[commentary and actual analysis (oooh, shiny) on this week’s sales rankings]

I know you folks don’t bother to read comments, or any of my posts for that matter unless a kind friend of the blog happens to post a link, so I’ll take it as a given that you missed the discussion & commentary on last Monday’s article “Soliciting Feedback.” As might be guessed from the title, I was looking for a few opinions on how the rankings looked to my intended audience, and if any changes in format were needed or wanted in the name of clarity.

As predicted, I had next to no response. One of my two respondents mentioned that a “a single, long list” might be best, and while I could transition to a Manga 1000 and screw the derivative charts, I actually really like the derivative charts. (In fact the whole mess — two weeks of data entry, the spreadsheet, the top 500 & series rankings — might be described as the necessary-but-disposable-math to the get to the Midlist 500.) In this case I decided to interpret the request as a call to move all the charts into a single, very long post — and also make it that much easier for other bloggers to link to.

If one really wanted to, one could still separately link to The Manga Top 500, the Top 50 Series, the Publisher’s Scorecard, and the Midlist 500, because I planned ahead and managed to remember/research enough HTML to keep that functionality, but as a blogger I know you’re all a lazy, self-centered lot and it’s almost asking too much to get the one link each week. Well, there ya go: I almost broke WordPress with it but yeah, it’ll fit in one post.

At the top of said post I’ve inserted a summary — the clips you’ll see above the fold if you were just browsing past the article on the Comicsnob main page. I don’t know if this strikes the proper balance between casual readers and the data-mining manga-freaks, but that was at least part of my intention in designing the new post format.

##

So let’s go back to Monday — some good stuff there. Speaking of the guy who responded to my formatting questions, Glenn Kardy took me at my word and asked an open-ended question about some of my results. Since it is not only of interest in and of itself but also points up some of the problems with and limitations of my methods, I thought I’d bump this to the main page for everyone to read:

Comment from Glenn Kardy
: April 15, 2008, 4:17 am

We’ve been watching your lists with great interest here at Japanime/Manga University. We think what you’re attempting is wonderful, and hope you will continue your efforts.

Answer to question #1: We feel a single, long list is always the easiest to digest.

Answer to question #2: An executive summary would be great — the more info and feedback, the better!

And, #3 (our open-essay question): We wonder why our imprint’s two most popular titles — “Manga Moods” and “The Manga Cookbook” — fail to make your lists, while our “Kanji de Manga Vol. 1″ is on the list week after week.

“Manga Moods” and “The Manga Cookbook” outsell “Kanji de Manga” by about 3-to-1 (according to the data we receive from our distributor and Nielsen Bookscan). And “Kana de Manga” (one of 7 volumes in the “Kanji de Manga” series) sells just as many copies as “Kanji de Manga Vol. 1″ but also never shows up on your lists.

Just curious — and hopeful!

Thanks,

Glenn

My Response

“Glenn:

“The persistence of Kanji de Manga vol 1 on the charts has an easy explanation, one I almost don’t have to reference the spreadsheet for: bn.com

“At bn.com, Kanji 1 comes in at ~#100 in their manga category this past week - a mark that it hits fairly consistently week-to-week, in fact. It’s number 103 this morning [edit: #102 at this very moment]::
http://browse.barnesandnoble.com/browse/nav.asp?No=100&N=989444&Ne=989444&visgrp=fiction

“I usually only look at the first 300 titles from any source but last week I just happened to load up a top 500 manga from B&N on April the 10th (I *might* be looking to expand my source data) and the Kanji de Manga box set was #338 in the manga category at that time.

“As one of my actual-bookstore-sources, and a big one at that, bn.com is checked three times a week, so the solid perfomance of Kanji 1 week to week has a cumulative effect that may be distorting, when compared to actual sales numbers.

“Here are some sales rankings (compared to all books, not just manga) to put things in perspective:

Kanji de Manga 1
B&N: 17,961
Amazon: 235,461
-

Kana de Manga
B&N: 21,906
Amazon: 162,519
-

Kanji de Manga vols 1-3 box
B&N: 55,979
Amazon: 371,204
-

Manga Cookbook
B&N: 75,850
Amazon: 36,491
-

Manga Moods
B&N: 303,583
Amazon: 253,940
-

Manga Sisters
B&N: 367,916
Amazon: 567,703
-

50 Things We Love about Japan
B&N: 746,418
Amazon: 761,723
-

“(similar data isn’t available from borders.com yet)

“It looks like Kana de Manga might/should show up occasionally in my Manga 500 as well, except…
at bn.com, Kana de Manga isn’t in any category at all — Manga or otherwise. Additionally, Manga Moods appears to be classed as just a Graphic Novel, not specifically manga (w/ art technique as a secondary). If B&N has a mechanism for publishers to request/submit corrections to title information, you might persue that.

