Ontario 2014 (June 12th)

[Hatman 🍁]

Careful with the subsamples, comrades. That way madness lies.

More accurate than applying swings province-wide.

Another problem with Brendan's map is he is using one poll. You need to do regional polling aggregates from multiple polling companies to get good numbers. For example, the EKOS poll gave me some lol-worthy results such as the Liberals winning Haliburton-Kawartha Lakes-Brock (the ghost of John Tory's failures lives).

[Hatman 🍁]

I guess the media craves "game changing" moments. Let's see if it has traction.

[Hatman 🍁]

It's true the old Liberal base in Timmins is shifting to the Tories. I guess you can call it the anti-NDP vote.

And on the whole, this thread demonstrates why I, personally, would rather steer clear of raw projection-type methodology, at least when it comes to Ontario--it works better in the UK and US, where political party demos and patterns have much more of a set pattern.  I like when there's wiggle-room for chance and miracles...

Oh, don't ruin the fun! Ontario has been very predictable in recent elections (save the 2011 federal election). Some ridings due swing with the flow, but some ridings have strong incumbents or strong candidacies that throw a wrench in any model. The *fun* is trying to pick up on those strengths. I hope to do that with my model.

[Hatman 🍁]

What are you all actually doing, mathematically, to get these projections?

I use a ratio projection on a regional basis, but altering results in ridings that have had by-elections (taking into account polling averages around the by-elections). Unlike Brendan, I will also be doing some riding level tweaks to get more believable results. Tweaks will be based on math as well though.

But what is the actual formula?

Or are you deliberately keeping the details secret, so that it can't be replicated, like a for-profit pollster?

No, I should be transparent for conflict of interest reasons (as I may or may not have access to internal polls that I cannot report on in my model)

In the ratio model, if a party doubles their support in a region, then they double their support in each of the ridings. It definitely has its flaws (which is why the Greens can often be shown to "win" seats), but I have seen it used in the polling industry (not by Forum Research obviously). I first started using it when I was running a Canadian election game in the Individual politics sub forum here.

[Hatman 🍁]

Yes, even 308 will have better numbers than Forum

[Hatman 🍁]

My Week #1 projection (same numbers as I reported on yesterday):

http://canadianelectionatlas.blogspot.ca/2014/05/2014-ontario-election-projection-week-1.html

[Hatman 🍁]

You're right DL, Wilf made the same comment on my blog. I will call it plurality next time.

[Hatman 🍁]

What are you all actually doing, mathematically, to get these projections?

I use a ratio projection on a regional basis, but altering results in ridings that have had by-elections (taking into account polling averages around the by-elections). Unlike Brendan, I will also be doing some riding level tweaks to get more believable results. Tweaks will be based on math as well though.

But what is the actual formula?

Or are you deliberately keeping the details secret, so that it can't be replicated, like a for-profit pollster?

No, I should be transparent for conflict of interest reasons (as I may or may not have access to internal polls that I cannot report on in my model)

In the ratio model, if a party doubles their support in a region, then they double their support in each of the ridings. It definitely has its flaws (which is why the Greens can often be shown to "win" seats), but I have seen it used in the polling industry (not by Forum Research obviously). I first started using it when I was running a Canadian election game in the Individual politics sub forum here.

Wait, sorry: upon further reflection, I actually still don't get it.

Suppose a party had 20% at the last election in a region and 60% in one riding in the region. Then according to the bold, if the party's support doubles to 40% in the region, they double in the riding to 120%? That can't be right.

You're right, it isn't quite like that. Their raw votes would double, so their percentage would be compared to the raw votes for all the parties in that riding. This would heavily skew things, as all of a sudden turnout in that riding would be huge (though, the end result could look realistic).

[Hatman 🍁]

Well, there's 107 seats, so I guess 2 per seat? Not sure why there needs to be 2.

[Hatman 🍁]

Straight from the CBC Ontario Votes website, the Ontario Riding Finder:



Anyone else recognize the map?

I noticed this already, it's... quite amusing. Guess you can blame the cutbacks?

[Hatman 🍁]

Straight from the CBC Ontario Votes website, the Ontario Riding Finder:



Anyone else recognize the map?

I noticed this already, it's... quite amusing. Guess you can blame the cutbacks?

It shouldn't be that expensive to copy paste stuff from their 2011 Ontario map

Referring to the lack of quality control resources.

[Hatman 🍁]

Love how they appointed a man as Minister of women's issues. A bit like appointing a White person to Indian Affairs... oh wait.

[Hatman 🍁]

Why is the NDP all of a sudden popular in Southwestern Ontario? Outside of Windsor, it's not exactly traditional NDP territory. Perhaps it's the economy?

[Hatman 🍁]

Do the dippers have a chance of winning this outright? Or minority government?

