Joel Whitburn criticism: chart fabrication, misrepresentation of sources, cherry picking


originally published on 3 March 2013; revised on 12 November 2015; latest edit: 17 November 2020


In a review titled “A Discographic Deception,” dated July 13, 1987, by expert discographer and sound historian Tim Brooks, of the notorious but influential venture into acoustic-era popular music chart fabrication by Joel Whitburn in his 1986 book Pop Memories 1890-1954: The History of American Popular Music, the author exposes flaws in Whitburn’s research methodology, and reveals numerous instances of Whitburn’s astonishingly bad habit of mispresentating sources. In case after case, according to Brooks’ analysis, Whitburn’s cited sources contain nothing like what he says they do, and specifically do not contain the periodic “lists of top popular recordings,” sales data, or ranking information he attributes to them, and which might have contributed to the chart data he claims to have derived from them. Whitburn’s source of “sheet music sales” data for acoustic-era recordings is unidentified. An edited version of response letter from Joel Whitburn, and a rebuttal by Brooks, follow the review.

Quotes from the 13 July 1987 Tim Brooks review “A Discographic Deception,” regarding Whitburn’s Pop Memories 1890-1954:

it must be said, the entire book is a colossal fraud.

But wait, you say, you didn’t know there were any popular charts in 1890? You are right. Whitburn simply made them up.

The great danger is that Whitburn’s apparently precise data, with its impressive looking sources, will be reprinted and enshrined as historical fact by others. This has already begun to happen…Whitburn has certainly been misleading in not making it clear that his “charts” are entirely speculative, and, as we have seen, none too accurate.

In a later scathing 26 November 1989* review/exposé by Tim Brooks of Joel Whitburn’s Pop Memories the author wrote,

This is a dangerous bookHere we have a whole book full of misleading information, presented in such a factual, almost statistical manner that it is bound to be quoted.

From the introduction of the book Popular American Recording Pioneers:1895-1925 by Tim Gracyk with Frank Hoffmann (2012), p. 11:

A few writers in recent years, following the example of Joel Whitburn’s Pop Memories: 1890-1954 (Record Research Inc., 1986), have cited precise chart numbers for early recordings — what records after being released were number one, number two, number three, and so on. It is a deplorable trend, and I never refer to chart positions. Primary sources provide no basis for assigning chart numbers. No company files tell us precise numbers; trade journals never systematically ranked records

At no time in the acoustic era was enough information compiled or made available about sales for anyone today to create accurate charts or rank best-sellers, and the further back in time we go, the more difficulty we have in identifying hits. Even if one had access to sales figures of the 1890s, a chart of hits means little for an era when records of many popular titles were made in the hundreds, not thousands or millions

All chart positions concerning recordings of the acoustic recording era are fictitious, and since they mislead novice collectors, they do much harm.

Billboard published the first singles popularity charts in 1940 (see the section “History, methods and description” in the Wikipedia article on Billboard charts). The Hit Parade chart first published in 1936 seems to have been something else, though the Wikipedia article on Billboard chart history contains contradictory definitions of the term hit parade. Wikipedia’s article on the term Hit parade begins by saying it is a ranked list of the most popular recordings at a certain point in time, but subsequently states that through the late 1940s the term hit parade referred to a list of compositions, since “In those times, when a tune became a hit, it was typically recorded by several different artists.” The latter use of the term suggests that the Billboard Hit Parade charts from 1936 to 1939 may have only ranked songs, rather than individual recordings, according to their popularity. I’ve yet to find any of these early Billboard Hit Parade charts online.

In section 1.4 of the article Review of Irving Berlin: A Life In Song, by Philip Furia, by music theorist and historian David Carson Berry, published in the Volume 6, Number 5, November 2000 issue of the journal Music Theory Online, the author says (link and italics added):

Beginning in 1935 (nearly three decades after Berlin’s first song), Your Hit Parade was broadcast on radio (and later on television); it offered a national survey, with chart positions based on various factors including sales of records and sheet music (although its evolving rankings formula was never clearly articulated). Thus, from 1935 onward, one can at least say that a song was number X according to that particular source. For songs released beforehand, however, there is no consistent way to derive such a ranking. Variety and other publications may provide ad hoc sales figures, or print very specific charts (for example, of record sales by a given company in a given market), but they offer nothing that would enable one to say, so generally, that a song was “number X” nationally.

