again, if the polls have to be trusted, it seems the main concern of Iggy will be to remain the official opposition! Looking at the Nanos poll, the Liberals don't even have the Atlantic as a safe source of seats anymore. And by being so far behind in Ontario, this party would be virtually wiped out of the map except in Montreal and Toronto.

The NDP is tricky. Polls have not been really nice to them recently, but this party is usually quite stable in elections. On top of that, their vote has become much more efficient the year, thus theirs seats are mostly safe. The downside of this is naturally that potential gains are difficult. I would say the NDP can expect to increase in Quebec (at least in term of percentages) and in BC. The latter is probably the best potential source of additional seats. Of course, with the Liberals so low and Iggy being impopular as he is, there is a slim chance to see the NDP actually becoming the main opposition to the Tories, at least in some provinces. If such a switch was to happen, the NDP could finally break the 20%-mark and become a real contender. But for that to happen, they need to make sure to remain stable in Ontario and the Prairies.

The Green party hasn't much to hope for. Yes we'll see other polls putting this party as high as 10% nationwide, and even in 2nd place in the Praries or Alberta some times, but at the end, on election day, they will likely get 6-7% max and maybe one seat.

Finally, the Bloc has an easy life as always. But an election could be costly for this party. In term of percentages, we could well see this party falling even more, to around 35%. Fortunately for them, unless one of the federalist parties increases a lot, they are safe in term of seats. With the Tories not focusing on Quebec anymore, the Bloc will try to play the card of "we are the only way to block a Tories majority", but this time, it seems not to be true. And let's not forget the possibility that the Tories will keep some key promises for Quebec for an election, such as the sale tax hamronization.

So here you go folks. We'll know by Friday if we have an election or not. It seems almost unavoidable but we never know. Forcing an election is never easy, but the key difference this time is that Harper likely wants to be defeated.

[Update] There is also a new Harris-Decima poll, but unfortunately, we don't have the provincial breakdown yet. Nationwide, the Tories stand at "only" 34%, 28% for the Liberals and 17% for the NDP. The Canadian Press uses the term "Deja vu", but I would like to see the numbers in Ontario before saying that. I'll update the latest projections as soon as I have access to those numbers.

Sondage Légermarketing 14 Mars: projections provinciales

Note: while this blog is mainly written in english, I've decided that posts related to Quebec's politics will be written in french from now on. It makes more sense after all.

Alors que le dernier sondage Légermarketing nous indiquait que le Parti Conservateur avait une avance de 13 points sur les Libéraux et fleurtait avec une majorité, les intentions de votes provinciales nous montrent un Parti Liberal de Jean Charest au plus bas, à seulement 27%. Le PQ de Marois continue d'être assez largement en tête, avec 37%. Il semble cependant que la barre des 40% devienne difficile à franchir pour la formation souverainiste. Au final, je crois que Marois peut être sereine quant au vote de confiance qui aura lieu d'ici peu.

L'une des raison pour laquelle le PQ semble incapable de dépasser 40% pourrait être le fait que Québec Solidaire récolte 11% des votes. La plupart des sondages récents ont placé la formation de gauche aux alentours de 10%. Est-ce que cela arrivera lors d'une vraie campagne électorale? Dur à dire, mais si tel était le cas, QS pourrait créer quelques surprises, surtout sur l'île de Montréal.

En utilisant ce sondage et le modèle, cela donne les projections indiquées. Le PQ remporterait facilement une majorité. Le PLQ tomberait à 34 sièges et Charest serait personnellement battu dans Sherbrook. Pas que ce soit vraiment pertinent, car à 27%, il démissionnerait de toutes manières.

L'ADQ reste en-dessous de ses précédents scores sous Dumont, mais grâce à la chute des Libéraux, ferait des gains en termes de sièges.

Quant à QS, ce parti conserverait Mercier, mais gagnerait en plus Gouin et Laurier-Dorion (de très peu). QS aurait également des chances dans Anjou et Sainte-Marie-Saint-Jacques, mais il faudrait un gros effort de ce parti dans ces deux comtés.