Unlikely given polling numbers / memories of 1990

[Hatman 🍁]

Mine were:

PC 65
GRN 46
LIB 43
NDP 39

Although I have the Liberal colour in my name, that's because when I signed up I mostly commented on US results where I would be a Democrat. In Canada I consider myself centre-right. The Liberals under Wynne have swung so far left they are basically in the same spot as the NDP on the spectrum. I am not quite as right wing as Hudak, but I believe in balance on the spectrum so if you swing too far one you need to balance it out the other way. The reason I only got 65 PC as I agree the deficit as a major issue and requires big spending cuts, but I unlike Hudak I am opposed to tax cuts prior to balancing the budget.

Federally I wouldn't vote Conservative though not because Harper is too right wing, but he's too autocratic. I wouldn't vote Liberal either as Justin Trudeau is a complete lightweight and is only leader because of his looks and father. I will probably vote Independent next federal election.

Independent? You better hope there's going to be an independent running.

[Hatman 🍁]

Mine were:

PC 65
GRN 46
LIB 43
NDP 39

Although I have the Liberal colour in my name, that's because when I signed up I mostly commented on US results where I would be a Democrat. In Canada I consider myself centre-right. The Liberals under Wynne have swung so far left they are basically in the same spot as the NDP on the spectrum. I am not quite as right wing as Hudak, but I believe in balance on the spectrum so if you swing too far one you need to balance it out the other way. The reason I only got 65 PC as I agree the deficit as a major issue and requires big spending cuts, but I unlike Hudak I am opposed to tax cuts prior to balancing the budget.

Federally I wouldn't vote Conservative though not because Harper is too right wing, but he's too autocratic. I wouldn't vote Liberal either as Justin Trudeau is a complete lightweight and is only leader because of his looks and father. I will probably vote Independent next federal election.

Independent? You better hope there's going to be an independent running.

That's federally, provincially I am voting PC, but federally I will go Independent or a third party. The Libertarian party always puts up a candidate my riding so although not nearly as ideological as them, that's probably who I will go for. I am though hoping Harper resigns and then I might vote Tory. I might also just vote for whomever is the best local candidate so definitely not Liberal since it's Adam Vaughan who I cannot stand.

Cool. I always like it when people vote for small parties. Luckily the NDP is close enough to me ideologically that I haven't really considered voting for any other party. Municipally I have voted for fringe candidates though.

[Hatman 🍁]

But have you ever voted for John C. Turmel?  I have.

I've had just one opportunity to do so (last August) and did not. I would've voted for him had I been old enough in the 1997 Ottawa-Carleton Regional Chair election. Alas, I was only 11.

My results:

NDP: 73
Grn: 68
Lib: 62
PC: 41

[Hatman 🍁]

Is it correct to assume that although pollsters are weighting by the regions that they publish, they are not further weighting within these regions? So a southwest sample could have, say, Windsor overrepresented one time and then rural areas overrepresented the next time?

I doubt it. If there's any regional weighting, it's likely based on their reported regions.

[Hatman 🍁]

Thanks for your insight, Al.

[Hatman 🍁]

Signs aren't legally allowed to go up until the 12th I think, but that won't stop campaigns from putting them up this weekend. A few people have recycled signs up from the August by-election (same three candidates). Including me, but I never took mine out of my window; no one can see it here on the 15th floor, facing away from the street.

[Hatman 🍁]

+1000, the Star's being itself as usual. More examples: Harper allegedly "smirking" in their private meeting on pensions... with Wynne herself as the source.

And people think Sun is bad

It is. Us poor dippers have no newspaper to read

[Hatman 🍁]

There is the Georgia Straight in BC. Besides the TorStar is more about defeating the PC's so if the NDP pulls ahead of the Liberals they will change their tune quickly. They did endorse the NDP last federal election.

I know; I went out and bought the newspaper for a keepsake.

[Hatman 🍁]

Some riding polls conducted by Oracle:

Sudbury:
NDP: 31.8
Lib: 26
PC: 15.2
Grn: 2.5
Undecided: 24.4

Sault Ste. Marie:
Lib: 27.5
NDP: 20.3
PC: 11.6
Grn: 4.3
Undecided: 36.2

No surprises. Cue Al with his concerns about local polling blah blah blah

[Hatman 🍁]

Some riding polls conducted by Oracle:

Sudbury:
NDP: 31.8
Lib: 26
PC: 15.2
Grn: 2.5
Undecided: 24.4

Sault Ste. Marie:
Lib: 27.5
NDP: 20.3
PC: 11.6
Grn: 4.3
Undecided: 36.2

No surprises. Cue Al with his concerns about local polling blah blah blah

im told its hard to poll rural route addreses

Does Sudbury and SSM have many rural people? Both are mostly urban.

Not really, no. One problem with rural areas is that rural postal codes often span multiple ridings making it hard to determine which riding they're in.

[Hatman 🍁]

The Star has now become a self parody: http://www.thestar.com/opinion/commentary/2014/05/08/andrea_horwaths_rightwing_populism_salutin.html