John Jackson Miller picks up his calculator again

twitter John Jackson Miller picks up his calculator again0facebook John Jackson Miller picks up his calculator again0google John Jackson Miller picks up his calculator again0pinterest John Jackson Miller picks up his calculator again0tumblr John Jackson Miller picks up his calculator againreddit John Jackson Miller picks up his calculator again0stumbleupon John Jackson Miller picks up his calculator again0email John Jackson Miller picks up his calculator again

Newsarama has started a new columns by noted sales figure collector John Jackson Miller. there is much to absorb in his maiden outing, but here’s a snippet: then and now.

January 2008 sales are up over those of January 1998, a month from deep in the comics recession when the “dead quarter” very definitely lived up to its name.

Jan. 2008 final orders versus Jan. 1998 preorders (est.)
Top 300 units: +3% (7.18 million copies vs. 7 million copies)
Top 300 dollars: +34% ($22.33 million vs. $16.6 million)
Average cost of comics in the Top 300: +23% ($3.22 vs. $2.61)
Average cost of comics in the Top 300, weighted by orders: +31% ($3.11 vs. $2.38)

Most, but not all, of the difference in dollars represented by the Top 300 comes from the increase in cover prices. But the overall market is certainly larger today — Diamond did not release indexed trade paperback data for the month, but the category was certainly far smaller than it is now.

Comments

  1. As Miller implies when referring to cover price changes, the dollar figure is misleading since it’s not inflation-adjusted.

    Adjusted for inflation, the top 300 in 1998 accounted for $21.37 million, for a more modest +4% which is commiserate with the +3% on unit sales.

  2. I wonder if using the January 2008 figures as they presently stand isn’t moot, though, given that there’s apparently been a major bend in the Diamond charts for the month.

    If they really calculated about five percent too much for weeks 2 through 5 of January, as I suspect, the Jan 2008 numbers probably aren’t as great as it seems right now.

  3. I saw the same bend — it took awhile to figure what was going on. I found the overstatement (if that is what it is) increasing by about 2% for each week in the month according to one publisher data set I was looking at. That looks like reorders to me — as if reorders from February Week 1 were allowed to enter the charts. February Week 1 reorders would affect the January Week 5 figures the most, with less impact on the earlier weeks. That would fit with what I observed.

    My bet is that we will find a commensurate understatement in the February charts, with fewer January titles climbing onto the bottom of the list than we might otherwise have seen. If that happens, yes, the monthly measures will be imperfect to that degree — though it would be difficult to see how to recalculate them without new data from Diamond.

    It is one reason I am also running quarterly stats on comichron.com, as these remove these and other monthly variations.

  4. Marc-Oliver Frisch Said:

    …as I suspect, the Jan 2008 numbers probably aren’t as great as it seems right now…

    I was thinking the same thing – several books looked like they made a slight unaccounted upturn…

  5. “That looks like reorders to me — as if reorders from February Week 1 were allowed to enter the charts.”

    That sounds very plausible, but I don’t think it’s the whole story. It doesn’t seem to cover the sudden sales jump for COUNTDOWN in the second week of January, for instance, from which point the book continues a fairly typical decline.

  6. The first-week Diamond indexes were at the least variance from the actual figures, I found — and at the least variance from each other, whereas as the month progressed, variance increased. The “Week 2″ order index numbers suggested by the actual sales were closer to the Week 1 number and more tightly compacted with each other than the Week 5 ones were — they were scattered more, relatively speaking. Which is another argument for something having to do with the end of January or the beginning of February.

    So I focused on the group that seemed most like “normal” and closest to actual estimates, drawing on additional known clues that suggested that the 745 number did indeed reflect what was sold for issues later in the month, if not exactly within the month. An example is that Phoenix Comicon variant of Hulk — a week 2 ship that sold out a 5,000-print run order. If you go from the size of the run, it stacks up right at 745, IIRC.

    There is another possibility, which is that the reports the publishers received accounted for a different period than the Diamond charts did — for example, if they pushed the report buttons at different times. That used to be the #1 cause of the variance we saw in the preorder days — you’d never, ever get the charts to square up, because the chart was always run later than the publishers got their reports and did not report the exact same period.

    A lot of strange things happen in that first ship week of the year traditionally, so I don’t know what happened with Countdown — just thinking about it, you’d expect the holiday week orders and shipments to be lighter. And there was the holdover sales from 2006 into 2007 — maybe the same thing happened and hit this week. It’s hard to know.

    Again, in the absence of a new chart, I would look to February to see what happens. Some real February sales slipping into January would not be a huge problem as long as we’re not tabulating twice.

  7. >>Some real February sales slipping into January would not be a huge problem as long as we’re not tabulating twice.

    That is, as long as we’re not counting the same copies as having been sold twice. As evidenced this month, we re-tabulate all the time!

  8. “That is, as long as we’re not counting the same copies as having been sold twice.”

    Well, as I recall, that’s precisely what Diamond seems to have done in the January 2007 numbers, so I’m not getting my hopes up!

  9. As Marc says, the January chart is clearly very odd indeed, and at this stage, I would treat it with extreme caution. There’s an awful lot of comics going up for no apparent reason, and I have great difficulty believing that an extra week of re-orders could provide an explanation for that.

    Somehow, AMAZING #549 and UNCANNY #495 have also made it onto the chart, even though they didn’t ship until February. This strongly suggests to me that something has gone seriously awry at Diamond’s end.

  10. Yeah, I looked at the extra titles as further support that we were looking at “January-plus” in some manner — though that couldn’t have been the whole explanation, as it was only those books. I do believe a week’s reorders could cover these percentages — 6-8% across the month after shipping is historically a common rate, and it would improve the Week 4 and 5 stuff here to a greater degree.

    Interestingly, publishers’ own tracking files from Diamond that I’ve seen don’t begin with the first Wednesday or even the first of the month, but track back nine days from that first Wednesday and run to ten days before the next month’s first Wednesday. So “October 2007,” for Diamond’s system, was actually Monday, September 24 to Sunday, October 28 — ten days before the first new comics day of November, the 7th. I believe that’s because those October 3 books were in the warehouse to be picked and packed the week of September the 24th. Somehow, the numbers have always squared up, so October numbers reflect October orders and shipments — but it shows that they approach the calendar differently. Maybe some of the first-week February orders shipped normally, but accidentally got entered into the system with the wrong dates.

  11. There’s also WILDSTORM: REVELATIONS to account for. The first issue shipped in week 1 of January, the second in week 3. If you believe the Diamond charts, issue #2 outsold #1. I don’t think that’s covered by subsequent reorder activity, either.

    Mind you, as I said, I find your theory that February numbers somehow entered the picture to be plausible, but I don’t think it’s the only thing that’s wrong with the figures. There are too many other bizarre things.

  12. Agreed. The problem was that the numbers as stated weren’t suggesting any alternative case. The first “warts and all” run of the numbers I did came out messy and the Week 1 figure fit some of it perfectly, so that’s where I (and, I think, Milton) went.

    While this has been passed up the chain, the Diamond revisions triggered by the public have tended to be simpler ones to address, like a transposition of the dollar shares for market shares, which has happened before. A revision here might be more likely to be triggered if publishers found in the list a reason to complain — if the chart were penalizing unequally.

    I think it’s just January that’s cursed — last year’s was messy, as you said, and January 2001 was the only one I can recall that was so chaotic it finally did prompt a second run.

Speak Your Mind

*