May 16th 2013: What if I had used only the best poll?

May 16th 2013: What if I had used only the best poll?
Among all the pollsters for the last BC election, Forum did clearly better than anyone else. Pollsters seem to go for the "late swing" excuse again, yet the best polling firm conducted its poll... one week from the election! Forum wasn't spot on, but they were close enough that a late swing could indeed explain the discrepancies. Kudos to this firm as they were already the only one to catch the much closer race in Alberta (while still being off though) and the only one to have the Liberals much higher than anybody else in Quebec. So while not perfect, it seems that Forum polls are at least within the margins of error (or very close).

So what would the projections have looked like had I used only Forum's numbers?
Well, the model and simulations would still have predicted a BCNDP victory, but the actual outcome would have been much more likely to happen. When I used an average of all recent polls, I had the BC Liberals with around 13-14% of winning. But even in the scenarios where Christy Clark was winning, 50 seats was really at the extreme of the intervals (and outside of the 95% confidence intervals).

But when using Forum's numbers only, here are the seat ranges (at 95% confidence) for each party:

BC Liberals: 30-52
BC NDP: 33-54
BC Green: 0-1
BC Conservatives: 0

May 15th 2013: Post-mortem for BC

The day before the election, I tweeted that the BC Liberals had more chances to win the election than the Boston Bruins had to come back against the Mapple Leafs (at 4-1 in the middle of the third period). Based on the polls, I had the BC NDP's odds of winning around 86-87%. Seems that the Bruis' win should really have prepared us for last night!. But I can at least argue that I technically had the Liberals victory as a possiblity (although the magnitude of the victory, with 50 seats for the Liberals, was really at the maximum of the projection ranges). However, the actual result was incredibly unlikely. At least based on the information we had.


Before I talk about how wrong the polls were, above is what the projections would have looked like if I had the correct vote shares. Details can be found here. As you can see, the model would have performed quite well, making the correct call in 75 ridings (so a success rate of 88%). Some of the mistakes include the ridings of Christy Clark, Oak-Bay-Gordon Head (where I didn't think Andrew Weaver would get such a big boost but it was a possiblity as well in the model) or a couple of close races in the lower mainland. It also seems the model would have made quite a lot of mistakes on Vancouver Island, probably because the regional swing (of the Green especially) was so different from the provincial swing. Since some of these 10 mistakes cancel out each other, the overall projections would have been really close. The fact is that if you came to my blog and input these vote shares in my simulator, you'd have won your political pool! So "my part of the job", i.e translating votes into seats, is clearly well done. Let's not miss the achievement here: by having the provincial vote shares only, the model could have correctly called the election as well as 88 % of the ridings. On top of that, if you look through the pdf, you'll see that more often than not, not only the correct call would have been made, but the vote shares in the riding are very close to the actual ones. It just shows how important the provincial swing is. So translating votes into seats is totally possible, with very little information (but a lot of work from me). Now, even more than before, the key is really to find or project the correct vote shares.

That, honeslty, shouldn't be my job. Pollsters are there for that. We've got a lot of polls during this BC election, including like 3-4 during the last two days. They were all showing the same outcome: BC NDP ahead by 5-7 points (except Forum who had the BC NDP up by 2 only, but they released the poll about a week ago, so they might just have been lucky). So even by averaging the polls, you'd still have been far off. If I accounted for the true statistical uncertainty due to these polls (i.e: margins of error), my simulations would never have allowed for the possibility of a BC Liberals win. Indeed, when you combine polls, margins of error go down. It's only because I add a lot more randomness that I technically had the actual outcome as a possibility. My only regret is that two ridings that were called with probabilities of 100% went wrong. I thought I was accounting for enough uncertainty, but it seems I'll have to add even more, both at the provincial and riding levels. The probabilities should account for the possible error of the polls. They already do, but not fully it seems.

Honestly, it's quite depressing and frustating to work so hard on building a model just to be so wrong because all the polls are providing you with incorrect information. In the US, during the last election, you didn't have to be Nate Silver to predict the election. A simple poll average would have been (almost) spot on. So pollsters in this country should really ask themselves why they can be so wrong. And not just one or two polls, but all of them! It's the same thing that happened in Alberta (and to a lesser extend in Quebec) last year (we could also include the underestimation of the Conservatives last federal election, although some polls were quite close). It's almost funny because people argue a lot about the correct methodology (online panel vs IVR vc phone calls), but the issue lies elsewhere. I'm sure pollsters will come out (they have started already) talking about how there was a late swing and blablbla. But to me, the problem is they don't poll the right people. Ekos is trying to identify likely voters but is obviously failling miserably.

