The Ultimate Averaged Chart - The BBC Chart Re-Imagined

[Metalweb]

The problem with the NME book seems to be that there's no way to tell what's an error and what's a deliberate change to the original charts!

We may suspect Glow Worm on 10 Jan is simply a mistake but we don't know for sure...

[MrTibbs]

It is! That is the single error I've found in the whole lot. He has even gone away and added in the proper 1960's album charts and made so many corrections over previous stuff. I'm not going to say it's error free (What work is?) but that's literally the only one I know of.

That's reassuring Lonnie, before I commenced the Ultimate Averaged Chart I got really annoyed when discovering chart position errors in books and ranted about 'carelessness'. I know different now. For all the checking I do with each UAC before posting occasional errors slip through the net and thankfully get picked up for correction despite all my best efforts. So I now understand how difficult it is to totally eradicate errors.

I think there is an acceptable level of margin for error in any book containing thousands and thousands of numbers but some publications had so many errors that it was really just sloppy compilation with little checking.

[kingofskiffle]

I'm disappointed to hear that, I accepted Graham's book as the more accurate volume over the NME book.

It is! That is the single error I've found in the whole lot. He has even gone away and added in the proper 1960's album charts and made so many corrections over previous stuff. I'm not going to say it's error free (What work is?) but that's literally the only one I know of.

[MrTibbs]

Actually, the Graham Betts book is wrong here. We did discuss this error I think when creating the book.

I'm disappointed to hear that, I accepted Graham's book as the more accurate volume over the NME book.

[MrTibbs]

The NME book also appears to be wrong for 7th February. Graham Betts' book again lists the correct chart. NME book omitted Nat King Cole - Faith Can Move Mountains from joint #10, and, Eddie Fisher - Everything I Have Is Yours at #8.

And on 20th December 1952 the NME book also omits Kay Starr - Comes Along A Love climbing from #6 to joint #3.

[kingofskiffle]

NME Scans

9 Jan
https://www.dropbox.com/s/jegxhgkd00...20Pop.pdf?dl=0
16 jan
https://www.dropbox.com/s/2w7pqr8tu9...20Pop.pdf?dl=0
23 Jan
https://www.dropbox.com/s/16v97gx40b...20Pop.pdf?dl=0
30 Jan
https://www.dropbox.com/s/8c0fn5fd9e...20Pop.pdf?dl=0

[kingofskiffle]

Graham Betts' book Official Charts And Hits The Fifties may solve this. His chart for 9th January agrees totally with the NME book as you posted above Lonnie. Mills Brothers entering at 10. BUT his chart for 23rd January only agrees up to #9. Then he lists :

10 RE Nat King Cole - Because You're Mine
11 10 Tony Brent - Walkin' To Missouri
12 NE Eddie Fisher - Everything I Have Is Yours
12 NE Tony Brent - Got You On My Mind
12 12 Nat King Cole - Faith Can Move Mountains

So it would appear the NME book is wrong here and The Mills Brothers did indeed chart on 9th January.

Actually, the Graham Betts book is wrong here. We did discuss this error I think when creating the book.

[kingofskiffle]

Actual Graham, the Glow Work/Worm is down to a new keyboard and auto correct.

The record should be on 30 Jan, as shown in scans at the time. If the source of the charts used to make the book was the NME book then errors occur. Hit Singles Vol 17 for example has the record down as entering on 9 Jan. Complete Book on 30 Jan. OCC website lists wrong. So I use this error see if I agree with the error checking nature of the publisher.

[MrTibbs]

Graham Betts' book Official Charts And Hits The Fifties may solve this. His chart for 9th January agrees totally with the NME book as you posted above Lonnie. Mills Brothers entering at 10. BUT his chart for 23rd January only agrees up to #9 with one omission. Then he lists :

6 RE Max Bygraves - Cowpuncher's Cantata (joint six with Louis Armstrong)
10 RE Nat King Cole - Because You're Mine
11 10 Tony Brent - Walkin' To Missouri
12 NE Eddie Fisher - Everything I Have Is Yours
12 NE Tony Brent - Got You On My Mind
12 12 Nat King Cole - Faith Can Move Mountains

So it would appear the NME book is wrong here and The Mills Brothers did indeed chart on 9th January.

[Graham76man]

I am confused! Does the original NME of 9 January list the record or not?
I'm guessing it doesn't, since the British Hit Singles book people would have been working with original papers and they have it listed on the 30 January.

