There were 5,040 possible outcomes after the judges' scores at the weekend, and by analysing those that put Patsy last and Gav 6th, we can conclude the following:

- Nothing meaningful about Matt, Pamela and Kara. They could have survived by coming bottom with the public and could all have come top with the public.

- Scott didn't finish bottom with the public (neither did Ann or Gavin, but we knew that already)

- The highest Gav finished with the public was 2nd(!) albeit there's only one combination that gives this outcome

- Gav's most likely position with the public was 5th (39%)

- There's a 79% chance that Ann finished 3rd or better with the public

- There's a 6% chance that Ann finished 5th with the public

Rolling forward to next week, there are 720 possible outcomes (excluding judges' ties), but if we assume the public voting patterns are repeated, then we can rule out all thoses where Scott, Gav or Ann come last, Ann second last, Gav top etc. and reduce the number of outcomes to 180 for a given set of judges' scores. (Assume also that Gav isn't finishing 2nd in the phone vote, as this has only a 1 in 433 chance of actually having happened.)

Assume that Gav and Ann finish 5th and 6th with the judges next week - not unreasonable.

Of the resulting 180 outcomes, 18 see Ann going home, but to do this, she would have to finish 4th in the phone vote. So there's a 79% chance that she's not going home next week!

There's exactly a 25% chance she'll be in the bottom two next week though, so the conclusion here is that if she is in the bottom two, don't get too excited, as even if she does, she's most likely not going to be last.

Gav appears to be "toast". The are 98 outcomes that would see him going home, of which 20 involve him finishing 4th or better in the phone vote, so even if he is more popular than the two dancing ladies, he's still got a 20% of getting the "red card".

If Kara finishes 3rd with the judges again, then there are 34 outcomes that see her going home. Likewise if Pam finishes 3rd. These all involve finishing last or last but one with the public, so I've not considered Scott for this fate (as he's definitely not finishing last with the public) or Matt based on anecdotal evidence of his popularity.

The scene is set for Kara and Pam to top the judges' scores, Gav to be marked highly for his VW, maybe beat Scott (if Scott has another zombie session) but still depart, albeit on the all-round satisfying high note other than Ann enduring for another week.

What is amusing is that after a couple of hours of number crunching, we end up concluding what gut instinct would have told us anyway!

Gut feeling says to me - whatever the stats - that Ann won't go until she's properly caused outrage by knocking out a real contender. Since the semi-final is the one with the double elimination, unless she does that this Saturday, she'll be into the final, I reckon.

People who "hate Ann" - as per the histrionics on this forum - should be voting for Gavin like crazy this weekend, if you ask me!

If she's out of the bottom two this week then there could be fireworks!

The judges need to engineer Gav into 3rd place on their scores behind the two least popular "good" dancers, with the two most popular good dancers 4th and 5th. Then, Ann could only avoid the chop by topping the phone vote, even if Gav only comes 4th in the phone vote. If she's doing this, she's going to win whatever happens.

Gav to outscore Matt and Scott. That would be the day that Strictly dies but is immediately resurrected!

But only 432 combos leading to the result we got. Just flagging that up.

"if we assume the public voting patterns are repeated" is the interesting bit, to me. Votes moving from couples eliminated to others, switching of allegiance, previous non-voters pitching in, all of these can throw things out significantly.

Indeed. I should have added that.

This is in danger of planning for WW2 based on WW1. Success of any kind brings new admirers and supporters.

The Anti Ann Alliance failed to make a united stand and was picked off one by one. Only two more chances for tactical voting diverting votes from unbeatable celebs to reinforce beatable celebs.

May be too late.

Well you have to assume something! Assuming patterns are repeated is least unjustifiable at this stage of the game.

FWIW, I'm not sure the potential switch of votes from the eliminated pair will make much difference yet, as there aren't very many of them.

Which is why Craig has to grit his teeth and get the 8 paddle out for Gavin!

Bluntly speaking, 4 judges can collude far more successfully than a notional anti-Ann alliance can vote tactically.

From the published results, all we know is that Ann is more popular than Gavin and Patsy.
Gavin is more popular than Patsy.

Looking forward, that leaves us with Ann is more popular than Gavin.:yawn:

This is what we know about the number of points they could have got from public vote this week:

Kara - 1, 2, 3, 4, 5, 6, 7
Pamela - 1, 2, 3, 4, 5, 6, 7
Matt - 1, 2, 3, 4, 5, 6, 7
Scott - 2, 3, 4, 5, 6, 7
Patsy - 1, 2, 3, 4
Gavin - 2, 3, 4, 5, 6
Ann - 3, 4, 5, 6, 7

I was rather disappointed, I have to say. If Scott been judged below Patsy then Gav would most likely have got the chop and we'd have got some insight into Ann and Scott's relative popularity.