I'll continue to build model and offer projections. After the last Quebec election, I decided to add uncertainty as to account for the possibility of big mistakes in the polls. I'll thus continue that way and hopefully, polling firms will improve on their parts.

May 13th 2013: Final BC projections

[Update: even though we've got three new polls today (Ekos, Angus-Reid and Ipsos), adding them to the projections doesn't change anything. Actually, even if I was to use these three polls only, the projections below would still be valid. So I won't be updating the infography or the pdf]


Given the polls, past results and the electoral system, taking account the uncertainty due to theses factors, Too Close Too Call projects that the BC NDP will win a majority government on May 14th 2013, while the BC Liberals will form the official opposition.

Adrian Dix and his party have 87% chances of winning the most seats and 86% of getting a majority. For the BC Liberals, it'd take a mix of a serious (and systematic) bias in the polls as well as some "luck" or efficiency with the electoral system in order to keep power.

Just a couple of weeks ago, the possibility of a BC Liberals surprise didn't even exist. So Christy Clark has at least succeeded in bringing her party back in a position where they not only can avoid a landslide victory of the BC NDP, but can even dream of beating it. Still, with the odds being around 13%, you have to like risk to gamble money on the BC Liberals.

If you want the riding-by-riding projections, you can find them here.

Let's take a look at each party. First, the very likely next governming party in BC, the BC NDP of Adrian Dix. After three consecutive elections won by the BC Liberals, all with smaller and smaller margins, the BC NDP is poised to come back to power. While the odds are very good for this party, let's not forget that 4 weeks ago, they were projected to win a majority 100% of the time. They even had the possibility of winning as many as 80 seats. So, did Adrian Dix make mistakes during this campaign? I wouldn't say so. I believe it's more a return to the natural order. In particular, the BC NDP didn't drop in voting intentions (a little bit, but at almost 45%, they are very close to their average of the last 6 months). What really happened are the smaller parties (Green and Conservatives) declining. The only negative point is that the BC NDP wasn't able, it seems, to benefit from some of the Green support going away, at least not looking at the net effect. It's well possible some of the Green supporters did indeed go back to the NDP, but at the same time, some NDP supporters moved to the Liberals. At the end of the day, the net effect has been a very stable NDP.

Christy Clark started this campaign with very little hope of winning it. With around a 20-points deficits, it would have taken a miracle for her to remain Premier. Still, an effective campaign and the collapse of the BC Conservatives gave her a chance. It'll likely not be enough, but she'll make sure the BC NDP is at least facing a real opposition in Victoria. And while she was projected to lose her own seat for the best part of the election, we now project that she'll indeed be back in Victoria. For how long? And will it be as the leader? We'll have to wait and see.

The BC Green party has great expectations this election. But while polls still look good (especially considering they could get a positive provincial swing despite running 24 less candidates, thus meaning that in ridings where they do have candidates, the swing is definitely positive), a decline towards the end of the campaign puts the chances of getting MLA(s) in jeopardy. Their best chance has always been with Andrew Weaver in Oak-Bay-Gordon Head but it'll take a significant "candidate-effect" for him to win. Something comparable to what Elizabeth May achieved last election. By Canadian standards, this is pretty unique. I believe the Green will be in the race but will ultimately fail. The main reason being the BC NDP polling so high on Vancouver Island and showing no sign of decline.

As for the BC Conservatives, the trend of the last year is quite disastrous. They went from being in the race to beat the BC Liberals for second place to most likely finishing last (among the four parties) with no MLA. An objectively weak campaign marked by candidates withdrawals and other controversies, along with what has been considered the weakest debate performance, all point to a bad night for this party and his leader, John Cummins. At 6-7% provincially without a full slate of candidates, the party could be in the race in some ridings around Kelowna or in the Interior, but as the Green, they should ultimately fall short. Still, they could be the party with the best provincial swing, something not rewarded (yet) with this electoral system, but something not negligeable.

