NTV and Telelink released the only independent poll of the campaign on Monday and with a week to go in the 2011 general election, things are on track for a historic election.
For starters, let’s look at the Telelink party support numbers:
- PCP 35%
- NDP 15%
- LIB 08%
- UND 42%
These numbers are ones you can trust for accuracy and reliability. In fact, once you read along here and look at 2007 you’ll understand the real reason why CRA and other numbers are pure crap on a cracker.
Next, let’s take a look at the 2007 poll numbers.
In 2007, Telelink’s election poll turned out these numbers:
- PCP 42%
- LIB 08%
- NDP 3.5%
- UND 31.7%
The actual poll results on election night, as a share of eligible voters was:
- PCP 42%
- LIB 13%
- NDP 05%
- DNV 38%
By comparison, CRA’s August 2007 poll (adjusted to show percent of all responses) was:
- PCP 62%
- LIB 14%
- NDP 06%
- UND 18%
All opinion polls in this province survey eligible voters. The polling firms don’t report their figures that way. They make it seem like they are talking about share of popular vote. But if you look at it, they simply talk to anyone eligible to vote. That means you have to compare their poll percentages to the share of eligible votes a party got on voting day.
Incidentally, if you looked at the popular vote numbers and the ones CRA actually reported (as a percentage of decideds) their accuracy doesn’t get any better.
So compared to CRA, Telelink was almost spot on for everything, except the Liberal vote.
And with all that in mind, let’s look at what we can see in 2011.
Record Low Turnout
For starters, we can expect a record low turn-out at the polls beating the previous record set in 2007. Given the way the Telelink numbers compared to actual then, we could be looking at half the population not bothering to get out to vote.
In patronage-addled political cultures like Newfoundland and Labrador, voting is one of the ways people pay the patron back for his benevolence. They also need to turn out to vote to show their continued loyalty to the Boss or to signal their allegiance to a new one.
Not surprisingly, in the 18 elections from 1949 to 2007, turnout was above 69% in all but three. Turnout in 1949 was 95%.
So in years when the turn-out was low, you have to wonder what the heck was on the go. What do 1956, 1966 and 2007 have in common with 2011? One thing they don’t have in common is overwhelming satisfaction with the party in power at the moment.
Tory support high but dropping
Conservative support sits at 35% and that’s likely where it will hold. What’s most interesting over the past four years is the way even CRA polls have picked up a decline in Tory support. When you take out the misleading twist of giving the numbers as a share of decideds, the Tory support has dropped dramatically since early 2010.
The Race for Second Place and Other Bullshit
While plenty of people in 2011 will be talking about the low Liberal number in the Telelink poll, you already know it’s exactly the same number the Liberals turned up the last time out in the Telelink poll.
On voting day in 2007, the actual Liberal share of eligible vote turned out to be almost double that number. And as a share of popular vote, the Liberals turned out three times what turned up in the polls. They held three seats at the end of the night and picked up another one later on.
The polls don’t tell the story and the media reports on those polls sure don’t tell the real story.
You can also see that when you consider that the NDP polling number in 2011 (15% to 18%) is roughly what the party had in the mid-1980s. They had two seats.
The final seat count will depend very much on what happen on the ground this week.
The Conservatives have undoubtedly dumped as many bodies as they can into the seats where they are under pressure. Some of those seats are in St. John’s and others are spread across the island and into Labrador.
If the Liberals and New Democrats learned anything from the last time, they are pushing back hard as well.
There is no race for second place. That’s a fiction invented by the media. The real election race is being played out in those pressured seats – as many as a dozen or so – across the province. The Liberals pose the bigger threat to the Tories because they are potentially viable in more seats. That’s just a function of the electoral math.
What you will see Tory partisans doing is pushing hard on the theme that the Liberal Party support is collapsing. They want people - especially Liberal campaign workers - to become demoralised and stay home.
The Tories are getting plenty of help from Liberal gaffes. They are also getting help from the news media who report fiction – like the race for second place or the importance of debates - as if it was fact.
What is really happening and what the news media often report are two very different things.
And what happens next week after all the voting is done, well that won’t look much like the media projections to date and it may well be a lot less dramatic than people are assuming.
- srbp -
2 comments:
Ed, you said:
You can also see that when you consider that the NDP polling number in 2011 (15% to 18%) is roughly what the party had in the mid-1980s. They had two seats.
Do you mean opinion polls or election results from the mid-80s?
In 1985 for example, when the NDP had its short-lived electoral 'mini-surge', it got 14.4% of the actual vote. Not 14.4% of all eligible voters, because obviously not all eligible voters voted.
The NTV/Telelink poll from Monday shows that 42% of all eligibles are undecided.
So, if you trust that the NTV/Telelink poll shows roughly equivalent results to what will happen on October 11th based on the past accuracy of NTV/Telelink polls, and if we can safely assume that either (a) all of those exact same undecideds will stay home on election day, or (b) those undecideds who do end up voting will vote in exactly the same proportions for each party as the decideds (the other 58%), then the following is true:
- The NDP will not get 15% of the decided vote. It will get 15% of the decided+UND/DNV "vote", which translates into roughly 26% of the decided/actual vote (15% of a group of 58 out of 100 is not 15% any more);
- For the Liberals, 8% of 58% is 14%.
- And for the PCs, 35/58 = 60%
14% for the Liberals is more or less what MQO found; the only main difference then between MQO and NTV/Telelink's findings for decideds is a 6-8% discrepancy between NDP and PC decideds.
Otherwise I agree with everything, especially the bits about how this will translate into seats, although I would say that with Liberal numbers so low, even a highly efficient vote won't help it get those seats it is otherwise capable of getting. But both opposition parties better pray that they are extremely popular in swing districts and extremely unpopular in PC safe seats.
The 15-18% was the actual vote on polling day. I'd have to back and double check those numbers specifically because I got those from another source.
In 2007, the Telelink poll found 3.5% of respondents support the Orange option. On polling day, they captured 5% of eligible vote.
Based on that, I'd suggest that they should get roughly what Telelink found this time as a share of eleigible.
The only problem in all this for all parties is what turns up at the polls.
The one big weakness for the orange team is that they are - if usual patterns hold - short on bodies on the ground. GOTV is not easy without bodies and their overall campaign hasn't had the sort of features to it, including the intensity of advertising that in the federal election took up any slack in places like Quebec.
That said, every party is short of bodies in this election. No one is getting volunteers which measn that they all have problems with GOTV.
How that translates into actual votes on the day remains to be seen. Still, based on past elections, the NDP can capture a higher share of vote measured any way and it won't make as much difference as a modest shift in Liberal vote mostly because of where the vote is.
My seat guess - based on past elections and the current environment - would be something that looks pretty much like it is today. Add one more Dipper (maybe) and trade off one Grit seat in one place for another somewhere else (maybe).
Much depends on whether the Tories could redeploy their resources into areas where the races are tight either way and pull out every last possible vote.
Regardless of all that, a record low turnout means that the two opp parties failed miserably to capitalise on an opportunity no matter what the seat count is.
Post a Comment