With a couple of newly released polls from Nanos, Ekos and Harris-Decima in the last few days, it is time for a major update to the seats projections. These new polls don't agree on everything (some show a double-digit lead for the Conservatives, some only a 7-points one), but they usually agree on something: the trend in Ontario. It seems the Conservatives are losing ground on this province. They are still first of course, but the lead is now closer to 5-6 points (even closer in some recent polls), as opposed to 10-points and more at the beginning of the election. What this means in terms of seats is naturally that the majority is slipping away from the Conservatives. They are still projected to secure a stronger minority than currently, and a majority is definitely a possibility given the number of clsoe races, but the trend is not going up for this party. They would need to win 23 out of 31 close races, a really high percentages of victories.

For the Liberals, they seem to have gone up in the Atlantic and as I've said, in Ontario. The last Ekos poll also shows them quite high in Quebec, but numbers are a little bit off in this province with the Bloc at only 28%. So I wouldn't trust these numbers too much. Nevertheless, Michael Ignatieff is still very far from even dreaming of forming the government. Best case scenario seems to bring Harper to another minority, and a smaller one than in 2008. What would that mean for the next House of Commons? I have no idea. Seriously, with a coalition all but ruled out, it seems that we would get the same minority government as before. I always wonder what would be considered as a loss or a win for Ignatieff. A Conservatives majority would sure send him back to Harvard, but how many gains does he need over Stephane Dion's results in order to claim some kind of victory?

The NDP is slightly going down, even thought this party is resilient as far as seats are concerned. Thanks to a large increase in Quebec, Jack Layton could get another MP from this province. That would already be quite an achievement. They are projected to win 11 out of 16 close races, a number a little bit too high. If you have an election pool, I would suggest to put the NDP around 32 seats, not 35.

The Greens don't have any projected MP, nor do they even have a close race. Even May would fall short. And since it seems she definitely won't be part of the debates, the potential for grotwh isn't big for this party.

Finally, the Bloc is down a little bit and would lose a couple of seats, largely due to the increase of the NDP and the Liberals.

The riding-by-riding projections are available here.

This week, my article in the national post will be about the GTA and Vancouver, in particular: can the Tories really get new MPs there. It will be available on Friday, like last week.

Seems like I'm contributing to another newspaper and I didn't even know it! Well seems like the newspaper didn't know that either! This morning I got an e-mail from Nanos asking me to change the name of one of my files (on google docs) cause... The Standard thought it was a document from Nanos, providing polls in every riding!

I mean, seriously, how can a journalist really thinks Nanos hosts its (crappy) pdf on Google docs? With no methodology, no details? And how can they believe Nanos could poll every riding?

As I've said, funny thing. Seems a couple of Conservatives got excited for nothing as these projections were using a single Nanos poll puttng the Tories way higher than the average in Ontario.

The same way polls are provided with margins of error, it would make sense to provide seats projections with some sort of uncertainty. But how to proceed? I'll explain here how I proceed and why.

1. Using the MOE in each province.

One logical solution would be to make projections using the MOE of the polls, or the average of polls since this is what I'm using for the "official" projections. It should be noted that projecting a federal election is more difficult than a provincial one because we need to estimate and apply the model in many provinces. And because pollster usually have a 1000 observations sample size, the polls are much more accurate when these 1000 observations are from one province only (like from Quebec) than from the entire Canada. Therefore, even by using an average of polls, we are still left with really big MOE in some provinces like the Atlantic or the Prairies. Here is an example.