En passant, la fameuse règle qui voudrait que le PQ obtienne davantage de sièges que le PLQ, à pourcentages égaux, ne semble plus être vraie. Je posterai bientôt une analyse démontrant que si le PLQ et le PQ devaient être à égalité, le résultat de l'élection ne serait probablement pas une facile victoire du PQ, comme en 1998. Tout cela pour dire que le PQ, bien que bénéficiant d'une confortable avance, n'est pas à l'abris de se retrouver avec une très faible majorité, voire une minorité. Cela dépendrait de la capacité du PLQ à rester au-dessous de 30%.

New poll from LegerMarketing, 14th March 2011

We are getting really close to the federal budget and the polls keep coming. And they keep showing a huge lead for the Conservatives. The last one, from LegerMarketing, show a 13 points lead nationally for the Tories, and a big 14-points lead in Ontario. Getting high numbers in this province is naturally the key for a majority (since it seems pretty obvious that the Conservatives have given up any hopes they had to increase substantially their number of MPs in Quebec.

The Liberals are down pretty much everywhere, with the Atlantic being their only dominant region in this poll. At only 27% in Ontario, they would win only 29 seats in this province. Can you imagine that in 2000, this aprty could win all but two seats in this province?

The NDP holds up pretty well everywhere, except in BC where this poll puts this party last. On the other hand, this is another poll placing the NDP very high in Quebec, at 20% for second place! We have seen this trend going on for a while and I'm convinced that the NDP will be a real contender in this province next time. Probably not in term of seats (the party can only dream of 2-3 max), but at least in terms of percentages. Imagine this aprty with Thomas Mulcair as leader...

The Greens are projected to win one seat, Guelph. While I'm always skeptical of polls for the Green party, I have to wonder why the heck did Elizabeth May chose to run in BC... I'll post about that another day. But I sometimes think that she isn't actually interested in being elected.

At the end, if we indeed go in election, it's gonna be very interesting because we would start it by having a virtual Conservative majority. And this would be new. Yes in 2008, we were wondering if the Tories could get a majority or not, but it never was very likely (until the middle of the night of the election!). But this time, it is completely possible and it should therefore affect the dynamic of the campaign. Speaking of which, according to you, what does Igniatieff need to do to remain leader of the Liberals? I think it's pretty safe to assume that a CPC majority would send him back to Harvard. But what about another minority? How many seats does he need to gain compared to Stephane Dion? I'm not sure about this.

Second choices: what if?

In its latest poll, Ekos asked the second choices, as well as asking who they voted for in 2008. I always find those questions very interesting.

For the retention rate, we see that the Conservatives are the best. They would keep 77.3% of the people who voted for them in 2008. The Liberals would keep only 63.8%, while the NPD and Bloc stand at respectively 61.6% and 73.8%. Surprisingly, the "lost" voters of the Liberals would actually go the the NDP (13.3%) and the Green (14.2%), more than to the CPC (11.5%). This is one of the only good news in recent weeks for Igniatieff: it seems he has a pool of potential voters in the Green and NDP. In case of an election, I think it's likely that those people could quickly come back to the Liberals if te prospect of a Tories majority becomes real. Of course, for the Lib, keeping the voters of 2008 is hardly enough to make some gains in term of seats. But it's a start.

The second choices show how the Conservatives are different. Indeed, they are the party with the most voters that have no second choice at all (as much as 48.6%!), while this number is only around 25% for the Liberals or NDP. I was also surprised to see that among Green voters, the second choices are split almost evenly between the NDP and the Liberals. I was expecting the NDP to have the edge here.

So using those numbers, let's do a little bit of math. In particular, let's see what would happen if the Bloc didn't exist in Quebec, and what would happen if the Green party didn't exist. For each scenario, I will also input those calculated numbers into my model, even though the model is not meant for that. I'm assuming that people without a second choice would simply not vote.

