BBC make it up as they go along. I understand in past years Round 1 for three is the same as ordinary Saturdays: judges + phones. Round 2 for two is pure phones.

The poll doesn't look like a reliable indicator of actual votes cast to me. I'm convinced Gavin is getting far more than 4% of votes.

I thought the judges marked, but the marks didn't make any difference.



The thing is, I don't care all that much who wins (though I'd like it to be Kara). I do care who gets to dance a showdance, because I want to see all the dances from the best couples, and especially from my favourite couple -- so the first round is the more important one for me.

If the judges are inclined to favour Pam, ... hmm, ... maybe I want them out of it.

I think for this poll to be scientifically accurate it has to be based on people who are voting on Strictly this series and not just people who watch the show.

Apparently all 3 couples will do the showdance regardless of elimination, but the eliminated couple may dance in a different frame of mind?

The rules on BBC website are as always a masterpiece of ambgiuity. The last precedent was 2 years ago, with Tom, Rachel, Lisa. On that occasion unquestionably judges marks were aggregated with marks from the previous, controversial, Saturday. This judges aggregate was then aggregated with phone vote rankings. But this was a one-off due to the 3-3-1 fiasco.

On Saturday the Final comprises of two rounds with an elimination after round one. For my money then only Round 2 is the true Final. Anybody with info please let us know, whether marks do count in Round 1. It would seem pretty pointless to have two rounds, if both are decided on phone votes only, with judges' marks for reference. By this stage many fans are committed supporters, they would vote the same both times unless there came a meltdown in the Showdance.

If you can organise a reliable poll the bookies will want to pay you tens of thousands of pounds for the info.

You wonder why the bookies don't organise one then or is that illegal?

Could it be that the GBP are actually voting on the dancing? Surely not.

Yes, and take into account how many times such people actually vote.

Nothing illegal about organising a private poll - Tories and Labour both organised their own polls. For the sampling to be representative the organiser will need to know EVERYTHING about the demographics of voters, who will be watching Strictly and what will be on ITV, who multi-votes how many times, who indeed will be in and not Xmas shopping on the 18th. Pollsters with money no object cannot even get the General Election right.

Apparently both Kara and Matt are available at 5/4, Pamela at 6/1. So anybody who knows the outcome for certain , can be 125% richer.

The 2 professionally conducted X Factor polls, YouGov and ICM , although not perfect, were very good.
Matt winning, Rebecca 2nd and One Direction a poor 3rd.
It was closer in the end between Matt and Rebecca than these 2 polls had said, but that can be put down to Matt underperforming due to his voice being worn out.

It looks like we're heading for an exciting SCD final with the 3 finalists all having a strong support base.

It's anyones Glitterball this year

Go Pamela

wow whomever the judges put last could end up winning it!

I wonder where gavin and scotts votes will go?

Bookies odds can be a very poor guide, as we saw last night with XF.
They had One Direction as a very strong 2nd favourite and even favourite at one time.
The reality is that they were never anywhere near the top of vote, at any time and were never in a position to win.

I suspect the odds are wrong with SCD too.

I'd have Matt as a clear favourite as he should pick up alot of the Scott and Gavin vote. With Pam & Kara not very far behind in joint second spot.

No, that's just wrong.

If it's an ICM poll, it's a poll which has been conducted by a professional organisation. It's not just some random poll on an internet forum; it'll have controls and presumably a sampling methodology.

That's not to say the DS poll is irrelevant - they're both data points.

And, of course, there's no guarantee that "popularity" = "votes", although there does appear to be a strong correlation between the two.

So, if this poll is correct, it means that there's a genuine three-horse race <insert Carthorse joke here>, which I think is a first for a Strictly final, at least for a while.

... yeah but any poll is biased as the samplng will be restricted to those who haven't hung up or run away from those poors sods with the clipboards.

I'm trying to imagine the 'control' group for Strictly fans and have gone cross-eyed with the effort.

I think they tend to use email these days. I know YouGov do. I'm not sure about ICM.
The full results still aren't up on their site yet.

If ICM selected a sample to accurately reflect UK demographics then it will be far out. Regardless of who sits watching Strictly Final, there is a long tradition accepting that those who actually pick up the phone and VOTE is overwhelmingly female, some say 90%. The DS SCD forum reflecting this gender ratio would therefore be a far better nationwide sample than ICM. Not sure if X Factor electorate shows the same gender ratio as Strictly. This gender ratio has never been scientifically proven, and I have never heard of an SCD exit poll.

