Comments on: Ebb and Flow of Box Office Receipts Over Past 20 Years http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/ Strength in Numbers Sun, 12 Feb 2012 04:57:37 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Moritz Stefaner http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49851 Moritz Stefaner Fri, 14 Mar 2008 09:30:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49851 Here is another variant of the visualization principle I did for my <a href="http://well-formed-data.net/thesis" rel="nofollow">master's thesis</a>: <a href="http://well-formed-data.net/archives/77/emerging-topics-update" rel="nofollow">Emerging topics</a> Here is another variant of the visualization principle I did for my master’s thesis:

Emerging topics

]]> By: Egg Syntax http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49850 Egg Syntax Fri, 29 Feb 2008 02:14:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49850 Ha! I'm pleased. I think I get some sort of points. "VisPoints"? "Data Points"? ;) Lee Byron, AFAIK, studies with Ben Fry at CMU, and I definitely see Fry's influence in this work. Ha! I’m pleased. I think I get some sort of points. “VisPoints”? “Data Points”? ;)

Lee Byron, AFAIK, studies with Ben Fry at CMU, and I definitely see Fry’s influence in this work.

]]> By: Nathan Yau http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-1893 Nathan Yau Wed, 27 Feb 2008 15:59:23 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-1893 @egg syntax: definitely right on that one! the lastfm stuff was the inspiration for the graphic: http://www.portfolio.com/views/blogs/odd-numbers/2008/02/26/q-amp-a-anatomy-of-a-graphic @egg syntax: definitely right on that one! the lastfm stuff was the inspiration for the graphic: http://www.portfolio.com/views.....-a-graphic

]]> By: Egg Syntax http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49848 Egg Syntax Tue, 26 Feb 2008 17:45:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49848 Visually this reminds me most of Lee Byron's piece which shows listening histories for individuals from last.fm. See http://www.megamu.com/lastfm/ Visually this reminds me most of Lee Byron’s piece which shows listening histories for individuals from last.fm. See http://www.megamu.com/lastfm/

]]> By: Hadley http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49847 Hadley Tue, 26 Feb 2008 02:34:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49847 nick: it's basically a stacked area chart, with some smoothing of the raw data. i don't think the horizontal position is that important, i suspect it would be easier to read (but not as pretty) with a flat baseline. nick: it’s basically a stacked area chart, with some smoothing of the raw data. i don’t think the horizontal position is that important, i suspect it would be easier to read (but not as pretty) with a flat baseline.

]]> By: Jono http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49846 Jono Tue, 26 Feb 2008 02:15:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49846 It's pretty, but I'm not sure it's not too pretty. One thing which surprises me is that it seems to show a film as peaking some time after it opens, and not on the first weekend. It's possible this is just an artifact of the smooth spline they used, but still it's misleading and therefore a bad visualization. But really I think it's wrong. Look at that last big movie on there, National Treasure. It's biggest daily gross is on opening day (Friday 12/21). The graph clearly doesn't read that way. The label "Dec" is clearly centered on the month, so the "e" is about on the 15th, so that peak is probably on the 21st. So why the smooth ramp up which is at least a week long, if not longer if you count the hidden bit. Only sampling on weekends and then fitting a smooth spline is a BAD visualization! It doesn't help that when you mouse over a film you don't see the whole "shape" of that film, as previously noted. It’s pretty, but I’m not sure it’s not too pretty. One thing
which surprises me is that it seems to show a film as
peaking some time after it opens, and not on the first
weekend.

It’s possible this is just an artifact of the smooth spline
they used, but still it’s misleading and therefore a bad
visualization. But really I think it’s wrong. Look at that
last big movie on there, National Treasure. It’s biggest
daily gross is on opening day (Friday 12/21). The graph clearly
doesn’t read that way. The label “Dec” is clearly centered
on the month, so the “e” is about on the 15th, so that peak
is probably on the 21st. So why the smooth ramp up which is
at least a week long, if not longer if you count the hidden
bit. Only sampling on weekends and then fitting a smooth
spline is a BAD visualization!

It doesn’t help that when you mouse over a film you don’t
see the whole “shape” of that film, as previously noted.

