Tories' CPP Voluntary Contribitions - Good Politics, Good Policy

cybercoma · June 2, 2015

Based on the latest analysis from threehundredeight.

When calculating the averages, we end up with this:

http://www.threehundredeight.com/

Did you even read your link?

"The projection still lags the polls a little, in part because May has been a quiet month (10 polls had been done in April, while we're up to seven for May). Polls that show a three-way race currently take up 61% of the projection. But the NDP has averaged 29.5% in those polls, whereas they have averaged just around 23% in the remaining polls taking up 39% of the weight."

All of the recent polls show a three-way tie. It's not some anomaly when several pollsters are reporting the same thing. The three-way tie is where the electorate stands today. The aggregate is including polls before the tie had been recorded.

Smallc · June 2, 2015

Did you even read your link?

There's a reason a rolling average is used. If you weren't being partisan you'd admit that.

cybercoma · June 3, 2015

There's a reason a rolling average is used. If you weren't being partisan you'd admit that.

Of course there's a reason, but you're completely ignoring the trend and acting as if the rolling average would be the reality if people went to the polls today. The last several polls are the current reality.

Smallc · June 3, 2015

Of course there's a reason, but you're completely ignoring the trend and acting as if the rolling average would be the reality if people went to the polls today. The last several polls are the current reality.

All of those polls have the Conservatives either in the lead or tied for it in the last few days.

cybercoma · June 4, 2015

All of those polls have the Conservatives either in the lead or tied for it in the last few days.

Whatever you say. Evidently, I should just add statistics to the list of things you don't seem to understand.

msj · June 4, 2015

Whatever you say. Evidently, I should just add statistics to the list of things you don't seem to understand.

What an odd ad hominem.

One does not need be a statician to follow the link to see that smallc's statement is backed up by the recent post.

Why not focus on the poll results rather than attack the person making the claim?

cybercoma · June 4, 2015

One also doesn't need to be a statistician to understand how 308 and Nanos's lag works.

msj · June 4, 2015

Then explain it and stop the ad hominem BS.

cybercoma · June 5, 2015

Then explain it and stop the ad hominem BS.

I already explained it. There were less surveys in May, leaving more older and now outdated surveys, given the immense shift in such a short time, in the rolling average. There's also the issue of Nanos having "stickier" surveys that don't shift as easily. In other words, Nanos uses a similar technique to Grenier, which results in twice weighting to older results. The current surveys show a statistical tie. It wasn't a blip or an anomaly, as several different pollsters reported the same statistical tie. There's no way to say the CPC are currently in the lead, unless you give more weight to older, outdated surveys.

Regardless, it's too early to say anything about the election. This is probably just glow from the Alberta election and we'll likely see two more major shifts before people go to the polls anyway. But as it stands today in this moment, there is a three-way statistical tie.

msj · June 5, 2015

If you go back and carefully read what smallc has actually posted you will see that what he has said is all correct.

You are the one going way out to try and prove the CPC are not tied for the lead in polls and do not have the most seats projected (130 at that time) by the website smallc linked to.

IOW, whatever the heck you are reading into smallc's claim is just not there.

cybercoma · June 5, 2015

Where in what I've said do you read that I said the CPC are not tied for the lead? I'm explaining the problem with the seat projections and emphasizing what the current polling has stated. And that's a three way statistical tie, which I've said over and over. Smallc is the one who keeps saying the CPC are in the lead and that's just not the case at the moment.

Topaz · June 5, 2015

Today, the strategist for the Tories said the Tories could win the election by saying over and over again, we can keep Canadian safer than the opposition parties. Now, if one looks at this, Harper is one of reasons so many people are becoming terrorists in Canada, HE'S the guy who sent over troops to the ME, he's the one, the one guy tried to get to on Parliament Hill. He didn't go after Canadians , in his mind he went after the source. I can't see Harper returning to power when 60-65% want him gone before he does anyone damage to Canadians lives.

Smallc · June 5, 2015

I don't want him gone (given the alternatives). At the moment, according to the polls, just enough other people feel the same way (given that the riding distribution favours the Conservatives).

msj · June 6, 2015

Where in what I've said do you read that I said the CPC are not tied for the lead? I'm explaining the problem with the seat projections and emphasizing what the current polling has stated. And that's a three way statistical tie, which I've said over and over. Smallc is the one who keeps saying the CPC are in the lead and that's just not the case at the moment.

I'm going to go with 308's methodology and go with smallc mentioning that the CPC were leading the seat projection at 130 seats over whatever back pedalling you are doing here.

