The Mystery of Posts With a Long Tail

Each week, we get from Google Analytics a report of which individual posts resulted in the most visitors to RBC. Our archive holds over 12,000 posts, but the leaders are almost always posts from the current or prior week (note that post-based entries are a rarity, people usually enter by the main page or from links by other bloggers).

However, three old posts are there every week, and it is interesting to speculate why.

The most remarkable performance is The Elixir Broccoli of Life, which James wrote 10 months ago. Yet it has lead 100-200 people to RBC every week since.

The second most steady performer is Drinking Alone Does Not Necessarily Mean You Have a Drinking Problem, which funnily enough was posted only a week after the broccoli post. It’s been there almost every week since, and caused 103 people to come to RBC last week.

I have heard that seeking health advice is one of the most common reasons people first start using the Internet, and that may be why these pieces have such a long tail (They differ from most of our health posts in providing individual advice versus, say, analyzing a new health care law). Or maybe there is a website in China that links to these posts and we are just riding the wave along with whatever else they pick up as a gazillion people a day there get Internet access.

However, the health explanation breaks down for the post with the third longest tail. The 60 or so film recommendations I have written unsurprisingly almost never show up except in the week I wrote them. With one appropriately mysterious exception: My review four months ago of Cabinet of Dr. Caligari has never been out of the top post list since it appeared. It’s not as if reviews of a 90 year old classic film are in short supply so this is hard to explain…maybe A.O. Scott gave it some love at some point but didn’t wire me to let me know, which would be just like him.

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College London. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over thirteen thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

21 thoughts on “The Mystery of Posts With a Long Tail”

  1. I expect that tails will get longer in general when you permanently institute the ‘recent comments’ feature which briefly appeared last month.

  2. I’ve been wondering about this myself and was thinking of a post. Two other oldies of mine which pop up regularly in the site meter are Is Barack Obama black? (from 2008!) and Sunday rhinoceros blogging a year later. Neither of these have anything to do with health.

    Two possible mechanisms are search, and slow viral dissemination via email or social media. I can just about understand search for the Obama post – “Barack Obama black” in Google will give you my post on the third page – but the search strings that would get you to the broccoli and the rhinoceros are frankly incredible. So I go for viral dissemination.

    It’s also possible that students are guddling around in our archives to find ideas for essays and term papers. This would presumably show up in the Google Analytics data.

    The RBC welcomes all readers, including zombies, grave-robbers and adolescents.

    1. The RBC welcomes all readers, including zombies, grave-robbers and adolescents.

      Yes, but we draw the line at adolescent zombie grave-robbers.

      1. Exactly.

        Here in sunny Jersey (Channel Islands) a benevolent, and wealthy, citizen (and he truly was both) left a large property to the States of Jersey (island’s government) to be used “for the benefit of aged, infirm and needy residents of the island”

        In a States meeting the Minister in charge of property said that this meant that persons benefitting form the bequest would have to be aged AND needy AND infirm. I said, with respect, I have studied linguistics, and what you have just said is utter nonsense. Commas, in lists such as this, denote additon: aged people, infirm people and needy people can all benefit from the bequest.

        Oh no, he said, that is the LEGAL advice we have received. So there you have it, lawyers are . . .

        It matters a lot, because said building lies next to the new Town Park in our capital, and it would make a fantastic cmmunity centre linking out to the park, which has been designed for this to happen.

        BUT some . . . lawyer is making this well nigh impossible.

        Hopefully, with the passage of time the people in charge change, the legal advice quietly gets forgotten (or it changes 180 degrees) and we can all get on with it.

        1. Not just in Jersey.

          During my mother-in-law’s final illness, she executed a Power-of-Attorney appointing my wife and me “singly and jointly” as her attorneys-in-fact. I was told on several occasions that my wife’s signature was required as well as mine, because it read “singly and jointly” rather than “singly or jointly”.

        2. You’d think lawyers would adopt Boolean operators, for clarity of meaning and because it would require elaboration of language and hence obscurantism, which they seem to enjoy for the in-group culture it creates. Clearly, your lawyer antagonist is of the opinion that they already have adopted the Boolean “AND”.

          1. The irony is that in several cases it was a clerical functionary rather than an attorney doing the parsing.

            The wording on the POA was that provided for by the State. State law provides that Court costs and reasonable attorneys fees must be paid by the Respondent when someone has to obtain a Court order to take permitted actions, provided the statutory language is used.

            On the other hand, when the executor is dying of lung cancer, there isn’t time to let things wend their way through the courts.

        3. That sequence would have been a likely place for the good old lawyer’s ‘and/or’ – often a horrible expression (and of course completely illicit in an obligation, since one does not know whether one has to do one or both of the choices in order to comply).

          Part of the problem is what the person interpreting (or his/her client) *wants* to happen. If the Jersey authorities wanted to open the land to people who were any of aged, infirm or needy, or any combination of those characteristics, it would be about impossible to prove them wrong. Likewise with the power of attorney: if the exercise of the power by one of the attorneys were accepted, it would be about impossible to prove that the other interpretation needed to be made.

          Lawyers are paid to reduce risk, and I suppose that doing nothing i.e. not going along with what the petitioner wants, is marginally less risky than accommodation. But one is allowed to consider policy reasons for interpreting in one way or another.

    2. You have to consider personalized search. If somebody or a friend of theirs has the RBC listed in Google Reader, search results from may get prioritized over generic search results.

      I know I had that happen to me when I used Google Reader (personalized search results are flagged as such).

  3. For the movie review post, maybe the reason is that Google knows that you have links to a book on climate and two on drug policy, and the kind of people looking for movie reviews of a 90-year-old film (university and film school students, hipsters) are likely to be interested in those things, too.

    As an experiment, maybe try writing a post on something on a lot of syllabuses?

    1. Could be, but I shudder to think that any self-respecting doctor would keep squid in his cabinet.

  4. I notice a bunch of the older posts, including those you link here – from say 2009 – say “comments are closed” and have no comments shown. But surely there were comments at one time?

    1. Comments close at some point — I don’t know why. I assume no comments are purged when posts close, although I just work here.

  5. You can use “Secondary Dimension” in Google Analytics to track the referral sources for a particular post.

    Also, typo: third paragraph, should be “led,” not “lead.”

  6. This is the post I return to repeatedly, usually to remind myself of the comprise/consist distinction (although you seem to have changed the address since I last clicked on the bookmark on this computer: I had to poke around for several minutes to find it!).

Comments are closed.