Footnote: IF the record was an addition, it would have to have been issued in December of 1952 (nearly all records were released in the first two weeks of the month at this point). The 45 Worlds 78 section shows it was.
http://www.45worlds.com/78rpm/record/05007

A late January entry would have been odd for a December record, but it could happen.

[Graham76man]

Minor errors like Glow "Work" for "Worm" here in your text - he he!

[Metalweb]

The introduction to the NME chart book actually states:

"Under intense pressure of producing the best music weekly in Britain, mistakes were sometimes made in compiling the chart. These have been corrected wherever possible. In very rare cases (three only, we think) a record was missed out of a chart. In these instances we have guessed the likely position and inserted it as an extra."

[kingofskiffle]

In that case no, as it is still listed as 1 week on chart.

What is fun is that is is clearly an error when you look at the right chart two columns over.

https://www.dropbox.com/s/5n4u7nme6z...EBook.jpg?dl=0

And yet it still got copied as not being an error!

[Splodj]

I have only seen the NME chart book in libraries, but I thought it had the charts week by week so the real week of Glow Worm's entry would be shown anyway.

It must be possible to identify the 'minor errors' corrected. Although I suppose mistakes made when compiling the book might be misconstrued as corrections!

[kingofskiffle]

I’d agree with all that. Even assuming that the weighting/combing is wrong (I’m not saying it is - read on!) this side by side comparison is, as others have said, extremely valuable. I like seeing when records first charted and where on the different charts. It’s also interesting seeing different peak positions.

Now, I do feel that the weighting and combining is correct and is as accurate as it can possibly be, given the level of information available. We do not know all and if we did then of course we would do things differently. If we knew, for example, accurate sales from the various shops this list would be different in how it was compiled. I think the data we have is accurate based on Alan’s research. I think it’s really useful having that from a point when some data was kept or remembered. History has a habit of getting re-written (sometimes by accident) when people mis remember facts. Some ‘facts’ get taken as correct simply because the source that used them was accurate for other things so this ‘fact’ must also be correct. I’m sure all authors try and ensure accuracy, but memories do change.

A case in point: When did the Mills Brothers chart with Glow Worm in 1953? Was it 30 Jan or 9 Jan? The compilers of the first Guinness books said 30 Jan, as did all sources issued before about 1992. Now, after 1992 NME printed their chart book and said in the front they had correct a few minor errors (but neglected to say which ones - not knocking them, they didn’t have to). Now Glow Work is listed as 9 Jan. Other works duly corrected they data based on this book. This is reasonable to do, since the publisher was NME, holder of those charts and therefore accurate, right? Only the original printing shows Glow Worm not charting until 30 Jan. I have the scan. This is an error on the part of the NME book when it was printed.

[MrTibbs]

Even if some may disagree about the number of stores used, this aspect of the charts (lining them all up side by side) makes it an extremely valuable and informative resource. Thank you again for creating this chart.

Although of course the number of stores is open to debate it is as far as possible research based on Alan Smith's in-depth knowledge and contacts in such matters and well documented and accepted including this site.

The exact numbers used are less important than the proportion they create between each chart which is as accurate as is possible using available research info. I think it is fair to say that none of the charts from this era could be considered 100% accurate. All are a guide based on data which could not be substantiated.

However by putting them all together and taking an average gives, I believe, the best result achievable without accurate data from the time which frankly no longer exists.

Factoring in proportions is important and I know this view is supported by many of you fellow 'chartoholics' on here. The music paper charts of the day were not equal in respect of sampling so the BBC methodology was too simplistic by treating them as such. So the further layer of factoring in 'proportions' was adopted to provide a more balanced and fair composite chart.

The end result produces a result not a million miles from the one compiled by the BBC but it smoothes out ties and inconsistencies, even errors, and gives a more balanced outcome. Or so I believe.

[braindeadpj]

Interesting how the last 2 no.1s have been stuck at a prior position (3 and 4) for 3 weeks before either jumping to no.1 or climbing to 2 then 1. They did the same on the NME chart, but the current no.1 behaved differently on the RM chart, actually dropping before climbing back up - not yet to reach no.1 (the previous 1 by The Dreamweavers, behaved the same on both charts).

[braindeadpj]

I have to say that although the main focus here is on producing a robust averaged composite chart I also find the fact that all competing charts lined up side by side makes for interesting reading and fascinating comparisons that otherwise might go unnoticed.
The thread therefore also provides a detailed and useful database of all these main charts, whichever were available at any given time, to be all together in one place to give an overview as opposed to being presented separately.