From where did the national (US) chart positions for years prior to 1935 or 1936, that can now be found all over the web, come from? As outlined above, they sprang from the imagination of Joel Whitburn. There were no national charts during this era for best-selling or most popular records; therefore, no national chart positions. Why make them up?

Popular websites which routinely and extensively incorporate Joel Whitburn’s fabricated pre-1936 “chart” figures include:

  • Wikipedia
  • — via “charts” drawn from, a site which is presently down


Links to the aforementioned reviews and articles, and other relevant criticism

Pre-1936 popular music chart fabrication, and misrepresentation of sources:

  • Tim Brooks July 13, 1987 review, titled “A Discographic Deception,” of the Joel Whitburn book Pop Memories 1890-1954: The History of American Popular Music (1986)
    • But wait, you say, you didn’t know there were any popular charts in 1890? You are right. Whitburn simply made them up.
    • “…the entire book is a colossal fraud.
  • Tim Brooks, 26 November 1989 review of the Joel Whitburn book Pop Memories 1890-1954: The History of American Popular Music (1986)
    • This is a dangerous book
    • Here we have a whole book full of misleading information, presented in such a factual, almost statistical manner that it is bound to be quoted.
  • The Mainspring Press Record Collectors Blog (link broken as of 30 January 2017)
    • The charts referred to were actually fabricated by pop-culture writer Joel Whitburn, and were quickly recognized by astute collectors and researchers as fictitious.”
  • from Review of Irving Berlin: A Life In Song, by Philip Furia, section 1.4, by music theorist and historian David Carson Berry, published in the Volume 6, Number 5, November 2000 issue of the journal Music Theory Online.
    • (referring to Pop Memories 1890-1954, by Joel Whitburn) There is much misinformation to be found in books like this, however, which fabricate single-number chart positions; rankings often are based on sources that may have reported company-biased or only regional information, and, in extreme cases, information may be merely anecdotal.

Fictitious chart performance profiles for mid to late 1950s recordings, created by combining cherry-picked data from multiple charts:

  •  forum at; 21 November 2011 thread
    • Whitburn doesn’t only cherry-pick the peak positions but all other chart information as well. He takes the peak from one chart, weeks spent on chart from a different chartHe does it this way for every record charted in any of these [Billboard pop] charts between 1955 and 1958. So what you see in Joel Whitburn’s books for that period is indeed a cherry-picking from four different charts published by Billboard during that era.” —


* The article at is dated November 26, 1989. However, the page URL bears the date 1990, and a note at the bottom of the page, following the article and its ten footnotes, indicates that the page, which consists almost entirely of the review, was “last modified on November 5th, 2011.”

11 Comments (+add yours?)

  1. Rick Heinegeber
    Jan 23, 2015 @ 17:50:56

    This was great to read, I never liked Whitburn. The “token” payment system for his database is just plain retarded, as opposed to a monthly subscription. I always wondered what kind of scam he had going with Billboard to get access to all their record information. Some sort of exclusive deal I am told. I have been using the Bullfrogpond music database for songs from 1890-1945 but I believe they simply use Whitburns sources. Which begs the question, What is a reliable source for releases pre 1946? It is also sad that all Billboard Charts apart from the year end 100 are locked away in a vault somewhere. I would love to follow the historical Disco / Dance and Alternative Charts somehow.



    • doc
      Jan 23, 2015 @ 21:15:02

      Hi Rick,

      It is also sad that all Billboard Charts apart from the year end 100 are locked away in a vault somewhere.

      Beginning in 1940, popularity charts have been published in the weekly issues of the magazine. The length of the published charts has varied over time, and data for non-charting records is presumably archived somewhere.

