Wednesday, April 29, 2009
I had mentioned before that, while looking at things like how many copies the #1 title each month had sold over time (and that data appears here) was interesting, we could learn some more by looking at the changes across the entire distributor Top 300 Comics lists. To show what I mean, here's a quick comparison of March 2009 with a couple of previous months.
First, let's look at March 2003, which was the first March for which Diamond reported Final Order figures. We can go earlier, but let's keep things apples-to-apples for the moment. March 2003 is an interesting month to look at for a number of reasons. 2003 was a mini-slump in the business, so there are similarities there; the March Top 300 sold just a quarter-million copies more than in March 2009 (where the fewer number of copies ordered still sold for more than $2 million more). So the number of copies represented in the Top 300 is fairly close; March 2003 sold about 5% more copies.
But let's look at where those extra copies were — and were not:
The blue line is the actual unit sales of every comic book, from 1st to 300th place, in March 2003; the red line, for March 2009. They're both sloping arcs, as we might expect. But to see the differences more closely, let's look at the front and the back halves separately:
At ranks #1-8, the March 2003 books are ahead, and in some cases, far ahead of the March 2009 books. Batman's "Hush" was still going, and Marvel's Ultimate books accounted for four of the top eight. But orders fell dramatically from eighth to ninth place in March 2003 — from 84,450 copies to 61,400 copies — allowing March 2009 to move ahead. From there, the items down to about 56th place trade off — sometimes March 2003's selling better, sometimes March 2009's titles selling more. From 57th place to the middle of the list, we see the items in March 2003's list selling better — just slightly better in the lower half of the Top 100, as much as 26% better in the 130s and 140s. You can see where the gap widens and narrows.
But now let's look at the second half of the chart:
With the scale altered, we can see something quite different about the second half of the Top 300 in March 2009. The arc of 2003 is closer to a line in 2009, with very little variance from the mean. At 192nd place, the comic books in 2009 outpace those of 2003 — and they just clobber the low-end titles once we reach the mid-200s, selling as much as 42% more copies as their earlier counterparts.
So what's going on? The arc of 2003 slope has been flattened — with sales from the top of the list and the upper midlist pushing down toward the lower portion, and making sales on the whole list more evenly distributed. There's a simple reason: Look again at the table for March 2003. The 200s are populated by familiar names like Archie and Antarctic, but also smaller publishers like Gutsoon and Aerosol. Marvel is pretty much not to be found there: it had 57 entries in the Top 300, and only 1 item in the 201st-300th place range. In March 2009, it had 99 entries on the list — and 22 in the bottom third of the chart. Marvel, DC, Image, Dark Horse, and CrossGen (the top five) combined for 215 slots in March 2003; 237 in March 2009 (with IDW in place of CrossGen).
Simply, the top comics in March 2003 were selling better — and there was more strength evident in the upper midlist. But comics from the majors are taking more rack positions than they were, and so we're likelier to find them appearing further into the list than before. The March 2009 chart represents only 5% fewer comic books sold — but the audience within appears to be fragmenting and spreading across more titles than before.
This difference between the "shapes of the market" is quite apparent here, and can be found in other months — though it grows less pronounced in comparison with more recent months. A look at March 2006, when the market was doing very well and the Top 300 sold 31% more comics, shows the difference in slope, but the first thing one notices is that the market outsold March 2009 at every single position:
Though even here, the difference in performance between the first half and the second half of the chart can be seen:
The slope in 2009 is much more uniform.
I haven't done a regression analysis on these charts, but a large explanatory factor in how high the curve starts on the page from month to month — on the Y-axis — is likely the number of comics shops, since it's their orders that are positioning these lines. Major editorial events most greatly impact the sharpness of slope at the very far left on the charts; the number of major publishing offerings, the uniformity of slope on the right side of the chart. The slow, ongoing transformation of the chart from an arc to a line — particularly in that second half — is interesting, however, and may be worthy of a second look by someone.