John Sides posts this graph:
You will perhaps not be surprised to hear that I have no comments on the substance but I have some thoughts on the presentation. I’d bound the y-axis at 0 and 100% (currently it goes beyond these limits), also I’d put the year labels between the hash marks rather than on them (think about it: on this scale, 1995 is a time period, not a single point), also I’d put percent signs on the y-axis (e.g., “25%” rather than “25”) for some useful redundancy. But other than these minor comments, I think the graph is beautiful.
The year-labeling issue is not completely trivial, especially when trying to interpret when the series ends. I’ve noticed that people often have difficulties representing time on the x-axis of graphs. Other times, you’ll see, for example, a graph going from 1950 to 2000 with 50 little hash marks and tiny slanted labels at 1951, 1953, 1955, 1957, etc. Instead of simply labeling every 20 or 25 years.
I think it depends on the length of the time period the data was collected over. If the data was collected in one day then on that scale it makes sense to treat the time data as essentially continuous and so the labels on the tick marks make sense. If the data was collected over a month or longer then it makes more sense to to think of time as discrete and to have the year label in the middle … but then I'd also label the months or quarters on the graph as discrete entities too.
In part I guess it depends if you are look at the "big picture" i.e. the trend. Or interested in the finer detail, trying to relate the data to other happenings in time e.g. when the surge occurred.
Big picture or trend, annual observation or daily, with a data set of this breadth, the tickmarks should appear between the years (between 12/31 and 1/1) and the year labels should be centered between the markers.
This is interesting: I work at an evaluation-center, where we often make figures like theese. I often advocate that øprofile plot's like this is very fitting for this kind of data (all though I' prefer slightly bigger data-points/tickmarks). Most of my collegueas however believe that since the data is not continuosly, a barchart is more appropriate, because the chart only has data for four different years, not eg. every week.