Fun With Numbers: Do Elements of Ecchi Content/Fanservice Provide a Boost for Blockbuster Shows?

One of the things I love about giving panels/talks at cons is the Q&A component. Sure, every time there’s at least one guy whose question that’s a recitation essay, but you also get a lot of stimulation; observations beyond your field of view from people outside your normal circles of interaction. One of the best questions I got during Ohayocon came after the Myth of Fanservice panel talking about the results of this article. I don’t remember the exact wording, but it was to the effect of “I get from this presentation that shallow fanservice doesn’t sell, but fanservice that’s a component of an otherwise good show actually annoys me more, because it makes it harder to recommend shows. What about fanservice as a minor component of a show, rather than the central one?” At the time, I replied that it was an interesting question, but that I hadn’t tested it and couldn’t come up with a way off the top of my head.

After I thought about it for a while, I realized there’s actually a fairly intuitive way of getting at this question. Since exactly how fanservice/ecchi elements a show has to include before being a fanservice/ecchi show varies from person to person, it was very possible that one could get a snapshot of that spectrum by looking at how two separate databases with varying standards classified a show. As fate would have it, myanimelist (ecchi genretag) and animenewsnetwork (ecchi+fanservice themetags)* classify shows as ecchi in ways that are different enough that one can split shows from my original black/white sample into 3 meaningful categories.**

1. No/Minimal Ecchi (Not tagged under either system)

2. Ambiguously Ecchi (Tagged under one system, but not the other)

3. Unambiguously Ecchi (Tagged under both systems)

Theoretically, if a show is really heavy on the fanservice, it’ll end up being in category 3, and if it’s got naked men wrestling behind one-way glass, it’ll end up being in category 1. If there’s room to dispute how much fanservice a show has and/or how central it is, it’ll more likely end up in category 2. And by comparing those 3 categories, we should be able to get some idea of how much fanservice as a component does for shows with other notable selling points. The breakdowns for the categories can be found on this doc, and analysis can be found after the break.

The full sample contained a total of 761 shows, with 394 in the late-DVD era (2005-2008) and 367 in the Blu-Ray era (2009-2012). The average show sold 4566 disks per volume over the 8-year period, 3552 in the late-DVD era, and 5654 in the Blu-Ray era.

The Minimal Ecchi sample mirrors that full sample very closely. The overall average of the 608 non-ecchi shows was 4424 disks per volume, with 337 late-DVD shows selling 3590 and 271 BD-era shows selling 5460. Since the sample makes up the majority of total shows, it’s not surprisingly that they behave very similarly, though it’s worth nothing that pulling out both tiers of ecchi show has little impact on the average.

The 83-show Unambiguously Ecchi sample shows less variance between eras. The overall average of 4995 disks per volume is split between 29 shows averaging 4560 from 2005-2008 and 54 shows averaging 5229 from 2009-2012. Though the modern era has, to some degree, raised all boats, unambiguously ecchi shows have benefitted from this phenomenon much less than the field has.

The 70-show Ambiguously Ecchi sample, arguably the star of this column, posted an overall 5214 disks per volume average. This has very little to do with its performance in the DVD era, where 28 shows averaged 1972 units per volume, and very much to do with the Blu-Ray era, where 42 shows averaged a backbreaking 7453 units per volume. 

It would be easy to counter that 4 of the shows largely responsible for this turnaround, namely Bakemonogatari, Sword Art Online, and both seasons of Horizon on the Middle of Nowhere, are very much not about that. All are primarily shows built around a real, character-driven plot.*** That said, while I was primarily watching Horizon for Marshawn Lynch reasons and carry a pair of 8oz gloves around cons primarily to throw down with those who talk smack about it, it’d be a significant stretch to say that the show tried to minimize the amount of fanservice it included. Ditto for the other two, albiet to a lesser extent. The point of this category was to include shows using some degree of fanservice to boost an otherwise solid show, and by all accounts, the ceiling of such an approach is pretty high.

So yes, there’s probably some merit to the idea that some ecchi content can help grease the skids if you’re trying to hit it big. If I had to posit a theory as to why, I would guess it had something to do with characterization. It is possible that the same attributes that make characters compelling to watch also make fanservice scenes featuring them more appealing. If it’s merely a phenomenon that forces people to try to write better scripts, I personally can’t complain. But it’s almost certainly true that scenes like that are a reason why anime often faces disinterest in many circles. I don’t have a strong enough moral compass to necessarily judge what it means in the long term, but I can say that fanservice as a noncentral element is probably here to stay for the immediate future.

*Yeah, I had to combine the two. Taken as a set, they are consistently applied between seasons.