Let's focus on BC, a province with 36 ridings. The Nanos polls have around 150 observations for this province, meaning MOE of 8-points. Let's see how much we can shrink this MOE by averaging most recent polls. The last Ekos poll had 302 observations for this province, thus a 5.6-points MOE. We also had a recent Leger poll, with 349 observations and a 5.6-points MOE. Finally, we also had a Harris-Decima poll with MOE of 6.2-points for BC. Those are the four most recent polls (note: I consider the Nanos as one poll only since every day, the update consists of adding around 400 new observations only. Hence, Nanos has one new full poll every 4 days). If we average these four polls, we get a MOE much smaller of course (here, a big assumption is that the only difference between these polls is the sampling size. In reality, we know that other things matter as well, such as the methodology (online or by phone for instance). But I'm not taking that into account). The MOE for the "meta poll" is 3.2-points only (if you wanna know how to do this calculation, simply sum the square of each MOE, take the square root of the sum and then divide by the number of polls). So basically, at 95% confidence interval, the percentages for BC look like this (note: technically the MOE is NOT the same for every party, it depends on the level of support. But pollsters and media usually use the approximation and give the same MOE to everybody):

CPC: 40.2% ±3.2%

LPC: 22.3% ±3.2%

NDP: 24.1% ±3.2%

Green: 11.3% ±3.2%

If you do the same calculations, you get a meta MOE of 3.8% in the Atlantic, 2.3% in Quebec, 1.9% in Ontario, 3.6% in Alberta and 3.4% in the Prairies. Therefore, to calculate the upper bound for the Conservatives, we should input the upper bound of the confidence interval in every province. What we are supposed to do with the other parties is not that clear. Should I also input the Liberals at the lower bound? It would make sense since if one party is doing better, it means the other ones are doing worse (in each province, the votes for the 4 four (5 in Quebec) main parties usually sum to 96% or even more, leaving very little space for the "others" parties). Let's try first to simply input the Tories at the max, leaving the others at the average.

Using the last 4 fours polls mentioned, the projections would change from 148 to 157 seats. Inputing them at their lower bound for the percentages, they would get 136 seats. This means that the 95%-confidence interval for the Conservatives is 136 seats to 157 seats.

The best case scenario for the Liberals would increase their seats from 75 to 82. The worst case scenario would be 71 seats, meaning the 95% CI is 71-82.

So overall it makes sense. It requires a lot of calculation of course. But is there a simple way to find these confidence intervals? Let's look at solution number 2.

2. Using the "close races".

What I call a close race is what others might call a marginal seat. This is a riding won by a margin of less than 5-points. Using the four polls mentioned, the Tories are projected to win 131 safe seats and be involved in 38 close races. For the projections, they would win 17 out of those 38 races. For me, a 100%-confidence interval would be 131-169. It is way bigger than the 95% CI calculated above. So what can we do? Well, there is no way that a party would lose ALL the ridings where this party was close. Neither is there a chance to win them all. If you look at the past elections, parties usually win between 30% to 70% of close races. Let's apply this here. If the Conservatives were to win only 30%, they would win 131+30%x38=143. If they were to win 70%, they would be at 158. That would give us a CI of 143-158, not so far from the above calculations. For the Liberals, the same interval would be 58+12=70 and 83. Here we are even closer.

At the end, this method of simply attributing more or less close races is of course less scientific. You can't use MOE to back this method. But it seems to provide a good approximation of the uncertainty in the projections and they require way less calculations. On top of that, one could argue it is more reality-based (as opposed to being based on some theoretical statistical concept).

For the final projections, I will probably do the two calculations. But for now, I'll simply mention the number of safe seats and close races.

Note to my English readers: since I firmly believe in a bilingual Canada and since this post is about Quebec, I decided to write it in french. You can use Google translate if you really have to, or better, try to read it anyway. Posts in french will remain the exception.

A quoi ressemblerait le Québec au niveau fédéral si le Bloc n'existait pas? Quel parti en sortirait gagnant? Afin de répondre à cette question (à laquelle j'ai déjà tenté de répondre sur ce blogue, mais en utilisant une méthodologie moins adaptée), je vais procéder ainsi:

1) Utiliser les dernières projections pour le Québec.

2) Dans chaque comté, je redistribue les votes du Bloc selon les second choix exprimés par ses électeurs, tels qu'indiqués dans ce sondage Ekos.