Scenario 1: No Bloc.

actual no Bloc seats
cpc 15.4% 22.3% 20
lpc 21.7% 30.6% 36
ndp 12.5% 23.8% 13
green 11.3% 19.3% 6
bloc 35.9% - -
other 3.2% 4.0% 0

In term of percentages, the NDP would be a clear winner. However, in term of seats, the Tories would actually get all the forteresses of the Bloc (in rural Quebec). The Green would get as many as 6 seats, but this is only because in this poll, they were already high. As opposed to the projections using this Ekos poll, the CPC would gain 12 seats, the LPC 19. the NDP 12 and the Green 6.

Scenario 2: No Green Party of Canada

Here I'm assuming that the second choices are the same in every province, since Ekos doesn't provide us with the provincial breakdown.

actual no Green seats gain
cpc 35.2% 37.9% 137 0
lpc 27.8% 31.2% 84 -4
ndp 14.9% 17.9% 34 2
green 10.1% - - -
bloc 8.8% 9.6% 50 2
other 3.1% 3.5% 0 0

In term of seats, not many changes, even though the Green Party seems to be costing the NDP, 2 seats and 2 other seats to the Bloc, both at the expense of the Liberals. Since the Tories don't suffer any overall loss, it seems the Green party is only affecting the distribution of seats among the other parties. If you want to compare province by province, here is the breakdown:

Ekos projections:

Ekos No Green:

You can see that at the province level, the absence of the Green party would also affect the Conservatives. But at the end, it seems that the Green party is not big enough (yet?) to actually change the colour of the government or cause substantial seat cheanges.

New Angus-Reid and Ekos polls

We just got two new polls, from Angus-Reid and Ekos. The latter shows a little fall back for the Tories, while the former actually show a big lead for this party. Using only the AR poll, the Conservatives are projected at 154 seats, thus a majority (remember I don't project the three territories) while the Liberals would get only 71. On the other hand, by using only the latest Ekos poll, the Tories stand at 137 seats only. When using those two polls as well as the other recent ones from Nanos, Ipsos and Harris-Decima, we arrive at the projections illustrated in this post. Details are here.

So it seems there is some kind of volatility. AR conducted the survey online between March 8th and 9th, while Ekos conducted its survey between February 24th to March 8th. That could explain the reason, even though given the recent scandals affecting the Tories (court decision about the 2006 election and the in-and-out funding scheme; Oda; etc) we could have expected the opposite: the survey conducted recently only should show the Tories lower. In any case, if you actually apply the margins of errors, most changes are not significant.

One thing to notice in the AR poll is the 20% for "other parties" in Atlantic Canada. Did I miss something there? Was there an independant candidate who announced he would run? The talk of a new party? Because 20% is really high.

Also, the two polls show the Bloc falling slighlty in Quebec, to around 35%. They would still get a lot of seats of course, thanks to the stupid electoral system and the division of the federalist vote, but there could be something there. How low can the Bloc actually go?

So, does it look like we are going in elections? I really don't know. There are talks that the Liberals could use their opposition day, March 21st, to bring the government down (or at least try to) but I find it very unlikely that the three opposition parties would trigger an election one day before the budget. After all, the general population isn't that aware of the recent scandals.

I'll post right after about the second choices and the retention rate of each party, regarding their 2008 voters. Some interesting stuff there as well, from the Ekos poll.

As requested: comparisons with 308

Here is a comparison with 308. I used his numbers and inputed them in my model. You can compare these results to 308's ones here. Both models give similar results, even though there are still differences in the Atlantic and in Ontario. For the former, my guess is it depends if we assume that the effects such as the ABC campaign are gone or not. I might revise my decision to assume that 50% of those effects are gone later. Depending of some polls during the next campaign.

At the end, with 308 and DemocraticSpace (whcih I'm sure we'll be active during the campaign), I think people have enough choices regarding riding-level projections (you also have other sites, such as cdn election watcher). I personally see the different models as all providing some truth. A little bit like it's better to have more than one poll, it's better to have more than one model. Even though, at the end, for one specific election, one model will of course be better. And so far, only my website allows everyone to use the model.

