Yes, but the AI is based on people who are watching the episode by choice. They're watching it because they already like it.

So an episode of Dad's Army (although old) will have been watched by people who find it funny and the AI will reflect their enjoyment of the show.

If you forced 5000 people to watch it who normally wouldn't, the AI would be much much lower.

Anything scoring 85 or over is in the 'excellent' category.

Oh I see your point.

I'm surprised by how low the AI scores have been for this, consistent 84s and 85s. I know that's still excellent but I'm just a touch surprised that Hide and Journey haven't scored higher...

This is why AI is generally higher these days than the sixties and seventies. Most people are watching what they choose to watch rather than just what is on.

I love Dads Army and when I watch one of those repeats I enjoy myself and laugh heartily. Even during some episodes where the acting isn't good and the jokes are cheesy. Not that it ever happens of course. Because all Dads Army is brilliant! AI=90 fanboi minimum.

Don't they actually give a score out of 10 and then we're given this averaged out index figure out of 100? So, really, it's dropped by tenths in people's estimations if I understand it correctly.

The thing I enjoy is that despite the Moffat haters insistence that all his episodes are now nowhere near as good as under RTD's stewardship, they still get the HIGHEST AI scores of the series. Yet they still claim he is the one who is alienating the casual viewer! Hilarious.

When you say the highest AI scores, which particular episodes are you referring to??

Well, off the top of my head, The Bells of Saint John has the highest for 7b, 87 I think. I'm pretty sure his episodes are among, if not THE, highest during the three series he's been in charge. I stand to be corrected.

Ah right, my mistake, I thought you meant SM's series AI's were higher than RTD's series ones.

Ah no, sorry if I didn't make myself clear. Also, I was right In Series 7a 'Asylum' was top with 89 and 'Angels' second with 88. Series 6b 'Wedding' was joint top at 86 with a couple of other episodes. Series 6a 'A Good Man' and 'Impossible Astronaut' were joint top with 88, just in front 'Moon' with 87 etc.

Considering that Moffat's Who is so widely criticised, that's a little surprising, but that said, most people tend to hate on individual episodes not written by the Moff. But from a personal perspective I thought all those eps except Good Man were among the best period, consistent 9s and 10s from me so I'm really happy to see them all do well in the AI.

Yes and no.

Yes, in that the difference between an AI of 87 and one of 84 is the difference between an average panellist scores of 8.7/10 and 8.4/10.

No, in that this masks that, since panellists score in whole numbers, and a mid 80s score probably reflects a clumping around 8 - 10 scores, a sufficient number of panellists need to drop their scores by one to make the difference.

Remember that 'Cold War' had a lower post 'on the day' timeshift, but a higher overnight. Possibly reflecting a number of timeshifters watching earlier than they usually do. Which can make a timeshift look like it's dropped when it hasn't.

In the end 'Cold War' had virtually the same consolidated ratings as 'Rings'

'Hide' had a lower timeshift than we've had for a while, but not an unusually low one. A loss in overnights (which may well be primarily down to a loss of casual live viewers, the most easily lost component of the viewership due to competition and weather) is not automatically made up by timeshift. Sometimes it has, sometimes it hasn't, and it isn't always easy to see why.

But, at this point, the lower (though still within the usual range for Doctor Who these days) timeshift for 'Hide' is an exception rather than part of a trend. We'll have to see if that changes.

Certainly hitting the mid 6 millions is entirely normal for a Who series. I have an earlier post in this thread showing just that.

Any update on Timeshifted viewing for Journey to the centre of the TARDIS?

The ratings in the US have gone up again this week.

Sorry about the lack of updates recently guys, JttCotT is currently up to 6.203m

Thanks Yog101

How does that compare to this time last week?

I don't understand why you assume that it's peculiar that something old can be seen to be good.
Why would it's age reflect the quality of it?

If you can get films from the 1930s or older which are better than some films made this year, then why can't it work for television shows too?

It's either about the same or very very slightly up. Probably wont breach 7 million.

I thought so too.
I remember in the earlier years when it was brought back the occasional episode got less than 85, even one or two getting about 79. I think some of the best episodes have had about 84 and around that AI.

Other TV programmes can get extremely low AIs compared to any given episode of Doctor Who.

I believe that there's a range scoresheet somewhere that denotes how good the AI score for a programme is depending on what range its score falls in.
I think that it's something like anything over about 75 is excellent, and anything over say 90, is exceptional.
It goes something like that anyway.

I think 85 is an AI score which people should be very happy about. Many other programmes would envy an AI of 85.

Around 6.5m again. Not too bad...Could be better, but not too bad at all.

It's added more than Cold War had at this time, the figures for Hide at this point weren't posted. I think it will end up around 6.6m, same as Hide despite being 0.1m less in the overnights.

From a ratings perspective, the change in companion has had no effect on the US ratings. Here are the BBC America 9pm overnight ratings averages for S7.1 (the last five Amy & Rory episodes) and the first five episodes of S7.2 (the first five non-Christmas episodes with "modern Clara"):

S7.1 (5 episodes) - 1.13 million / 0.44 18-49 demo
S7.2 (5 episodes) - 1.11 million / 0.46 18-49 demo