]]> By: Dave http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49845 Dave Mon, 25 Feb 2008 22:55:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49845 It's interesting that you can almost see that the post peak slopes are sharper starting in the 90's when the average run in a theater got much shorter. It’s interesting that you can almost see that the post peak slopes are sharper starting in the 90′s when the average run in a theater got much shorter.

]]> By: Bruce Nash http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49844 Bruce Nash Mon, 25 Feb 2008 22:24:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49844 >>> You might also have noticed that the flows get a lot bigger as you move forward in time, even though the numbers are adjusted for inflation. Strange. I wonder why that is. <<< Interesting question. Ticket sales did rise through the 1980s, but sales have basically been flat since 1998 (see, http://www.the-numbers.com/market/ and http://www.natoonline.org/statisticsadmissions.htm), so the apparent growth could stem from a number of factors. 1. Maybe the figures are adjusted for retail inflation, not ticket price inflation (which has been slightly steeper over the years, I believe)? It's not clear from the description, but given that the top-grossing movie in their chart made an inflation-adjusted $862 million, and my numbers have Titanic making $863 million ( http://www.the-numbers.com/movies/records/inflation.php ), I think they probably are using ticket prices. 2. Maybe the growth really occurs between 1986 and 1998, and it's actually pretty flat from then onwards. 3. It still looks to me as though there's growth after 1998, and I think that's because the studios have concentrated more of their resources at key times of the year -- the Christmas Holidays, and Summer Holidays in particular. So the peaks are higher on the graph, the troughs are a little lower. One criticism of the chart is that it's hard to actually measure these peaks and troughs. It would be nice if they provided a background scale. Regarding the placement of each movie (top or bottom), it looks to me as though they place each new movie on which side of the graph is smaller at the time it was released. That's probably the simplest rule to keep everything in balance. Nice site, by the way! Bruce >>> You might also have noticed that the flows get a lot bigger as you move forward in time, even though the numbers are adjusted for inflation. Strange. I wonder why that is. <<<

Interesting question. Ticket sales did rise through the 1980s, but sales have basically been flat since 1998 (see, http://www.the-numbers.com/market/ and http://www.natoonline.org/statisticsadmissions.htm), so the apparent growth could stem from a number of factors.

1. Maybe the figures are adjusted for retail inflation, not ticket price inflation (which has been slightly steeper over the years, I believe)? It’s not clear from the description, but given that the top-grossing movie in their chart made an inflation-adjusted $862 million, and my numbers have Titanic making $863 million ( http://www.the-numbers.com/mov.....lation.php ), I think they probably are using ticket prices.

2. Maybe the growth really occurs between 1986 and 1998, and it’s actually pretty flat from then onwards.

3. It still looks to me as though there’s growth after 1998, and I think that’s because the studios have concentrated more of their resources at key times of the year — the Christmas Holidays, and Summer Holidays in particular. So the peaks are higher on the graph, the troughs are a little lower.

One criticism of the chart is that it’s hard to actually measure these peaks and troughs. It would be nice if they provided a background scale.

Regarding the placement of each movie (top or bottom), it looks to me as though they place each new movie on which side of the graph is smaller at the time it was released. That’s probably the simplest rule to keep everything in balance.

Nice site, by the way!

Bruce

]]> By: Nathan Yau http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49843 Nathan Yau Mon, 25 Feb 2008 21:33:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49843 I haven't seen any other formal naming of the chart other than ThemeRiver in the paper linked in the post... the baby name wizard, however, is known as a stack graph. I haven’t seen any other formal naming of the chart other than ThemeRiver in the paper linked in the post… the baby name wizard, however, is known as a stack graph.

]]> By: nick http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49842 nick Mon, 25 Feb 2008 20:32:00 +0000 http://flowingdata.com/2008/02/25/ebb-and-flow-of-box-office-receipts-over-past-20-years/#comment-49842 does anyone know what this type of graph is called, or a program that can create them? a similar, if less elegant visualization demonstrating frequency of baby names: http://babynamewizard.com/namevoyager/lnv0105.html does anyone know what this type of graph is called, or a program that can create them?

a similar, if less elegant visualization demonstrating frequency of baby names:
http://babynamewizard.com/namevoyager/lnv0105.html

]]>