**Since I just published a note about how inconsistent classifications for sequels can be a problem, it’s only fair to mention that myanimelist inconsistently classifies Nogizaka Haruka no Himitsu, Koihime Musou, and the Monogatari series, while ANN inconsistently classifies Medaka Box. The tags are generally large enough that this error rate doesn’t necessarily disqualify them, and there are somewhat reasonable arguments as to why those series are classified differently, but it’s worth being up front about that fact regardless.

***Sword Art Online executed that plot very poorly and had a crippling case of Eden of the East disease, but it did have one of the most compelling first episodes of the year. A lot of people who aren’t me agree that it has a touching story, which is more relevant to the discussion at hand than my personal opinion on it.

Advertisements

7 thoughts on “Fun With Numbers: Do Elements of Ecchi Content/Fanservice Provide a Boost for Blockbuster Shows?

  1. The topic of whether fan service benefits anime or not is one that comes up pretty often so it’s interesting to see some straight statistics. What I wouldn’t mind knowing is if there’s a difference in success between shows that feature ecchi as the main draw (eg. To Love-Ru), shows with ecchi that focus on story (eg. Sword Art Online) and shows focusing on story with very little ecchi involved (eg. Death Note). Of course it would be very difficult to properly categorize shows (and that’s without the question of whether slice of life shows can count as “story-focused”) and would take a lot of work, so it probably isn’t worthwhile >.<

    • In practice, the categories as they ended up in this analysis ended up producing a similar split, which is a good part of why I used this method. At least, Death Note was cat 1, SAO was cat 2 and To Love was cat 3. The general result here is that all 3 categories succeed in the modern era to some extent, though side-element-ecchi shows seem to have the highest rate of hit production and hence the highest average sales.

      A common trait about the shows you named is that they were all 6k+ sellers. Most anime don’t do nearly that well, regardless of what genre they’re in. And in general, there’s a wide distribution of sales averages for any genre big enough to have its own tag on ann. Their slice-of-life tag checks out well enough to be usable, so I will probably be making use of it at some point.

      In general, I find crowdsourcing to be more reliable than by-eye categorization anyway (more opinions=more likely to be near the mean). The key is making sure those crowdsourced tags aren’t all being applied by a single person the way ann’s moe tag seems to be.

  2. These are some very interesting numbers and I’m glad you took the time research this. The problem with many shows is that there are certain scenes that don’t further the story, aren’t used for humor or add to the show in a meaningful way. Bath scenes, awkward chest shots, or the fabled and omnipresent beach episode are all part of many of the best and highest selling anime even if the show wouldn’t normally be classified as ecchi. (Look at Oda’s female character designs as a simple example of this. Is One Piece ecchi? No in general it isn’t, but all of his female characters have abnormally large breasts and many wear bikinis.) Evangelion for example is known for it’s story and philosophy, but it also had the female members of the cast wearing bath towels or less in many episodes that seem unnecessary at times. This is my biggest problem with the industry, that studios don’t always trust that their work is enough to make 20k+ shows and feel like they have to add a little something extra.

    Also perhaps some observations come from those that you might have considered in the past your usual circles of interaction, you never know 😉

  3. Just a question…. It’s about NouCome though. How were LN sales before the anime aired, and how were they after the anime aired? I read somewhere that the LNs didn’t get a boost, but is this true? If so, is the anime considered a total flop?

    • http://myanimelist.net/forum/?topicid=586299
      v4: 8548 volumes in 2 weeks (April, after the anime was announced)
      v5: Did not chart
      http://myanimelist.net/forum/?topicid=676921
      v6: 12,211 volumes in 2 weeks (October, right before e1 of anime)
      http://myanimelist.net/forum/?topicid=905651
      v7: 20,253 volumes in 2 weeks (January, after anime ended)

      The LNs got a pretty clear boost, if not a huge one like Kore wa Zombie or Campione. At a minimum, that’s 8000 copies per volume, and it’s probably more (the charts don’t show us the tail, and we don’t have any barometer to measure it). It’s hard to take that 996 disk average and spin it into a net-profit story; even if the tail were double what we saw, that’s still only like 16000*500=8 million yen in LN gross, or 24 million yen over the 3 volumes that will presumably come out this year (based on previous release frequency). Relative to a super-lowball 100 million yen budget (10 million yen for 10 episodes), the LNs would have to continue for another 3-4 years (possible) and sustain the boost over that entire time (not impossible) to be in the break-even discussion.

  4. Pingback: ICYMI: Highlight Posts of 2014 | Animetics

  5. Pingback: Active Engagement Through Timed Comment Breakdowns: GJ-bu | Animetics

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s