“By sales rank, Kana de manga should be doing about as well as Kanji 1, but since it’s in the wrong category it never comes up on my b&n source charts.

“Japanime titles also occasionally appear on source charts from Chapters and Buy.com, but not other sales sites.

“Additionally, it looks like the Cookbook is doing much better at Amazon than at B&N. Since Amazon uses some weird keyword thing as opposed to actual categories there shouldn’t be a classification problem like that discovered for the Kana de Manga title. (The Manga Cookbook *is* classed as manga at bn.com, though, so no worries there)

“Here are my rankings from Sunday

252. ↓-1 (251) : Kanji de Manga 1 - Japanime’s Manga University, Jan 2005 [121.5] ::
568. ↑3 (571) : Manga Cookbook - Japanime’s Manga University, Aug 2007 [27.8] ::
660. ↑20 (680) : Manga Moods - Japanime’s Manga University, Dec 1899 [19.8] ::
659. ↓-168 (491) : Kanji de Manga vols 1-3 box set - Japanime’s Manga University, Nov 2006 [19.9] ::
1270. ↑new (0) : Kanji de Manga Kana de Manga Special Edition: Japanese Sound FX - , Dec 1899 [0.6] ::

“The Manga Cookbook misses the main chart but does rank #456 on the Midlist 500.

“final note: I don’t have Neilsen BookScan numbers, and likely never will. I rely on sales ‘charts’ (comparitive rankings) to try and estimate how manga titles are doing compared to each other. And my source data is rankings based on online sales — just a fraction of the total sales, and one that may vary as a percentage of total sales by quite a bit, from publisher to publisher and even from title to title. Another source of errors are the sites themselves: ‘bestseller’ rankings may reflect historical performance in addition to actual current sales — I don’t know how long of a time frame each site considers, and the bestseller ‘formula’ used will also be different for each site.

“To Answer Your Question:

“Manga Moods and the Manga Cookbook are selling much better in stores than they appear to be doing on-line. It’s a tough problem to have, I know, but I’m sure you guys will cope somehow.

“Though… The Manga Cookbook *might* just be ready for a push, up and over that curb, if you can think of a way to promote it.”

##

Aside from recycling previously written material to bulk up the commentary post (…what?) (and I would have been fine leaving the entirety of the post at that for this week) we’ve another point which I should probably cover, with the concommitant opportunity to look more deeply into the numbers and see what comes up:

Vampire Knight (at only 4 volumes) this week unseated Fullmetal Alchemist (16 manga, 5 novels, a few art books — not to mention the anime) to take our #5 series slot. Part of the reason is of course the extra weight I give to the top-ranked volume of each series when calculating the series score (#8 Vampire Knight vol 3 vs. #38 Fullmetal Alchemist vol 16) but there is also the fact that despite being outnumbered, the four volumes of Hino’s latest work are all placing in the top 25.

If only someone were tracking manga sales week to week so we could all see…

Yes, of course I’m being facetious.

Vampire Knight 4 - Viz Shojo Beat, Apr 2008
3/23 265. ↑31 (296) [106.5]
3/30 231. ↑34 (265) [136.3]
4/6 95. ↑136 (231) [253.5]
4/13 36. ↑59 (95) [387]
4/20 25. ↑11 (36) [411.5]
-

Vampire Knight 3 - Viz Shojo Beat, Oct 2007
3/23 55. ↓-9 (46) [320.2]
3/30 47. ↑8 (55) [355.6]
4/6 20. ↑27 (47) [445]
4/13 15. ↑5 (20) [489.7]
4/20 8. ↑7 (15) [529.1]
-

Vampire Knight 2 - Viz Shojo Beat, May 2007
3/23 24. ↓-3 (21) [426.7]
3/30 21. ↑3 (24) [445.3]
4/6 14. ↑7 (21) [483.1]
4/13 12. ↑2 (14) [507.7]
4/20 11. ↑1 (12) [518.2]
-