Avec le Bloc, les résultats sont les suivant:

	votes	sièges
PCC	20.3%	11
PLC	19.0%	13
NPD	16.1%	1
Verts	5.4%	0
Bloc	37.7%	50

Une fois les votes du Bloc redistribuées, voici les résultats:

	votes	sièges
PCC	29.1%	20
PLC	29.1%	23
NPD	30.8%	32
Verts	10.9%	0
Bloc	0.0%	0

Le NPD sortirait grand gagnant et ce petit exercice démontre bien à quel points les électeurs Bloquistes sont proches des électeurs NPD, sauf pour la question constitutionnelle bien sûr. 33% des électeurs Bloc ont le NPD comme second choix, loin devant les autres formations politiques (les détails sont dans le sondage Ekos).

Voici les détails. Je n'ai pas fait en sorte que la somme des % soit 100% (i.e: redistribuer ceux qui n'ont pas de 2e choix et ne voteraient pas). Cela ne change pas le gagnant de toute manières. Il est quand même remarquable de constater que le NPD remporterait 31 des 50 sièges du Bloc, alors que Libéraux et Conservateurs se partageraient le reste.

Si le Bloc n'existait pas, le NPD aurait carrément une chance de devenir l'opposition officielle avec autant de sièges au Québec!

Note: Every day, a post is dedicated to the newly released polls and projections are made using those polls only. This shouldn't be considered as the official projections of this site as they are less accurate since based on less polls. The "latest projections" based on an average of most recent polls are available in the right column of this site and are updated every 3-4 days.

We are lucky today since we not only get the daily update from Nanos, but we aslo get a poll from Legermarketing with a huge sample size of 3000. They don't tell the same story everywhere though. The Conservatives keep increasing in the latest Nanos update and now stand at 42% nationally! They seem to have taken the lead in the Atlantic (but carefull with the small sample size) and regain a more comfortable lead in Ontario. As for the Liberals, the boost they got 4 days ago is definitely gone.

In the other poll, the story is a little bit different. The Tories stand at "only" 37% and they lead in Ontario by only 39% versus 34%, which would represent an improvement for the Liberals as compared to 2008. Leger also place the Liberals first in the Atlantic. One thing that surprised me in this poll is the answer to the question "which got the best campaign so far?". I was expecting the Liberals to score better than they do.

I haven't updated the main projections yet as I would rather wait for one more poll, for instance from Angus-Reid. But if you wanna know, the Conservatives has slipped just below a majority (I give more weight to the Leger poll than the Nanos one since the sample size is 3x). By using the two new polls only, we get the displayed projections (details are here). While the Tories are a little short of a majority, they are still projected to win 135 safe seats and be involved in 28 races. So the majority is still possible, just less likely. What this poll shows is the ceiling the Tories have reached in the West. Being at 39% nationally is nice, but with only a 5-points lead in Ontario, it doesn't really help. They really need to either increase in the Atlantic or in Ontario in order to secure a safe majority. But if you ask me, I would still bet on a Tory majority at the moment as I believe this party will be able to target the ridings it needs.

For the Liberals, one thing that seems to come from nay polls is the fact they ar eback in the Prairies. Ok they are still far behind the Conservatives, but they are definitely up compared to 2008.

Nothing else to say so far. I'll probably write a more analytical post later today. I'm thinking about looking where the Bloc is in danger.

The boost the Liberals got 3 days ago, when they jumped 4-points to 32%, seems to be going away slowly. Of course, I'm well aware that if you apply the margins of errors, most (if not all) the changes observed so far are not significant (especially at the provincial level since the MOE are so big). But writing a post everyday about unsignificant changes would be kind of boring. So I don't bother much about this, I take what we see.