Finally, let's nor forget there are a couple of independents candidates that have a real chances, including 3 incumbents (only 1 of them actually elected by running as ind. in 2009). Independents are very hard to project and I won't feel bad if my model is off in these ridings. It's honnestly more a guess than anything else.

I'll post later a short posts "by the numbers" to complete this one. Expect a lot of probabilities and, obviously, numbers.

May 11 2013: Technical adjustments when a party isn't running a full slate of candidates

May 11 2013: Technical adjustments when a party isn't running a full slate of candidates
Two of the four main parties in BC aren't running candidates in the 85 ridings. In fact, the Green party only has 61 candidates while the BC Conservatives have 56 (or 60 if you include the 4 candidates who are on the ballot without the Conservatives identification, due to an error in registration with BC Election). For projections, this creates essentially two problems.

1) Polls (usually) don't account for that and therefore overestimate the support for these parties.

Except for last night's Angus-Reid poll who made full use of its online panel to offer respondents only the options available in their riding (i.e: if you live in a riding without a Green candidate, AR would not have offered this option to you), most (if not all) polls simply ask the same generic question to everyone. So there could be a small but significant numbers of respondents who declare wanting to vote for the Green party for instance, even though there isn't a candidate in this riding. After all, it's quite reasonable to assume that most respondents don't actually know if there is a candidate for each party in the riding and will only find out on election day.
 When that's the case, the support for such a party is clearly overestimated. The effect could be particularly important with the Green and Conservatives parties since they aren't even close to having 85 candidates.

On the other hand, if the polling firm accounts for that (as did Angus-Reid), then the support for the Green party is unbiased. And if you look at the AR poll, you indeed find the Green and BC Cons below their usual levels in other polls. But in this case, it actually underestimates the actual level of support in ridings where there is indeed a Green candidate. This post from BC Iconoclast explains it clearly.

In average though, since most polls don't account for this, expect the Green and Conservative supports to be overstimated. I don't currently correct for this in my projections but will make sure to do so for the final ones.

2) The provincial swing is affected.

My model, as most of these type of models, use the provincial swing to make projections in every riding. While I do account for regional and other effects, I have to make additional adjustments for the Green and BC Cons. Here is why: the Green party got 8.2% of the votes in 2009, with 85 candidates. There are now 24 ridings where there isn't a candidate anymore. In these ridings, the average level of support received by the Green candidates was 7%. So you see that the BC Green party is "efficient" and has decided to run candidates in ridings where it has a better chance this time around (specifically, in ridings where there is a Green candidate in 2013, the average result for this party in 2009 was 8.4%). Still, these 24 ridings with an average of 7% account for a good part of the overall 8.2% level of support. In fact, without these ridings, the Green would only have received around 6% of the votes province-wide. So when the polls (assuming we corrected for the effect explained in part 1) show the Green party at (say) 10%, it means that in the ridings where there is indeed a candidate, the provincial swing isn't 2 points (10-8), but 4 points (10-6). If you don,t account or correct for this, then you clerly underestimate the Green candidates in your projections.

To take a similar example as BC Iconoclast, I currently have the Green party at 18% on Vancouver Island. So quite similar to most polls (which shows that my regional coefficients to transpose the provincial swing actually work). But this 18% is an average out of the 14 ridings of the island. In reality, the BC Green party only has candidates in 11 ridings. So if I average over these 11 ridings only, I find the Green standing at 23%.

In conclusion, remember that most polls will have the Green and Conservatives supports wrong. But count on me for making sure to account for that in my projections (as well as in the simulator that you can use on this site)

11 Mai 2013: Le PLQ sous Couillard toujours confortablement en tête.

Depuis que Philippe Couillard a remporté la course à la chefferie du PLQ, ce parti trône en tête des intentions de votes au Québec. Le dernier sondage Léger pour le Journal de Montréal continue cette série. Les Libéraux sont à 35%, le PQ à seulement 27% et la CAQ tombe encore davantage et se retrouve à 19%. Clairement, Couillard réussit à aller chercher des appuis chez les deux autres partis.


En termes de projections, vous avez les résultats affichés. Les projections par comtés sont ici. En termes de probabilités, le 7 points d'avance des Libéraux leur donnent environ 90% de chances de remporter l'élection. Les chances d'une majorité Libérale sont d'environ 55%.