New projections

New projections
Here are the latest projections using an average of all the recent polls. Since most of them gave the Conservatives a big lead, the new seat projections are naturally more favourable to the Tories. They are getting very close to a majority. You can see the details here.

By the way, I'm glad to see that 308 has finally updated his model and is now able to make riding-levels projections. I'm particularly glad to see that his projections are now much closer to mine, especially regarding the Liberals and the NDP. I have been complaining for months that 308 was underestimating the NDP and overestimating the Liberals. So it's good to see that 308 has finally admitted there was a problem.

Oh I would like to just add one comment about the recent debate regarding polls. I love polls, of course. But I have to admit that pollsters make me sad sometimse. In particular, I hate when the pollsters improvice themselves as political analyst and give us, along their new numbers, the reasons as to why a party increased or not. Not only are those changes within the margin of error most of the time, the fact remains that to assess whether one news really hurt a party or not, you need much more than a simple poll. After all, I completed my undergrad with a minor in political science and I had to read electoral studies. When you do, you realize how complicated it can be to determine the effect of one issue. So I'd really appreciate if pollsters could simple improve their methodology, calculate the margin of error correctly and stop thinking they have a ph.d in political science. For instance, I think that one big mistakes of almost all polls is that they don't interview comitted voters only. How can you interview 1000 people, get their vote intentions and think it's perfect? When you know that on election day, more than 40% of people won't vote?

What does the swing look like? Proportional or not?

When a party's support increases in a province, how does it translate in each riding? The answer to this question is crucial as to which forecasting model to choose. So let's look at some graphs.

Let's look at the Conservatives (a party who experienced quite a lof of changes at the provincial level during the last 5 year), in Quebec (where the model usually performs at its best, both for the federal and provincial elections) during the last election (2008). The next graphs displays the swing at the riding level, sorted by the level of support during the previous election:

Remember that province-wide, the Conservatives dropped from 24.6% to 21.7%, thus a decrease of 2.9 percentage points, or -11.8%. Therefore, if the swing was uniform, we should see all the dots on the same horizontal line, at 2.9 (red line). On the other hand, if it was proportional, we should see the dots as forming a straight line decreasing (representing -11.9% of the previous level of support) (green line). As we can see, both assumptions would clearly be wrong. There is considerable amount of variation between the riding-level swings. You have an easy-to-understand graphical representation of how those two simplistic models could be wrong.

By the way, you get similar plots if you use other parties and/or provinces. Just to prove that I didn't use the one example that proved my points, here is a similar graph for the Liberals in Ontario, in 2008.

Here, I didn't bother showing the proportional change line, since it is obvious it wouldn't fit the data. Rather, I added the projection of the model. While I agree it isn't very clear that the model is performing better, what I would like you to notice is that the projected values are not on a straight line. Using regional effect, incumbency or the level of support last election, (for each party) my model is able to better forecast the outcome, without having to rely on an arbitrary assumption such as uniform or proportional changes.

Let's go back to the CPC in Quebec. Another aspect that might be surprising is the fact that the Conservatives actually increased their share of votes in some riding, despite a drop at the province-level. How to explain this? Well, first of all, individual effects could play a role, such as a new candidate or other specific events (for instance, what happenned to Maxime Bernier). Second of all, we are looking at shares of votes here, meaning that even though a party lost some votes, it could still increase its share if the other parties lost even more votes. For the model, I project every party and then make sure the percentages sum up to 100%. However, on e downside of the fact that riding sometimes go against the provincial trend is that my model could lead to weird results. Indeed, since I'm estimating my coefficients, the model could give parameters such that the support for one party would decrease in one rding when this party increases at the provincial level. This can be a potential problem but since I'm using more than one election, hopefully those weird effects will not biase the estimates.

At the end, it is really difficult to find evidence that the swing is proportional, especially in Ontario and Quebec. This is the reason the proportional swing model performs significantly worse than any other models, incuding the linear one, as shown in my methodology pdf. And this is why I find it weird when I see some websites basing their model on this assumption.

I'll post soon about how proportional models can even be worse for small of third parties.