However ICM did accurately predict the Ann vote where DS poll was far far out, so the signposts are mixed. Possibly this was a one-off aberration with Ann voters being not the traditional DS Strictly electorate, voters who gave up leaderless after Ann left? The Ann phenomenon still remains unexplained or unexplainable. Few admitted to voting for Ann whereas huge numbers went public in both DS poll and forum defending their votes for John Sergeant and Kristina.

Bookies odds are a very good guide. Not of the chances of winning, but of where the money is being placed. The chance of winning is mostly irrelevant; the art of bookmaking is to take enough money from those backing the losers to pay the winners plus a percentage for themselves.

The initial offering of odds is based on their prediction of who will be backed which has some relation to the expectations of the betting public, hence "mostly" irrelevant but not completely. Once the money starts being staked, the odds are adjusted to attract more or fewer bets according to the growing liability to pay out for any individual.

Why odds do make good guide is that anyone staking money is making a considered choice. Much closer to phone voting which is going to take (a little) time and money rather than answering a polling question where choosing any individual has no cost over any other.

The issue with this (and indeed any) poll is that we don't know who was asked. Was it just random people off the streeet or people voting in an Internet poll and were they asked if they were actual voters?

The key is, the numbers of people who actually vote is pretty low. X Factor have just released their voting figures for the series as being between 15-16 million. Not high really considering it's audience is around 15 million a week. That averages out at about 1.5million votes per week (less some weeks as more vote in the later rounds and final) which is about 10% of the viewing public. I'm guessing it's similar, if not lower, for Strictly. Any random person can give an opinion or choose a favourite, but do they actually vote?

Then, of course, there's the issue of which fans are more likely to vote than others. Do younger fans vote more? Older housewives? There are just too many variables to give a completely accurate picture. Even the DS poll can't be 100% accurate as you can only vote once on it.

I think I'm basically saying, you can take most polls with a pinch of salt quite frankly. Let's just wait and see what happens on Saturday.

The latest DS Favourite poll started this morning, gives 65% preference to Kara, 3 times as much as Matt.

That is an overwhelming ratio, not seen since Alesha's Final against Matt di A who had widespread antipathy because of the way he was rescued after his meltdown quarter-final. Alesha was then quoted by bookies at 1/3 then, a far cry from Kara today at 5/4.

Many signposts, but they do not agree.

The multi-vote issue is why I think the NOTW poll is not that useful. Unless we assume that multivoters are distributed evenly across fans of each contestant, and I don't think there's any reason to make that assumption.

The full ICM results are up now.
http://www.icmresearch.co.uk/pdfs/20...cing_poll2.pdf


The question was who you'd vote for, so nothing wrong with that.

Also Pam a steep increase in support as you go through the age groups . Kara a small decline. Matt consistent.

The question in the poll itself leads to ambiguity. 'IF you were going to vote for a celebrity on Strictly Come Dancing this weekend, which of the following celebrities would you vote for?' Doesn't mean they will though does it? Perhaps they should have asked will you vote and if so who for?

Thanks Riverman, that makes interesting reading. It really does seem to be neck and neck.

I would imagine a larger percentage would vote on the final, so as has always been assumed it will rest on the show dance and as each will bring a different performance element, the glitterball could go to any of them.

I'm not surprised that Pamela's vote rises as the age range increases. As an unfit woman in my 40s who loves to dance, I do find her inspirational in a way that Kara just isn't, despite the fact that I can see full well that Kara is the better dancer. I also remember her from 'Not The Nine O'Clock News' so she seems much more like a genuine celebrity to me rather than just Billy Connolly's wife.

Any discrepancy between the DS poll and the public vote can be almost totally explained I think by any age differential between DS forumites and the voting public. No one close to her age has been in the final before, so she's probably attracting a whole slew of votes from a demographic who don't play on DS and who haven't been incentivised to vote so much in previous finals.

It's going to be interesting.

Oh and I can tell you where at least part of the Ann vote came from. Despite my advancing years I am the mother of a five year old who LOVED Ann with a passion because she reminded her of her grandmother and who found her and Anton side-splittingly hilarious. Watching her sheer joy during her dances and her bitter tears when Ann was eventually defeated would have definitely incentivised me to throw a few votes her way if I lived in the UK. I suspect the 'kid' vote for Ann was huge.