Le PQ devrait espérer une sous-estimation de ses appuis ainsi qu'un peu de chance avec le système électoral afin de remporter l'élection. En effet, alors que le PQ remporterait le plus de sièges dans 90 simulations sur 1000, la formation souverainiste ne recueillerait davantage de votes que le PLQ que 7 fois! La concentration du vote Libéral fait en sorte que le PQ peut remporter l'élection en ayant moins de votes. Cela est connu, il s'agît simplement d'une autre illustration de ce phénomène.

La CAQ qui avait déjà un taux de conversion votes-sièges faible n'améliorerait pas cette situation. Dans les faits, la formation de François Legault aurait de fortes chances de se retrouver avec très peu de députés. Tellement que Québec Solidaire pourrait en fait la dépasser!

En effet, QS grimpe à 11%. Une possible sur-estimation des sondages hors campagne électorale, mais peu importe. Avec un tel résultat, couplé à la chute du PQ, la formation de gauche décrocherait 4 sièges (Gouin, Hochelaga, Mercier et SMSJ) et serait dans la course dans quelques comtés supplémentaires (Bourget, Crémazie, Laurier-Dorion, Pointe-aux-Trembles et Rosemont). La division du vote serait bien sûr un sujet d'actualité, mais serait parfois retourné. Dans Laurier-Dorion par exemple, QS a davantage de chances de gagner (35%) que le PQ (0%).

Bien sûr, les sondages hors campagne sont moins fiables. Néanmoins, avec un gouvernement minoritaire à Québec, il y a toujours une chance d'élections. Avec la CAQ si faible, il y a cependant peu de chances que Legault ne fasse tomber le gouvernement.

May 10th 2013: Don't be fooled, new polls don't show a 50-50 race

May 10th 2013: Don't be fooled, new polls don't show a 50-50 race
With the two new polls released yesterday, in particular the Oracle Poll showing a 4-points gap only between NDP and Liberals, some might be tempted to think that we now have a wild race where everything can happen. News medias sell more if they write about a race, remember the last US election?

By averaging the two polls published yesterday (so the two most recent polls), we'd get the NDP around 41%, the BC Liberals at almost 36% and the Green and Conservatives at 12.8% and 8.8% respectively. This is still a 5-points lead for the NDP. This is actually not significant. Indeed, for these levels of supports and with a sample size of 1000, the lead needs to be at least 5.4 points to be significant at 95% confidence interval). But given that we are actually using two polls with total sample size of 1800, the NDP lead is indeed significant. But remember, not being significant only means that it's possible for the Liberals to actually be ahead. It doesn't mean it's likely to be the case! If you prefer, it means that while the Liberals could be ahead of the NDP, it's just as equally possible for the NDP to actually have a very important lead of almost 13 points (margins of error work both way, remember that!).

If I run my 1000 simulations using these numbers, I get that Adrian Dix would win the election 84% of the time! Remember, this accounts for the uncertainty (or randomness) due to the polls as well as the electoral system. So it shows clearly that while we do have a much closer race than at the beginning of this campaign, the odds are still in the BC NDP favor.

Even if I was to only use the best poll for the Liberals (the Oracle poll), where Christy Clark would only be 4-points behind, I'd still find that the NDP would have 76% chances of winning. But there, any over- or under-estimation by the polls is becoming critical as each point translates into many more seats and chances of winning.

In conclusion, yes polls aren't perfectly accurate and can be quite wrong at times, but you have to wonder how likely it is. You can mention the Alberta election, but you should also remember that polls are not that wrong in average. My model actually accounts for that by going way beyond the traditional margins of error. So yes it's possible for the BC Liberals to win next Tuesday, but it isn't likely. Still, for Christy Clark, the simple fact that a victory is possible is already a huge improvement over simply 2 weeks ago!

May 10th 2013: Two new polls, both necessarily have mistakes

May 10th 2013: Two new polls, both necessarily have mistakes
Today, we've got two new polls from two firms that quite frankly, aren't the most well known in this business. The first one, from Hill & Knowlton strategies showing a somewhat similar situation as the other polls released recently: the BC NDP still first but with a lead of around 6-7 points "only". This poll has very original questions such as the voting intentions by types of vehicles driven. Is that fun? Yes. Is that useful? I'm not sure.

The second poll, from Oracle poll research, is more surprising as it shows a tight race between the two main parties. Indeed, the BC NDP would be at 41%, only 4 points ahead of the BC Liberals. With 1000 respondents, the NDP'S lead is almost non-significant.