I reckon we'll be down to working it out on our fingers before we know for sure! This is the downside (or upside, I suppose) of getting rid of the D-O. Each week, there's only usually one person whose popularity is known (i.e. the one from the bottom two the previous week) unlike the glory days of Rachel and Lisa where their regular apperances in the D-O flagged up their unpopularity with the masses and it would have taken a bizarre chain of events to get either one of them to win.

Sadly from Rachel's viewpoint, this bizarre chain of events transpired, only to be trumped by an even more bizarre one.

With, crucially, the chances of Ann's being less than 5 quite low and the chance of Gav's being more than 4 low, making Gav's chances of survival unless he finishes higher than 5th very low indeed.

Which is why the analysis that Gavin is toast this week is probably correct. He did beat Scott once, but that was in Scott's rumba week, so it is unlikely to be repeated.

As much as I want to see the back of Widdy (but not her backside any more, thank you very much), I really wouldn't be too bothered if Gavin goes before her, whereas I would be outraged if Kara gets eliminated next. So I won't be voting for Gavin, no matter what, and will, in all likelihood, be voting like mad for Kara.

I guess we'll never get to the bottom of this, but I wonder whether the anti-Ann brigade would favour a blatant stitch up to get Gav high enough up with the judges to save him, thus condemning Ann.

After all, if SCD is primarily a dance competition, should colluding judges be considered an acceptable price to pay to engineer Ann's downfall? Can a "wrong" be considered right, if it helps to right an even bigger "wrong"?

My view is that anyone who would accept such an engineered result is more concerned with getting the result they want rather than the integrity of the competition overall.

Although I wouldn't like to believe in engineered results taking place, last Saturday did make me think otherwise. I definitely knew Alesha wasn't marking according to the performances she was seeing (especially Scott and Matt). What is unfortunate is that she is not articulate enough or possess the dramatic ability to disguise such engineering as the other judges can. Now I definitely feel she, Len and Bruno will mark Scott in order to get him to the finals regardless of the performances (but I hope his performances will be good in order to justify the marks).

This. With only 6 couples left tactical voting is pointless at this stage as they are all vulnerable. Everyone should just vote for their favourite. For me that will also be Kara & Artem but if you are a Matt, Scott or Pamela fan then you should vote for them.

I love these kinds of threads.

I think if the leaderboard stays the same next week, Gavin will be out and Ann will be in the bottom two. That is scary, as the only thing I see being fairly non-static would be a bottom 2 bump for the next week. I don't think it would save her (I think she'll be out at the semi) but she could potentially take anyone out with her.

Off to crunch some more numbers...

I think if you are happy for the judges to engineer their scores to get rid of Anne then you can't complain when they engineer the scores resulting in your favourite going home. IMO engineering scores is wrong on any count (and I want her to go too),

Probably the only safe place for any of the top 4 to be on the leaderboard this week will be in the bottom 2 .... that way they can be sure of loads of votes to save them.

If Gavin & Ann are bottom 2 - as most possibly they will be - then they will get the votes, and one of the top 4 will go.

A very interesting analysis. Of course, even if you could ascertain the voting pattern, it doesn't mean that the same pattern will be repeated next week. Gavin might do an outstanding performance, while one of the big four might have a nightmare (although the blinkered judges will probably still award 9s).

I've a feeling that one of the big four could go this week, if Gavin either dances particularly well and gets among them, or has more votes than the person just above him on the leaderboard.

Personally I'd rather the marks were reasonable- but they've been all over the place a lot so I doubt that will change.

It is however probably reasonable to say that if your main concern is making sure Ann doesn't reach the semis (particularly if the public voting preferences mentioned on another thread are correct and stay the same), your best bet is probably to vote Gavin, unless something very strange happens to the leaderboard. Of course there are no guarantees. If your main concern is keeping your favourite in of course vote for them regardless of leaderboard position.

Personally I'd much rather see Gavin reach the semis than Ann- but I'll be out that night so I won't be voting anyway.

I think (stressing the think part) that would only work if Ann voters switched to Gavin. And I doubt that would happen.

On condition that 6 celebs earn 6 different marks, the top-marked celeb is immediately invulnerable.

If this Saturday Kara scores top mark 6, her minimum aggregate without a single phone vote would still be 7. Would anybody like to try work out a permutation for an aggregate of 7 to end up bottom? No, the top-marked celeb is immediately safe. Falling into second-bottom would do no harm.
In that scenario any vote for Kara would be better deployed voting tactically for Gavin, Pamela and Scott.

wrong post