Fun With Numbers: Attack on Titan’s US Release Sold 11,748 Copies on Week 1

OpusData BD data for the week ending in June 8th is now publicly avaialble, and since the BD/DVD combo packs are counted as BD releases (as they were for the Fairy Tail movie), that means we have sales figures for Attack on Titan’s part 1 release. The release, which performed the best on amazon out of any US release I’ve tracked in the past 6 months. Assuming the BD/DVD combo packs were counted the same way this company has previously done, Attack on Titan was the 14th best release among US BD content on this week, selling 11,748 copies.


This actually represents a severe underestimation for my rough-fit model, which pegged it at around 30,000 copies. While it underestimated DBZ season 3’s first week by 20%, this represents an overestimate of more than 60%. It’s possible that the series’ airing on network TV during solicitation made it more of an amazon-heavy title. Most likely, though, this underscores the fact that the model is still rough and prone to error. I’ll be tweaking the model a bit in the next couple of days to see if I can get a model that less severely overestimates this series while still covering the other two points of confirmed data (DBZ s3 and Aria the Natural). I should have at least one more chance to test that model when the series’ second part comes out in late September.

11 thoughts on “Fun With Numbers: Attack on Titan’s US Release Sold 11,748 Copies on Week 1

  1. 10k+ being a high water mark for an US license is really depressing. Listening to industry people lamenting the past in ANNCast, they were moving million VHS with Pokemon and hits are in hundred thousand range.

    • Long tails might mitigate some of that – I wish this hadn’t scrambled my model for another few months so that I could actually check that. Nowadays it seems like 100k lifetime is only a realistic figure for Ghibli movies. I do wonder how many of those hits they were talking about were TV series (as opposed to movies), though. Feels like it can’t have been *that* many (AS/Toonami titles and maybe 3 or 4 others).

      • In his ANNCast interview, Chad Kime talked about how 10k+ was the break even line in those heady days. (Which reminds me, didn’t Funimation tweeted a similar sentiment a few months ago?)

        Here’s my liner notes for that interview

        Chad Kime
        Head of Marketing, Geneon

        * A year and half at Books Nippon
        * 11 years at Geneon, originally Pioneer
        * Help Anime Insider start off with 25,000$ ad buy per issue
        * 2002-3 20% growth YOY sales
        * After sold off by Pioneer, 1/3-1/2 of sales returned

        Break-even line
        Trigun – 6000
        Vandread – 20000 (did not get the title)
        Demon Lord Dante – 30k
        “At that time 10k to break even

        Units (00:33:00)
        Pokemon – (expect 20000~30000, breakeven much lower) first month 1 million units; 8-10 million units, first 12 volumes; Pokemon still sold its 10k units in DVD
        FMA sold 10x more than Heat Guy J; cost Geneon same amount to license.
        No Go Nagai title ever sold more than 5000 units
        Psybuster – less than 100 units
        Rumiko Takahashi Anthology – 2 digits of sales; first volume sold ~300; second volume 80-90
        Samurai Champloo – 40-50k (10% increase after CN premiere); outsold Naruto on volume-by-volume for first year
        Rozen Maiden – 4~5k

        Viewitiful Joe – did not hit expectations
        Lupin III – sales 1/2 of expectation
        Tetsujin 28 – can’t sell; megabomb
        “Harem sales tend to sell”
        Dear S – sold okay; close to break even
        Hajime no Ippo – bombed; not even close to make money back
        Sugar – bombed, got nailed

        Amazing Nurse Nanako – made money
        Bottle Fairy – did not do well
        Cardcaptor Sakura – 15k subbed, 2k dubbed (expected 5k subbed, 100k dubbed); subbed fell off a cliff after Vol.4
        Fushigi Yugi – very very well at $200 box; made money before DVD released;
        Gate Keepers – didn’t do well
        Gungrave – Did okay
        Haibane Renmei – better than expected
        Ikki Tousen – made money
        Mao-chan – bombed
        Master Keaton – did horribly
        Mermaid Forest – Did okay
        Paranoia Agent – Did well
        Sherlock Hound – Did okay; didn’t come close to making money
        Dog of Flanders – “Target make us break even”
        Sakura Wars: The Movie – bombed
        Patlabor – bombed
        Submarine 707R – megabomb
        Ultra Maniac – did okay but didn’t break even
        Zipang – bombed
        When They Cry – Don’t do very well

        • Iirc, the Funi tweet was “a lot more than 3000”, which is a pretty broad range.

          “FMA sold 10x more than Heat Guy J; cost Geneon same amount to license.”

          Wasn’t FMA originally a Funmation license that every one was bidding on?

          The idea that Heat Guy J was the same price seems a bit odd. Even for per-episode, but especially considering the difference in cours. Maybe that was just their bid (one that Funi beat out).

          Doesn’t seem like many TV releases there hit 100k+ averages. Samurai Champloo being 50k after the TV bump doesn’t exactly give high odds to other contenders.

          • Geneon definitely expected (and paid dearly for) far more 100k+ series than they actually got. Listening to the interview again, Chad did only explicitly state Pokemon as their sole 100k+ hit.

            Maybe I might had overstated the number of 100k+ moon shots. Given that Champloo hit 50k, those numbers didn’t seem implausible as it is now. So what series would be the suspects that hit that mark? Eva, Pokemon, and DBZ seems like a lock. Maybe Cowboy Bebop too, given Champloo’s numbers. FMA? FLCL? Maybe initial releases of shonen titles like Naruto? Then there’re the 4Kids franchises. Even if most of their income came from merchandising, some part of their one time $200m annual revenue must have came from home video sales, right? Those titles looked plausible too.

            • Pokemon, DBZ, and Naruto I’d bet on, and probably YGO. I don’t know if any of the other 4kids titles really ended up in the latter’s league.

              Given how popular Champloo has continued to be in R1 (box rereleases of the series were topping amazon as recently as last Christmas, and it’s in the 10,000th-20,000th range even now), that 50k mark would make me a little wary of locking in on Eva/CB/FMA, though I wouldn’t be surprised if any of them did.

              FLCL might count if you count it as a TV series. It was only a 3-part release, though, a decent amount less expensive overall than regular TV series.

  2. Pingback: Fun With Numbers: Appleseed Alpha Sold 11,093 R1 BDs in Week 1 | Animetics

  3. Pingback: Fun With Numbers: Updated US Amazon Formulae | Animetics

  4. Pingback: Fun With Numbers: Attack on Titan Part 2 Sold 10,677 Copies on Week 1 | Animetics

  5. Pingback: Fun With Numbers: April 2015 US Amazon Data (Initial Numbers) | Animetics

  6. Pingback: The Attack on Titan TV Series Has Sold at Least 200,000 Copies in the U.S. | Animetics

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s