And the winner is...

So, which pollster had it right?

I recorded the last polls of the election here.

Here they are compared to the results.

Source	Liberal	PC	NDP	Green	Dates	Details
RESULTS	37.6%	35.4%	22.7%	2.9%	Oct 6	THE ELECTION!
Angus Reid	33%	36%	26%	5%	Oct 2-4	Online; Sample: 2223; 2.1% 19/20
Ipsos-Reid	41%	31%	25%		Sep 30-Oct3	Sample 1020; 3.1% 19/20
Abacus	37%	34%	24%	4%	Oct 3-4	Sample 1001; 3.1% 19/20

Abacus has the best call, perhaps because their sample was concentrated closest to election day.

The huge Forum Research Poll, also under that link, didn't report percentages, but did make a seat call of 45/45/17, which was damn close, and nailed the NDP count. Given that that poll Forum conducted the poll over the weekend, days away from the election, the reason for the discrepancy is clear.

Ipsos-Reid was way off, but also was working with the oldest data, and a smaller data set sprawled over many days.

For those who don't know, the final count was 53/37/17, the largest minority possible.

Polls-a-Ticking

Polls, polls, polls. The last ones are out for the Ontario election, and predictions range from a Hudak technical win giving way to a likely Liberal minority, to a Liberal majority.

Don't look at me.

For posterity, here they are:

Source	Liberal	PC	NDP	Green	Dates	Details
Angus Reid	33%	36%	26%	5%	Oct 2-4	Online; Sample: 2223; 2.1% 19/20
Ipsos-Reid	41%	31%	25%		Sep 30-Oct3	Sample 1020; 3.1% 19/20
Abacus	37%	34%	24%	4%	Oct 3-4	Sample 1001; 3.1% 19/20

I'm doubting the Ipsos-Reid poll the most as it's older. The Angus Reid has the best sample size, and contains the most recent data; but it's an online poll, which are still subject to much criticism.

To add to the noise, a huge Toronto Star poll of ridings 'too close to call' is predicting a hung legislature. The poll was conducted by Forum Research.

The Liberals and PC are predicted to get 45 seats each, the NDP 17. Fifty-four seats are needed for a majority.

The poll has a sample size of 23,000, and looked at 23 key ridings where the leader spread was under 5% in the last poll (done two weeks back). The poll was conducted Oct 1 and 2. If you're wondering about all the other ridings, know that most elections involve fighting over a minority of seats, with the remainder being so secure to almost not matter except in rare occasions.

With such a larger swing of numbers over such a short date span (Sept 30-Oct 4), and with the NDP showing strong numbers, many ridings are going to be determined by a sliver of the vote. Such small movements in ridings can't be caught by normal polling, so it seems safe to use the Forum Research poll as the best indicator.

Much like the last federal election, though, voting intentions firming up during the last few days are really going to matter.

This election was Hudak's to lose. I think he has, though McGuinty having a majority is quite uncertain. If McGuinty does get a majority, it looks likely to be with a heavily divided electorate, and I hope he governs accordingly.

Forum Research reports reduced Harper minority, NDP in Stornoway

From The Hill Times, the projection is Conservatives 137 seats, the NDP 108 seats if an election were held today, 60 for the Liberals and only three seats for the Bloc Québécois.

As for individual ridings, Gerard Kennedy is set to lose Parkdale-High Park to the NDP's Peggy Nash, and Beaches-East York as well to the NDP. Everyone's favourite Conservative Gerald Keddy is set to lose to the NDP out in Nova Scotia.

Liberals and Conservatives will lose seats to the NDP in Quebec.

There's even a loss of Conservative support to the NDP out West, though what that will do to the seat count is not clear.

There's still a few days to go, but the Conservative campaign is winding down, correctly expecting less air time for the campaign as hockey and the royal wedding take the final weekend over.

That also means there will be little action to change numbers. With the NDP having momentum, we're looking at the Bloc marginalized, and the NDP and the Liberals creating a majority.

I recommend an accord, but I'm getting ahead of myself.

---
Update:

EKOS has a different take, but not that different. NDP are a bit lower and the Bloc is holding some ground, but the NDP has surged well ahead there. All in all, I think we're waiting to see if NDP+Liberal = majority.

Two simulstaneous polls, two very different results

Proclaims the National Post: Tories 19 points up, NDP tops Liberals for second: Poll !!!

Yikes! Ipsos Reid has the Conservatives at 43%, NDP 24%, Liberals in third at 21%? Can it be?

A second poll by Ekos has the Conservatives at 34.4%, and the NDP and the Liberals tied at 24.7%.

Both were made by random sample phone April 18-20.

Both can't be right!

The answer is (likely) Ekos. Its sample size is 2156, whereas Ipsos Reid is only 1000.

One thing is common here: The NDP is up, way up, while Liberal fortunes are sagging.

Read the links. the regionals on Ekos are interesting.
========
UPDATE:

I thought I'd toss in the Nanos numbers. Nanos is compiled differently than the other two. 900 random calls a night are added to the two previous nights and averaged to get an ongoing rolling average. The most recent shows Con 39%, Libs 26.7%, NDP 22.1%

These numbers cover April 18-20 as well.

Conservative numbers appear to be all over the place, but I've been watching the Nanos poll for weeks, and the Cons have been stalled at 39% for a long time.

hard to say what is going on out there. The only consistent pattern is NDP support rising, and Liberal fortunes flagging.

Iggy's new low

Proving it takes more than scandals to get a lift in the polls:

A new public opinion survey shows support for the Liberals has slipped to 23 per cent – well back of the 38 per cent support for the front-running Conservatives.

Everyone is down, or stable -- except the Greens. I suspect the Greens are currently a refuge for many voters tired of the same-old same-old.