Vampire Knight 1 - Viz Shojo Beat, Jan 2007
3/23 31. ↑1 (32) [402]
3/30 23. ↑8 (31) [437.4]
4/6 17. ↑6 (23) [466.3]
4/13 18. ↓-1 (17) [465.1]
4/20 15. ↑3 (18) [502.1]
-

##

Fullmetal Alchemist 16 - Viz, Mar 2008
3/23 103. ↑133 (236) [252.5]
3/30 42. ↑61 (103) [377.6]
4/6 38. ↑4 (42) [389.6]
4/13 41. ↓-3 (38) [380.9]
4/20 38. ↑3 (41) [370.1]
-

Fullmetal Alchemist 15 - Viz, Dec 2007
3/23 57. ↑18 (75) [316.5]
3/30 50. ↑7 (57) [347.7]
4/6 42. ↑8 (50) [380.5]
4/13 47. ↓-5 (42) [349.3]
4/20 95. ↓-48 (47) [256.9]
-

Fullmetal Alchemist 14 - Viz, Aug 2007
3/23 86. ↑7 (93) [272.4]
3/30 85. ↑1 (86) [270.2]
4/6 91. ↓-6 (85) [256.3]
4/13 95. ↓-4 (91) [251.7]
4/20 91. ↑4 (95) [260.5]
-

Fullmetal Alchemist 2 - Viz, Jun 2005
3/23 78. ↑7 (85) [281.5]
3/30 79. ↓-1 (78) [281.4]
4/6 78. ↑1 (79) [280.2]
4/13 70. ↑8 (78) [296.5]
4/20 62. ↑8 (70) [310]
-

Fullmetal Alchemist 1 - Viz, Apr 2005
3/23 131. ↑26 (157) [215.6]
3/30 116. ↑15 (131) [239.4]
4/6 127. ↓-11 (116) [217.9]
4/13 165. ↓-38 (127) [182.2]
4/20 162. ↑3 (165) [182.5]
-

##

FMA is holding it’s own, particularly the new volume (which is doing quite well, all things considered) but in comparison, Vampire Knight is surging, blowing into the top 25 as actual sales catch up with preorders of vol 4 (gaining 200+ ranks in a matter of weeks) and other online shoppers fill in with the previous volumes.

There may be a certain give and take over the next few months, as new releases on either side continue to make big splashes and also affect the natural ebb and flow of backlist sales — Vampire Knight has one big advantage though: it’s currently serialized in Shojo Beat. Can anyone tell me if FMA is still in the monthly Shonen Jumps? I want to say no, but don’t read the rag so I could be wrong.

That and goth/emo teens really seem to like the vampires, any vampires. Been true for decades now, can’t say why. (I can guess… I might make some really good guesses, but since I don’t share the fascination I can’t say for sure)

This little revolution doesn’t mean a whole lot, necessarily, but I bring it up for two reasons: First, my Midlist 500 uses (or I should say, excludes) the top 5 series, so this changes the whole complexion of one of my charts
& Second, I wanted to demonstrate what the numbers could be used for, and why.

Sure, I just throw ‘em out there, hundreds every week. With over 1200 volumes coming up in what I’ll conservatively say are 200 individual series and about that many one-shots, there’s no way I could do this kind of analysis for every title, or series, or publisher.

But I could,

and so can you.

Have fun with the numbers, folks. That’s what they’re there for.

##

One more update for you, Glenn, to bring your books up-to-date with the new list:

251. ↑1 (252) : Kanji de Manga 1 - Japanime’s Manga University, Jan 2005 [120.9] ::
470. ↑190 (660) : Manga Moods - Japanime’s Manga University, Mar 2006 [40.5] ::
733. ↓-74 (659) : Kanji de Manga vols 1-3 box set - Japanime’s Manga University, Nov 2006 [15] ::
793. ↓-225 (568) : Manga Cookbook - Japanime’s Manga University, Aug 2007 [11.2] ::
1208. ↑62 (1270) : Kanji de Manga Kana de Manga Special Edition: Japanese Sound FX - Japanime’s Manga University, April 2007 [0.6] ::

Manga Moods moves up into the top 500 this week. What do you know, Bookscan’s numbers might even be right.