January surprise?

Last SES poll shows parties less than six points apart: 36-31 among "very likely" voters. For some reason, SES chose to break out the daily results from their three-day poll, something I don't recall them doing before. On two of the three days, SES shows Libs within three points - again, among "very likely" voters. Note of caution: sample size just 400 each night.

Meanwhile, Ipsos final, with sample size of 11,000, shows 11 point gap, 38-27.

Still waiting for final Strategic Counsel effort. Probably shows NDP ahead, if recent performance any guide...

UPDATE: Strategic Counsel/Globe/CTV puts Tories up by 10, 37-27. Sample size: 2500. Time for a change: 63%. Tory majority "good for country": 46% (down from 55% a week ago). Ontario: Cons and Libs neck and neck at 37-36, but NDP at 22. Most implausible data point: Libs further ahead in 905 than in 416? Second most implausible: three-way tie in BC?

That's Ipsos, Strategic Counsel, Leger, Decima, and Ekos, all with Tories about 10 up, vs SES at 6, gusting to within 2. Conclusion: Nanos isn't just peddling crack -- he's on crack. UPPERDATE: The Globe's report on its poll -- they're no longer burying them, I see -- contains this interesting nugget:

When support for the parties is overlaid on Monday's projected voter turnout of about 62 per cent, said Mr. Gregg, the Conservatives advance another three percentage points. This is because the people most likely to vote are older and more well-off and a significant portion of them will be Conservative supporters... On the basis of the poll results, Mr. Gregg predicted the Conservatives would win in the neighbourhood of 140 seats, the Liberals 75, the New Democrats 35 and the Bloc in the high 50s or low 60s.

Well, close.