      From 1940 until the institution of the Hot 100 singles chart in 1958, there were usually three separate Billboard singles popularity charts, and the number of singles in two of the three charts varied, according to Wikipedia, from 20 to 50, while the third remained at 20 during this period (from Wikipedia‘s page on the Billboard Hot 100):

      Best Sellers in Stores: ranked the biggest selling singles in retail stores, as reported by merchants surveyed throughout the country (20 to 50 positions). It is the oldest of the Billboard charts and dates to 1936.
      Most Played by Jockeys: ranked the most played songs on United States radio stations, as reported by radio disc jockeys and radio stations (20 to 25 positions).
      Most Played in Jukeboxes: ranked the most played songs in jukeboxes across the United States (20 positions). This was one of the main outlets of measuring song popularity with the younger generation of music listeners, as many radio stations resisted adding rock and roll music to their playlists for many years.

      These charts were replaced in 1958 by the Billboard Hot 100 chart (introduced 4 August 1958), which was supplemented in 1959 by the Billboard Bubbling Under Hot 100 chart (introduced 1 June 1959). The number of items in the latter chart has fluctuated over the years between 10 and 36.

      I’ve also noticed that the number of records (singles) in the Billboard Year-End charts fluctuated quite a bit in its early years:

      top 39-47, 1946-1948
      top 30, 1949-1955
      top 50, 1956-1958
      top 100, from 1959.
      (Derived from

      Which begs the question, What is a reliable source for releases pre 1946?

      According to Wikipedia, Billboard published the first “Music Popularity Chart” in 1940. I’ve briefly referred to this history above, in this page. Data from prior to 1940 rapidly becomes increasingly scarce and questionable. I’m unaware of a reliable source, or combination of sources, for earlier popularity rankings of recordings.

      Added 14 July 2015: There were no national popularity rankings of recordings that I’m aware of prior to 1940. But see my comments below regarding the Billboard “hit parade” launched in 1936, which may have ranked songs rather than recordings.



      • Rick Heinegeber
        Mar 16, 2015 @ 21:56:26

        Thanks for your reply Doc. Really like your site, lot of good info, very informative. I would have to say Annette Hanshaw is my fav ’20s gal. There is a great radio show where they play actual 78rpms from 1890-1930s, its called “The Ragged Antique Phonograph Program” on WFMU, you should check it out.

        Liked by 1 person


      • doc
        Mar 17, 2015 @ 00:03:31

        I’m unaware of a reliable source, or combination of sources, for earlier record popularity rankings, or “charts.”

        However, there is a gray area from January 1936, when Wikipedia says Billboard published it’s first music “hit parade,” to the advent of the Billboard “Music Popularity Chart” in July 1940. Also, I haven’t yet accounted for the apparent contradiction of the statement, quoted from the history section of Wikpedia’s Billboard Hot 100 article, in my comment above regarding the Best Sellers in Stores category: “It is the oldest of the Billboard charts and dates to 1936.” Was that a “hit parade” chart which ranked songs, or a singles popularity chart? I’ve yet to discover digitized Billboard charts from the period 1936 to 1939. If they exist, then presumably they are available in archived issues of the magazine from the period.



    • Mike
      Jul 22, 2016 @ 20:34:45

      You might want to check out the website, as they have most of the Billboard magazines since 1940, in pdf files that are free to look at. In addition, I believe they are also in the process of doing the same with Cashbox and Radio & Records, too.

      Liked by 1 person


  2. Eugene Overton
    Dec 24, 2016 @ 11:30:06

    Mr Whitburn book Pop Memories 1890-1954 and his subsequent update 1900-1940 (which is just a technical update in a larger book size format) is not a fraud. It is true that some of the criticism about recreating the acoustic era and pre world war 2 era is valid but not completely. Those people who are criticizing him and others need to be out there doing their own research and adding to the data not just running their mouths!