At first, you would think that the latest Nanos poll is only good news for the Tories. 41.3% nationally? 11-points ahead of the Liberals? Well not so fast, it depends where this lead comes from. And the increase comes mostly from Quebec where the Conservatives are at 26.7%. In the mean time, in Ontario, the 16-points lead observed the other day has been reduced to 45-34.5 lead. Still comfortable but maybe not enough to win a big majority (my post next week in the NP will be about what the Conservatives need in order to gain seats in Ontario and specifically in the GTA). As for the Prairies, increasing the level of support over there is really useless for Stephen Harper. Finally, this Nanos poll still gives the lead to the Liberals in the Atlantic. So at the end, despite a huge national lead, the potential gains for the Conservatives aren't that big. Let's focus on Quebec.

In 2008, with 21.7% of the votes, the Conservatives got 11 seats (including André Arthur). In 2006, with 24.6%, they got... 11 seats. Ok the Bloc's support decreased as well during the two elections and that explains partially why the CPC didn't lose seat. But the big story is: their seats are safe. That also means: potential gains are difficult! If we use the Nanos poll only with the Tories at 26.7% and the Bloc at only 35.8%, we get the following:

CPC: 13 seats

LPC: 17

NDP: 2

Bloc: 43

Surprinsigly, the Tories are not the party who gets the msot benefits from the drop of the Bloc. Their only gains would be Chicoutimi-le-Fjord and Louis-Hébert. They would also get close in the following ridings (and I'm being very generous in my definition of close): Drummond, Gaspésie, Haute-Gaspésie, Abitie-Baie-James, Québec and Montmagny-l'Islet-Kamouraska (won in a by-election). Yes if the Conservatives could get all these ridings, that would represent important gains. But where they stand right now, their national lead isn't that useful as far as gaining seats is concerned.

Of course, as I've said before, the Conservatives might not need a big lead everywhere. They could simply focus on the key ridings they want to win. In Quebec for instance, Harper could visit the 3-4 ridings where gains are possible. And that could well be enough to secure a majority.

By the way, I'll not do daily projections using this Nanos poll as the changes are not big enough. As always, you can simply use the simulator.

Note: Every day, a post is dedicated to the newly released polls and projections are made using those polls only. This shouldn't be considered as the official projections of this site as they are less accurate since based on less polls. The "latest projections" based on an average of most recent polls are available in the right column of this site and are updated every 3-4 days.

First of all, I would like to say two things. 1) Welcome to the new readers that came after reading my article on the National Post. I'll have such an article every week. 2) I thought about doing an April's fool with a post such as "Ignatieff can expect a majority according to latest poll!" but at the end, I decided not to. Maybe I would have attracted and fooled one newspaper!

So today we've got the weekly Ekos poll as well as the Nanos daily update. In the latter, the Liberals drop a little bit, after a big surge yesterday. The Conservative remain high and still have a very big lead in Ontario. The Ekos poll is more in line with the other polls and show the Liberals around 26%. One region where the two polls are widely opposed is the Atlantic. This is probably due to the sample size (seriously, having a rolling average of 99 observations is almost ridiculous. It means every day they add 30 observations and drop the oldest 30 interviews. The accuracy is clearly weak).

Based on an average of those two polls (with more weight given to the Ekos poll given the sample size), you get the dailies projections displayed in this post. The main reason the Tories are down is naturally the Atlantic (where the CPC is really behind according to Nanos). Hopefully, we'll soon get more polls to determine where the parties stand in this region.

One thing where both polls agree is about the Bloc, which is projected at 36%. That would mean this party would experience another drop in terms of percentages and lose a couple more seats. You can say what you want about the Bloc but there is definitely a trend going downward for this party over the years. Of course, thanks to the division of the federalist vote, this formation can still get over 40 MPs, but this is really the only reason. I said it before and I'll say it again, the Bloc could well disapeared if two reformes were implemented: abolishing public funding (as suggested by Harper) and the introduction of proportional representation. But this is another debate of course.

I really wonder what would happen if, on Election Day, the actual results was like this one. Would we really simply go back to another Conservatives minority? With the same instability and political drama every confidence votes? I find it quite crazy that it could actually happen.