Technically the timeshift would be a lot more than 1.5m, as you need to include VOSDAL. It's a long time since any Doctor Who episode jumped less than 3m from the 'live' figure. In fact it was The God Complex, which had 4.03m live, and jumped to 6.77m in the final count. (The following week, Closing Time had 3.93m live, and had a timeshift of exactly 3m to 6.93m.)

So far, only one episode has more than doubled its live audience. This was Day of the Moon, which had 3.38m live, and a timeshift of 3.92m to a total of 7.30m.

I'd also expect Cold War to jump up noticeably when we get the final figure on Sunday (or Monday). At the moment, with the figures being reported by Yog101, it's being dragged down by three minutes of weather, which is likely not timeshifting much at all from the 3m or so that actually watched it live.

Therefore, as the week goes on, that 3m or so for the first three minutes (between 6pm and 6:03pm) is becoming increasingly significant, and dragging down the overall average. This will be corrected in the final BARB total, when the precise start and end times are taken into account, which will make it appear as though there has been a sudden jump when we get the final total.

Some US ratings for 'Cold War'

Thanks for the info morph1970.

Any chance you could do a simple idiots guide on the time shift? Sort of what it is, how it works, when various figures come out etc.

I'd find it really useful as I have a general idea how the ratings work but would like to have a more detailed understanding.

Well, basically when you see the overnights reported each day, this is generally an average number of viewers across the advertised slot that watched 'live', as well as the viewers who watched a recording of the programme later that day (VOSDAL – Viewers On Same Day As Live), before 2am the following morning. This gives a rough guide to the number of people who watched the programme that day. However, it is still a rough estimate, because – to take a Doctor Who example – it might have been advertised to run between 6pm and 6:45pm, but it might actually have run between 6:03pm and 6:46pm.

The final ratings figure will therefore need to be adjusted for the precise timings of the broadcast, as well as adding in other viewers who watch a recording of the programme, up to exactly 7 days after broadcast. (In the case of Cold War, this will be 6:03pm this coming Saturday.)

At the moment, thanks to Yog101, we are seeing the daily updates for viewers who are watching a recording of Cold War each day. But those updates are still using the 'rough timings' of the broadcast. (Indeed, in the case of commercial channels, they also include all the advert breaks, where the average ratings figure is likely to have dropped, where casual viewers channel-hop.)

When BARB releases the final figures each Monday, all the necessary adjustments have been made – allowing for precise broadcast times, and stripping out the periods of the programme when adverts were shown. (They also work out a precise 'live' figure, which isn't available publically.)

The true timeshift figure is the difference between the two numbers above (ie: the final BARB consolidated rating minus the the final BARB 'live' figure.)

Here are some examples for Doctor Who:

The Impossible Astronaut
LIVE = 4.75m
OVERNIGHT = 6.5m
FINAL = 8.86m
Timeshift = 4.11m (a record for any show at the time, now beaten only by The Snowmen).

Day of the Moon
LIVE = 3.38m
OVERNIGHT = 5.4m
FINAL = 7.30m
Timeshift = 3.92m

....hopefully one day the figures will include iplayer viewings too.

They do include iPlayer viewings watched through a television, but not those on home computer or hand-held device etc.

Thanks for that.

Whatever happened to "live plus 7" figures that were going to arrive late 2011? (ie iplayer + live viewing + timeshift for 7 days after broadcast).

Thank you morph1970 - really useful.

Interesting.... Is that any TV that watches the episode by using I-Player or do you have to be part of the BARB panel?

You'd have to be part of the BARB panel, but the same calculations are made to make it representative of the entire population.

what whheni feel its a taaad ddull ssorry ssso bgt aaand the voicce maaay get beetterr vieweings ssorry ttto daamppen moods , but iplaayer maay help with viewing as people watch it later ax

Cold War Update: 7.083m

Still think it could struggle it's way up to 7.2m. Would be surprised if it got much better.

I think the bad press has affected the views of some casual viewers... But hopefully people will take back up to the next lot of stories... Especially the Neil Gaiman ep as that attracts others Who don't ordinarily watch...

Thanks Yog! Looks like it won't be going much higher. Can we blame Rings for that or Cold War itself? What will happen against tougher competition this Saturday?

Don't forget, it should still increase noticeably in the final figure when the first three minutes of the timeslot are stripped out. (It started at 6:03pm, but these figures haven't yet been adjusted for that.)

Well if you want to try and work it out,1800 to1802 is currently averaged at 5.308m

What's the average from 1800 to 1803 (or is that what you meant?). Actually, do you have minute by minute breakdowns already, Yog? In which case you can see 18:03 to 18:45...?

I didn't include 1803 as that's when the programme starts, right?

1800: 5.145m
1801: 5.348m
1802: 5.433m
1803: 7.068m


Yeah I can get 1803-1845. Was that when it definitely finished, or did it go on past that?

1803-1845 averages 7.163m

-----------------

EDIT: Actually, if I do a minute by minute breakdown of the 1800 clock hour, there is a massive drop off at 1844. So if I get the average of 1803 to 1843 it sits at: 7.243m

Thanks to Yog101 & Morph 1970.... Very interesting and useful information!

Ah, right! So 7.24m sounds more in the right sort of ballpark when the correct timings are adjusted. So it may yet reach 7.3m or even 7.4m with two days of recordings still to be added. Thanks Yog!

That looks better already thanks guys...

Will the figures for those watching the repeat on BBC3 tonight be added in to the final figures?

They're not, but I will be able to get the figures if it's of interest.