Technically, these two polls aren't that far from each other and could just be examples of the natural statistical variation. Although some of you could argue that the HK poll is based on an online panel while the Oracle poll is a more traditional phone poll. I personally see both types of polls as valid as long as the correct methodology is used (question, weighting, etc). Recent elections in this country haven't really helped settle this debate.

However, there is one big difference between the two: the Oracle polls has the 4 main parties summing to 100% while the HK poll has 3% for the "other" parties. These are really extreme examples and are perfect for what I wanted to talk about: even though you can't be sure how wrong a poll is, you can usually be sure that there is indeed a mistake!

What do I mean by that? Well, for the Oracle poll, we know that candidates from more than 4 parties will receive valid votes for sure. So there is no way the four main parties sum to 100%. On the other hand, having as much as 3% in this category is too high and typical of polls. In 2009, the other parties got less than 2% (except if you include the BC Conservatives in this category, which was the case of most pollsters back in 2009). With now 4 parties included in the polls, it's a lot more reasonable to assume that these four parties will sum to 98%, if not more. Similar numbers can be found for other Canadien elections.

So everytime you see a new poll, if you see the other parties getting more than 2% of the vote, you can tell almost for sure that this poll is off, at least by a little bit.

The second source of almost-sure error are the Green and Conservtives. Both are more than likely overestimated in the polls. Just look at the polls of the 2009 election and you see a constant overestimation for the Green (by about 2-3 points, which actually could explain the underestimation of the BC NDP). This usually occurs mainly because the supporters of these "smaller" parties are less sure of their choice. All polls show this very clearly. On top of that, given that this time around the Green party doesn't even run 85 candidates, it's quite likely this overestimation is at least as big as it was in 2009. Same for the Conservatives who run even less candidates.

So bottom line, conservatively, most polls out there are off by a good 5 points (1+2+2). Now the real question is really: which party is underestimated? If these 5 points are evenly distributed between Liberals and NDP, the latter keeps roughly the same lead over the Liberals and its chances of winning the election are mostly the same (the electoral system can play a role here, but for simplicity sake, let's ignore that). But recent elections all had a very similar pattern: only one party was polled wrong. Last federal election, the polls were pretty spot-on for the Liberals and NDP, but were quite off for the Conservatives. Last BC election, the polls were off for the NDP. Last Quebec election, the polls underestimated the Liberals by about 4 points while being much closer for the other parties. For the current election, how these 5-points will be re-distributed could make the difference between a landslide victory for the NDP or a surprising win of Christy Clark.

I don't think it's very likely that only one of the main parties will be underestimated. The overestimated Green support will likely go to the NDP (as in 2009) while the Conservatives will help the Liberals. Still, my objective was simply to show you that polls usually have obvious errors to spot. It's harder to correct them though.

I'll update the general projections soon. Adding these two polls didn't change anything much though.

May 7th 2013: Predicting the Green party

May 7th 2013: Predicting the Green party
I said it before but it's true: predicting the Green party is harder than projecting the main two parties, for a couple of reasons. First of all, there are a lot of ridings where they had a candidate in 2009 and don't run one in 2013. This creates some technical problems for the model and I have to account for it. But the main reason is that the provincial level of support (as measured by the polls) is more likely not the only variable that matters as the Green vote might be heavily concentrated in a couple of key ridings.

Think Elizabeth May. If you only looked at the provincial swing, there is no way she would have been elected in 2011. The federal Green party actually lost votes between 2008 and 2011 in BC. Yet, May won her riding (facing a cabinet minister!). How? With a strong example of candidate effect and an intense campaign from the party.

This scenario could happen again next week. There are two potential star candidates that could overcome a less-than-enough provincial swing: Jane Sterk (leader) in Victoria Beacon-Hill and Andrew Weaver in Oak-Bay Gordon Head. In both cases, they haven't run in this riding before. So let's take a closer look.