Even if some may disagree about the number of stores used, this aspect of the charts (lining them all up side by side) makes it an extremely valuable and informative resource. Thank you again for creating this chart.

[braindeadpj]

So far the length of the charts has varied between 22 and 28 and averages around 25 (24.81) meaning that are probably on average 5 unique records each week. Of course frequently the unique record will not be unique the following week, but other times...
Once you add MM, this may shift to 25 to 32 each week, so we'll be getting closer to a top 30 each week....
Interestingly the average for 1955 is: 25.62 while for 1956 it is 24 so far suggesting there is more agreement so far in 1956 than in 1955.

[Graham76man]

It is good to see the charts all set out in comparison, but it is only down to 30 on each chart. It would be nice to see the charts down to 50.

On the other site thing. You would expect that since the discographies are trying to show that certain records did sell. But if they stick to the "official" chart it would like that some records didn't sell at all.

[Splodj]

Agreed. Historic charts are always shown as a stream of one particular chart, and seeing them side by side reveals a lot more.

In 'discographies' online it is interesting to see how they sometimes see the need to depart from the 'official' chart positions. With the Kinks for example, their discographies mention that the first single (Long Tall Sally) charted in MM and the EP containing Well Respected Man (Kwyet Kinks) reached number one in the EP chart.

[MrTibbs]

I have to say that although the main focus here is on producing a robust averaged composite chart I also find the fact that all competing charts lined up side by side makes for interesting reading and fascinating comparisons that otherwise might go unnoticed.
The thread therefore also provides a detailed and useful database of all these main charts, whichever were available at any given time, to be all together in one place to give an overview as opposed to being presented separately.

[MrTibbs]

It's all change for next week April 7th as Melody Maker's Top Twenty chart is added to the calculations albeit with a much smaller sample size of 20. Note too that from April 7th there will also be a slight adjustment in NME and RM sample sizes too.

[MrTibbs]

Greetings Pop Pickers

Here is the next Ultimate Averaged Chart for Week Ending March 31st 1956

Here are all '' the uppers, the downers, the just hanging 'arounders '

The Ultimate Averaged Chart - Week Ending March 31st 1956							NME	RM	Total
Last	This		The Sound Survey			Stores	60	52	Points
Week	Week		The Top 24 Singles Chart					20	Scored
2	1	Rock And Roll Waltz - Kay Starr					1	2	2188
1	2	It's Almost Tomorrow - The Dream Weavers					2	1	2180
5	3	Poor People Of Paris - Winifred Atwell					3	3	2016
4	4	Zambesi - Lou Busch					4	4	1904
6	5	Only You - The Hilltoppers					5	6	1740
3	6	Memories Are Made Of This - Dean Martin					7	5	1672
7	7	Memories Are Made Of This - Dave King					6	7	1628
8	8	See You Later Alligator - Bill Haley and His Comets					8	8	1456
10	9	Chain Gang - Jimmy Young					9	9	1344
13	10	Theme From 'The ThreePenny Opera' - Dick Hyman					10	12	1128
11	11	The Great Pretender - Jimmy Parkinson					11	11	1120
9	12	Band Of Gold - Don Cherry					12	10	1112
15	13	Zambesi - Eddie Calvert					13	13	896
20	14	Theme From 'The Threepenny Opera' - Billy Vaughan					15	15	672
16	15	Jimmy Unknown - Lita Roza					17	17	448
14	16	Young And Foolish - Edmund Hockridge					14		420
NEW	17	Willie Can - Alma Cogan					18	17	388
NEW	18	I'm A Fool - Slim Whitman						14	364
18	19	The Trouble With Harry - Alfi and Harry					16		300
17	20	Seven Days - Anne Shelton						16	260
12	21	Rock Island Line - Lonnie Donegan					19		120
19	22	My September Love - David Whitfield						19	104
NEW	23	Nothin' To Do - Michael Holliday					20		60
NEW	24	A Tear Fell - Teresa Brewer						20	52
21		My September Love - Robert Earl
22		Dreams Can Tell A Lie - Nat King Cole
23		Tumbling Tumbleweeds - Slim Whitman

[MrTibbs]

It's a mystery to me how Winifred Atwell's horrendous tinny version of Poor People took off here while the vastly superior version by Les Baxter, a US #1, failed to even chart here. Sometimes there just ain't any justice !