I suspect that smallc, like many of us, think that 130 seats is closer to a "lead" than, say, 107 seats projected for the next party.

And that is the case at the moment.

cybercoma · June 6, 2015

I'm going to go with 308's methodology and go with smallc mentioning that the CPC were leading the seat projection at 130 seats over whatever back pedalling you are doing here.

I suspect that smallc, like many of us, think that 130 seats is closer to a "lead" than, say, 107 seats projected for the next party.

And that is the case at the moment.

Do you understand 308's methodology and why it's projection is behind right now? It still includes polls from as far back as April because less polls were conducted in May. It also includes a Nanos poll which includes its own lag feature. The 308 projection is currently outdated for those reasons. This isn't always the case. The problem right now is that there was a dramatic shift following the Alberta election. Every poll since then has a statistical tie and the most recent numbers are showing the NDP in the lead. Stay tuned for those. You'll enjoy that. Edited June 6, 2015 by cybercoma

cybercoma · June 6, 2015

I still suspect it won't last though. The NDP is getting a post-election bump from Alberta that's too soon to make a difference in the fall.

msj · June 6, 2015

I still suspect it won't last though. The NDP is getting a post-election bump from Alberta that's too soon to make a difference in the fall.

Nice covering yourself there.

Given that I'm a one issue guy in BC who hates the NG pipeline I would be ok with the NDP.

It does not change the fact that what smallc has stated is fact compared to what you have stated is opinion.

Yes we will see the polls change this way and that but let's comment on them after they have changed rather than add our own partisan conjecture to it like you do.

cybercoma · June 6, 2015

Nice covering yourself there.

Given that I'm a one issue guy in BC who hates the NG pipeline I would be ok with the NDP.

It does not change the fact that what smallc has stated is fact compared to what you have stated is opinion.

Yes we will see the polls change this way and that but let's comment on them after they have changed rather than add our own partisan conjecture to it like you do.

What smallc said is not a fact. The fact is 308 makes a projection and has a range of values. He's relying on the middle values. The FACT is that that the 308 projections are based on an average of polls. The most recent polls have a three way tie and the new EKOS poll actually has the NDP in the lead. That's where the country currently stands on the parties. There is no statistical support to say that the Tories are in the lead. They're either tied or behind. I don't know how many times I have to explain that the seat projections you're referring to are behind because they include outdated polling data from April. Those are the facts that you're ignoring when you say the CPC is in the lead because 308 seat projections have their middle values ahead of the NDP.

Smallc · June 6, 2015

http://3.bp.blogspot.com/-k6WbUPuXkA4/VXIOAYeusXI/AAAAAAAAWrc/yTb4FHuR-a8/s1600/Polls.png

The Conservatives are ahead in all but....2 polls. You're wrong.

Edited June 6, 2015 by Smallc

cybercoma · June 6, 2015

You have polls there going back to March. Are you even reading what I'm writing?

Smallc · June 6, 2015

Did you bother to look at the weighting and the actual numbers showing the projected vote share in each poll?

Smallc · June 6, 2015

The NDP has only been ahead of the Conservatives in 2 polls! There's nothing wrong with the model, based on it's accuracy. There's a problem with your partisan analysis.

cybercoma · June 6, 2015

I never said the NDP was ahead. Christ. Nevermind. You're the one claiming the Tories are ahead, which is completely wrong. Should I go back to 2003 polls and say the Liberals are ahead in the polls too? That's how stupid your argument is right now.

Smallc · June 6, 2015

Though within the statistical margin, the numbers show that using all recent polls (ignore the old ones if it makes you happy) the Conservatives hold a slight lead. Further, it is widely believed that the new riding distribution greatly favours the Tories. There's a 56% chance that the Conservatives would, if the vote were held today, fall somewhere above the medium number and below the high number, giving them in the range of 140 seats.

cybercoma · June 6, 2015

Though within the statistical margin, the numbers show that using all recent polls (ignore the old ones if it makes you happy) the Conservatives hold a slight lead.

Like I said earlier. I'm just going to go on the assumption that you don't understand statistics which is clear from this: "Within the statistical margin the Conservatives hold a lead." No. They don't hold anything. They hold a tie. There is NO difference between the scores when it falls within the statistical margin. Every single poll in the month of May was within the statistical margin and the recent poll from EKOS has the NDP in the lead. Your analysis is faulty because it relies on projections with built in lag that includes surveys from April and March. It is now June. Edited June 6, 2015 by cybercoma