Jane Sterk is the leader of the Green party for a second election. While she haven't been able to prevent a slide in the Green vote in 2009, current polls seem to indicate a positive swing for this party. However, polls don't account for the missing candidates in 24 ridings. Added to the tendency of polls to overestimate small parties, it might well be possible for the Green to experience a much smaller swing than expected. But what about the potential "star candidate" effect of Sterk? Well, it's hard to estimate. Not only is she running in Victoria Beacon-Hill for the first time, but the riding she ran in 2009 (Esquimalt-Royal Roads) was technically a new riding, created in 2008. We can look at her results for the 2005 election but she wasn't the leader yet. If we look at the 2009 results, it seems Sterk might have had a small and positive effect. Indeed, she got 16.8% of the vote. Comparing that to the ridings that previously composed this new one, and given the provincial swing, it is reasonable to assume that Sterk benefited from a small boost or bonus. But nothing spectacular or close to what Elizabeth May could achieve. Let's say it was between 5-10%. Looking at the current projections in Victoria-Beacon Hill, we can see that such a bonus would not be enough to beat incumbent (and former BC NDP leader) Carol James.

The best chance for a Green MLA seems to be Andrew Weaver in Oak-Bay Gordon Head. He's a well known university professor and Nobel prize winner. He's running for the first time in a riding where the Green got only 8.9% of the vote in 2009 (the riding was won by a small margin by Ida Wong, a Liberal candidate). I wrote about this before, but Andrew Weaver will need a boost or bonus of at least 15-points in order to win  this seat. By Canadian politics standards, this is less than Elizabeth May but more than what most other leaders or star candidates get. Still, my model currently gives him a bonus of 15-points (in average, in the 1000 simulations, this bonus varies between 5 and 25%, with a mean at 15%) but at the same time, the model heavily relies on provincial levels of supports (it does however includes regional effects that acknowledge that Vancouver Island is a good region for the Green party). In order to assess the real chances of Mr. Weaver, it might be better to look at other types of models. One good example is the Election Prediction Project. This one is almost the complete opposite of my model. Instead of transposing provincial swings into riding-level ones, it uses inputs from people in the riding. You can go and write your own entry, like: "I think the Green party will win as everybody around me will vote Green", or you can comment about the number of billboards, etc etc. Right now, for Oak-Bay Gordon Head, it's projected as too close to call (how appropriate, isn't it?). If you read the entries, you see a hard fight between NDP and Green, with the Liberals also in the race. This is pretty close to the current projections of my model. Bottom line: a close race in this riding where turnout will most likely play a crucial role. The Green will need to give everything they can in order to win this seat. I'll say this: if the Green party really gets 12% of the vote provincially next week, I think they'll win this riding.

Finally, I should also mention the riding of Saanish-North and the Island where Green candidate Adam Olsen might have a small chance. However, it seems the Green party is putting the money and effort in Oak-Bay-Gordon head for this election. This is probably for the best. In Quebec, the small left-wing party Québec Solidaire proceeded the same way to win its first seat. Then, 4 years later, it managed to win a second one. Small parties have to be patient. Getting a MLA with the electoral system is really hard.

At the end, we should also notice that polls have all shown the Green party 2nd or 3rd on Vancouver Island, far behind the NDP. The level of support for the Green party on the island is actually pretty close to my model (around 20-25% of the votes). This shows two things: 1) the regional effects of my model seem to work pretty well 2) The Green party isn't experiencing very different swings across the province. Specifically, it's not like the support is dropping somewhere and being transfered to the Island, thus leading to a really big and positive swing in this region. This decreases the chance of Green MLAs next Tuesday.

May 6th 2013: Before the last week, by the numbers

May 6th 2013: Before the last week, by the numbers
There is only about a week left to the campaign. So here are a lot of numbers giving you the situation at the begninning of this last week. In a nutshell: the BC Liberals have climbed back quite a bit, but for now, it isn't enough to prevent the BC NDP to regain power in the province. However, polls are now close enough that watching eleciton night might provide some suspense and surprises. If you want the latest projections, you can find them here.

Outcome
Probability
NDP victory
92%
NDP majority
90%
NDP with 70 seats or more
0.1%
NDP with 50 seats or less
41%
Lib with 20 seats or less
7.0%
Lib with 30 seats or less
44%
Cons with at least 1 seat
11%
Green with at least 1 seat
36%
NDP wins popular vote, Lib wins most seats
5%



Seats
Min Liberals
12
Max Liberals
52
Max Conservative
3
Max green
2



Other
Chances of Christy Clark being elected
78%
# riding where one party is projected to win 100% of the time
32 (26 NDP, 6 Lib)


Looking simply at the number of seats tells a big part of the story, but comparing this table to the previous one really shows how much uncertainty has been added in the last two weeks with the rise of the Liberals. We went from a sure win, sure majority for Adrian Dix to a situation where the Liberlas could potentially wins a majority (unlikely but possible).

The last week will really be crucial for the four parties. For the NDP, the game plan remains essentially the same: avoid making mistake, avoid scaring away some centrists voters and hope that the polls don't turn as wrong as they were in Alberta. For the Liberals and Christy Clark, it'll all about getting the vote out in the key ridings. But no matter how much effort they put into that, they'll still need a little help (in the sense that the polls must have been wrong).

The Green are difficult to predict. Province-wide, they are increasing but they are not running candidates in many ridings. But for them, it's less about the province than some key ridings. I'm really disappointed we never saw some riding-level polls on Vancouver Island (or I missed them). Right now, the model gives a bonus to Andrew Weaver. That is the only bonus and I'll look at that very carefully for my final projections. But Green supportes shouldn't get their hopes too high as all the polls have showns the NDP clearly leading on Vancouver Island.

Finally, for the BC Conservatives, the campaign hasn't been very good. They've had controversies, candidates who had to withdraw or forgot to register correctly with Election BC, etc etc. They are mostly stable in the polls except in the most recent Ipsos where they lose 4 points. In my opinion, this party is currently overestimated in the polls and will receive less than 10% of the vote. It could be enough for a couple of seats depending on the regional distribution of the votes. If it's heavily concentrated in Kelowna for instance, they could end up with 3 MLAs. But currently the most likely scenario is for this party not to get any.

May 2nd 2013: New Angus-Reid poll shows a much tighter race

If you have twitter and are following the BC election, you probably saw the new poll for CTV by Angus-Reid. After almost 3 weeks of campaign, it seems the race is getting closer as the BC Liberals are climbing up while the BC NDP is going down.

I'll update the general projections (that averahe the most recent polls) soon, but based on this poll only, the projections would like like this:


The riding-by-riding projections are here. As you can see in the pdf, Adrian Dix and his party now "only" have a 92% chances of winning the election. I say only cause until now, they were at 99-100% and had been there for a long time. Some might think that 8% is too little for the Liberals, but they are still polled 7-points behind the NDP. 7-points is significant and way outside the margins of error. For the Liberals to win, you'd still need the polls to be wrong. Not as wrong as in Alberta, but close. So 8% it is. The model and simulations already account for a lot of uncertainty in order to give the victory 8% of the time to the Liberals.

However, as I was saying in my previous post, I wouldn't be surprised if the race is actually even closer as I suspect the BC Conservatives and Green to be overestimated in the polls. One reason is the high number of ridings where these two parties don't run candidates. Polls don't account for that. A lot of people will show up on election day just to find out there is no candidate for the they wanted to vote for. Since the potential Green voters will likely vote blank or NDP and the potential Conservatives will blank or go for the Liberals, the net effect on the chances of winning for the two main parties is ambiguous. But I'd think that the Conservatives are probably slightly more overestimated than the Green. Also it's well possible the Green support is heavily concentrated in some ridings, especially on Vancouver Island. If the Green gets some seats there, it'll be at the expense of the NDP. So put the two effects together and the Liberals could well be only around 10 seats behind.

Finally, there has been a tendancy in this country, both at the provincial and federal levels, for polls to underestimate the center-right incumbent. Could it be the case with Christy Clark again? It's possible. I'm not sure why this tendency exists, but it definitely does.

The Liberals don't need to leap frog the NDP to win. The electoral system being what it is, Christy Clark could stay in power despite getting less votes. In the simulations (that, remember, account for the uncertainty due to the electoral system), the BC Liberals wins 85 times out of 1000. But their support province-wide is above the NDP only 33 times. It means that the Liberals can win even if they get slightly less votes than the NDP. In the best case scenario (for the Liberals, they win with 34.2% of the votes while the NDP gets 37%. This is an unlikely scenario, but not an impossible one. I pointed out in previous posts that the electoral system (and the distribution of the vote) was slightly in favor of the Liberals. At that time it didn't matter as the NDP lead was so important, but it's another story now.

So the last 2 weeks of this campaign will be crucial. Adrian Dix needs to stop the BC Liberals to increase more, especially if it comes at the expense of his party.

May 2nd 2013: Do we have a race?

May 2nd 2013: Do we have a race?
With the latest Abacus poll showing a reduced lead for the BC NDP, one may wonder if we may finally have a race in this election. Adrian Dix and his party are still heavily favored to win the next election, but for the first time since this campaign started, there is a small chance of a Liberal victory.

The latest projections (on the right column of this blog for the picture) are based on the latest Angus-Reid and Abacus polls, both conducted before the debate. They show relatively the same situation, even though Abacus has the NDP lead slighlty smaller. But as stated above, the probability of victory for the NDP is now "only" 99%. There is also a small chance of a minority government.

Looking at these two polls only, the Liberals would only have a 1% chance of winning the most seats. However, there are several factors that lead me to believe that the race is actually closer than expected. First of all, there is a slight upward trend for the Liberals (and a slight downward one for the NDP). But more importantly, I wouldn't be surprised to see the BC Conservatives fall quite a lot. Why? Well there aren't running a candidate in 29 ridings, including some where they really should (Boundary-Similkameen). In these ridings, the Conservatives voters will likely choose either not to vote or to vote for the Liberals. Additionnally, the post-debate Ipsos-Reid poll clearly shows that John Cummins is seen as having performed poorly at the debate. So all together, I think it's reasonable to assume the Conservatives will drop further and the Liberals will benefit from it.

Coupled to the fact the BC Liberals are quite high among voters above 55 (the kind of voters who actually cast a ballot on election day) as well as the surprising good hold of the Liberals candidates in a couple of riding-level polls, we could well have a final result much closer than what most would think. It's mostly conjecture of course at that point and we'll have to wait and see when new polls will be released.

Nevertheless, if the BC Liberals were to benefit from a drop of the Conservatives (while the NDP couldn't enjoy the same from the steady Green party), then we'd definitely have a race. Christy Clark and her party are in the "paying zone" where every additional percentage point can bring a lot of seats. If you use the simulator, you see for instance that a 3-points exchange from the Conservatives to the Liberals could reduce significantly the BC NDP's lead in term of seats.

The Green are projected at 0 seat but they are very close in the riding of Oak-Bay-Gordon-Head, mostly thanks to some changes I made to the model. First of all, I added a bonus for the "star candidate" Andrew Weaver in this riding. Second of all, as the Conservatives, the green aren't running candidates in 24 ridings. Therefore, if they REALLY are at 10-12% provincially, it means that they must be experiencing quite a strong swing in the ridings where they are running candidates. See that as a redistribution of the Green vote. However, this is contingent on the Green being at 10-12% provincially. The lack of a candidate in many ridings could well hurt them there.

May 1st 2013: New model and simulator

May 1st 2013: New model and simulator
I just uploaded the latest version of the simulator. Changes take care of the various ridings where the Green or the Conservatives don't have candidates, as well as 1-2 more independents candidates. I also slighlty changed the regional coefficients for the BC Conservatives as to reflect the fact that their support seems more uniform across the province than expected (by only looking at the 2009 results).

Whenever there is no Green candidate, the model assumes that the BC NDP will get 50% of the votes the Green candidate would have got, while the remaining 50% is simply lost. It's a complete assumption but one that makes sense, at least to me. Indeed, the Green Party voters have a lot in common with the NDP, but not 100% so that some voters will simply not cast a ballot because of the lack of a Green candidate. The same logic applies whenever the BC Conservatives aren't running a candidate, except that the votes then transfer to the BC Liberals (well, 50% of them). Polls generally show that the increased support for the Green comes from the NDP while the swing for the Conservatives is done at the expense of the Liberals. 

Most of the time, these changes don't matter, but you have the usual exception. For instance, the lack of a Conservative candidate in Boundary-Similkameen greatly benefits the Liberals. This riding was indeed one of the few the Conservatives could have won (given their current level of support and the 2009 results). So the controversy of last week which led to the withdrawal of the Conservatives candidate is a huge blow to this party.

Finally, I added the bonus to Green candidate Andrew Weaver. The bonus is 15-points and is taken evenly from the NDP and Liberals. I talked about this in previous posts. It's reasonable to assume that this candidate will get some kind of bonus/boost given his notoriety, but it's really difficult to estimate the magnitude of this boost. For now, I'm being generous and giving 15-points, but I reserve the right to change that.

You can try the new simulator here.

I'll incorporte these changes into the simulations model and publish new projections soon.