Wednesday, July 15, 2009

Poll Position: Conservatives

Three months ago, I analysed how the Conservatives were doing in the polls. It's time to return to this topic, and see how the Tories have been doing since April.

On April 8, one day before I wrote my analysis of the Conservative standing, I was projecting the party to win 133 seats and have 35.5% national support. Since then, they've lost 15 seats and 3.1 percentage points - an important drop. They've gone from a comfortable minority to having only one more seat than the Official Opposition; that is, if the Conservatives even form government.

Nationally, the Tories have been very consistent in their polling levels. The variation is only between 29% and 35%. Because the Conservatives have been within these two levels for virtually the entire period, it is difficult to see a trend. But the charts do show a very gradual incline in Conservative support levels, which bodes well for them in the future. Another positive factor for them is that the Conservatives haven't polled below 30% since the first week of May. They are starting to show some 34s and have been ahead of the Liberals in two of the last five polls.

In British Columbia, things have taken a turn for the worse for the Tories. Before April, they had a comfortable lead over the Liberals and the NDP, but since April they've been mingling with these two parties. They've varied between 26% and 45%, and since April the Conservatives have placed second or third in six polls out of eighteen, a significant change. Lately the Tories have dropped below the 40% level, something they last polled in the first week of June. They've dropped from a projected 24 seats and 41.5% in April to 20 seats and 36.4%. The Pacific province is becoming a problem for the party.

Alberta has remained solid, however. Since April the party has varied between 50% and 63%, and has stuck very close to the 60% level. In April, they were projected to win all 28 seats and 64.2% of the vote. Today, they are projected to win 26 seats and 60.2%. A small drop, but nothing catastrophic.

The Prairies have also remained stable for the Conservatives, but there is a slight trend downwards. They've varied between 32% and 59%, a very large variation indicative of the small polling sizes in this region. They only placed second once since April (on April 13), and have maintained themselves in the 40s, a healthy spot for them. Their projection over the last three months has changed from 21 seats and 48.7% to 21 seats and 46.8% - virtually no change at all.

In my analysis I targeted Ontario as a region the Conservatives needed to make some gains. From between April and mid-June, the Tories were stagnating in the mid-30s, and the Liberals opened up a significant lead. Since then, the party has varied between 32% and 43%, a large variation but the first time the party has polled in the 40s since before April. They're now competitive, and have placed first or tied for first in three of the last five polls. There has been a change in this province, and the Liberals and Conservatives seem to be in a dead heat. Nevertheless, they've seen a significant drop in the projection from 45 seats and 37.4% to 39 seats and 34.8%.

I also identified Quebec as a region the Tories needed to "rebuild their burnt bridges". They've actually been somewhat successful, and have lifted themselves up from the low teens (and even four times polling before 10%) to the mid-to-high teens. They still haven't polled higher than 20%, but are starting to show some 15s, 16s, and 18s, better than the 10s, 12s, and 13s they were polling the past. However, the trend is small, since the party was polling now and then in the mid-to-high teens prior to early June, when things started turning. So, conclusions can't be drawn yet. They've lost some ground in the projection. In April they were projected to win six seats and 17.2% of the vote, but now they are projected to win only four seats and 14.6%.

Finally, in Atlantic Canada the Conservatives are seeing some of their worst polling results. Since April the Tories have dropped and have been surpassed on numerous occasions (seven times out of 19 polls) by the NDP for second spot. This trend has become even more prevalent recently, where the Tories have been beaten by the NDP in five consecutive polls. The real break came in mid-June, when the Conservatives started consistently polling in the 20s. The projection, however, is only starting to move. In April the Conservatives were projected to win eight seats and 30.6% of the vote, while today they are projected to win eight seats and 28.9%. Not a big movement - yet.

The North is not projected with the use of direct polling. No polling firm polls the north. My model uses past historical results and the change in national support from the last election to make a projection. Demonstrating the drop in Conservative support nation wide, the party has dropped one seat and one point in the North.

Since April, the Conservatives have lost some serious ground. But they have had a few successes. They've managed to stop the slide in Quebec and are beginning to look towards growth. But if they party can't reach the 20% level, it will be difficult for them to compete in all but a few ridings. Considering the need to improve their support level by about a third in the province, it will be difficult to make such a leap. The party has also begun to compete with the Liberals in Ontario, and have erased the stable and significant gap between them and Michael Ignatieff's party.

But there are some other areas of considerable concern for the governing party. Stephen Harper has to do something to separate himself from the others in British Columbia if he wants to have a hope of keeping himself in power. The slide in Atlantic Canada should also be worrisome for the party. Having to fight for seats in the East (Ontario, Quebec, and Atlantic Canada) is nothing new for the party, but it looks like the fight will be the most difficult for the party since 2004. But now that British Columbia is opening up as a Second Front, Harper is beginning to find himself in a two-front war.