    There is no possible way any one person or one set of research would be able to be one hundred percent accurate on this era of music especially when there was NOTHING to start with! There were just the old records which people had listened to throughout those years with no points of reference. What Mr Whitburn did along with his handful of researchers which assisted him was to create a framework. They painstakingly attempted to obtain as much information as possible during the mid 1980’s, Since that time more data and information has come forth and other records, and charts have been created by others like Barry Kowals Hits of All Decades. He has charts of the HIT PARADE from 1935 as well as charts from 1900-1904 which have a great deal of more information than the top 5 chart data provided by Mr Whitburn. There are now numerous sources including the old magazines themselves now avaiable online. These include Billboard, Variety and other older sources.

    The main criticism I have about Mr Whitburn is that although he did a large amount of extensive research and created a powerful framework for artists and yearly hits for the pre 1940 era he has never updated or expanded it. The fact is for over 30 years he has contributed nothing to adding new or lost information which has emerged over the last 3 decades. So in my opinion one needs to look upon Mr Whitburns book and his research as the first part-the beginning only and to others for the rest of the picture of this vast musical era. There is more information and record data being found every day. So, while he created something which is now cited online and referenced. It should never be seen as the only source or definitive. It is just the overall framework. It is up to us to include as many other sources and information to supplement this and then and only then can one begin to get a more complete picture of all the songs by artists and missing potential hits.

    Mr Whitburn only gave us a small amount of recordings as he only created Top 3, then Top 5, Top 10, Top 15 Top 20 etc. If one could do enough research to create a Top 100 of each week from every year from 1900-1940 then we might actually have a much more precise or better look at all of the vast music recorded but not yet documented. There are a lot of artists and songs which are still lost to history as they never “charted”. In time this will change as more people do further research and attempt to document lost information.

    There is never going to be a BIBLE for this era. It will always be subjective. But the more information one looks at including Mr Whitburn and others who have spent their time and effort working on finding lost records and artists along with charts they have compiled. That information taken in total will give one the best picture if he or she is looking to find it. But remember there were no charts no nothing. At least now we have a much better insight to the music of the first 40 percent of the 20th century. And in my opinion that is all to the good!



  3. Joseph Scott
    Aug 30, 2017 @ 11:26:32

    “would be able to be one hundred percent accurate” What a strange way of describing the reality that he just made numbers up, just pretended to something that was literally statistically impossible. Pretending you’ve done the literally impossible is fraud.

    Liked by 1 person


  4. Ronald Volbrecht
    Sep 18, 2019 @ 07:23:00

    I am forever finding birth years to be wrong as well. The last 3 birth years of September born singers as relayed on hy local radio station, were off in Joel’s book by 1 year. Not a big deal, but now with your comments, it all makes a lot of sense. Thank you for your knowledge.

    Liked by 1 person


  5. tillywilly
    Jul 03, 2020 @ 14:17:45

    Eugene Overton makes some reasonable assertions
    I am no dummy when it comes to American Popular music research
    The simple truth is there is an enormous variation of data, and we all know the pre-WWII charts are never going to get you an accurate reading of no. 1 no. 2 etc

    I just went through The Billboard of January 4, 1936
    There is a chart for Sheet Music Leaders; then 10 Best Records for Week Ended Dec. 30; another presentation of the Sheet Music; and finally Radio Song Census.

    Each chart explains it samples only 3 of the top sources, so all the other companies and labels are not included. “10 Best Records for Week” are 10 releases each by Columbia, Brunswick and Victor labels. Decca, Okeh and the others are not in the survey. This chart sucks. But if you read Billboard over the years, you can see the continuing effort they made to provide what is a very popular product. Their charts got much better by 1940, and in January 1944, there was a quantum leap in popular, hillbilly and whatever they called race records, blues and r&b. The terminology also changed over the years, and coon records went away.
    Anyway, it takes years of study to know all the record labels, artists, composers, the record business, broadway musicals that most popular songs were written for so many years, I could go on and on. I have absolutely nothing against Whitburn. Of course his charts are not accurate, but they are a serious attempt to document a subject we are all very interested in. On Usenet, they took his work, and advanced it greatly with more sensible formulas and standards, while using all his questionable data. It gave me a list of records to listen to, and I am not bothered that many of my favorites never made anybody’s chart. Thank you to all the tireless researchers that posted massive amounts of data from defunct record labels, that I keep in spreadsheets, done my way of course!



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: