Fun With Numbers: Print Boosts’ Effect on Sequel Odds

It’s been well-established that an anime adaptation of a manga or light novel can be a huge boon to the source material. What’s a bit less obvious is whether or not boosting print material can fuel the production of a second season of said anime.There are at least a few reasons why it shouldn’t; for a manga get adaptation boosts primarily from the first season – afterwards, their sales tend to plateau or drop off (even if the series does get a second season). It makes sense on an intuitive level that there would be some sort of diminishing returns on subsequent seasons of anime; sequels tend to sell between 0% and 50% fewer disks, and people don’t tend to start watching anime from the second season onward. But whether or not those diminishing returns carry over to print sales, and if so to what extent, is a somewhat separate question.

In this post, I’ll be exploring that question, comparing the rate of shows getting sequels with and without print sales boosts over different ranges of disk sales, to get an idea of whether or not print sales boosts actually “matter” towards a show’s sequel odds.*

For the purposes of this column, a “significant” print boost is one where the sales after the first season of the anime has been available for 2 or more months show a 20% increase compared to the sales prior to the airing of the first season of anime (both figures averaged over 2 volumes if available), *or* a series that charts for the first-time post anime at a level that would have beaten at least one threshold for weeks where that series failed to chart by at least 1000 copies. If you want to check why a series is classified one way or the other, or you’re curious as to how exactly an individual adaptation from this period affected the sales of its source, you can find all the data used here in the following posts:

2010 Manga: https://animetics.net/2014/06/26/fun-with-numbers-anime-as-manga-advertisments-in-2010/

2011-2012 Manga: https://animetics.net/2014/07/08/fun-with-numbers-two-week-revisions-of-the-2011-2012-manga-adaptation-data/

2010 Light Novels: https://animetics.net/2014/06/27/fun-with-numbers-anime-as-light-novel-advertisments-in-2010/

2011 Light Novels: https://animetics.net/2014/01/15/fun-with-numbers-anime-as-light-novel-adapatations-in-2011/

2012 Light Novels: https://animetics.net/2014/01/15/fun-with-numbers-anime-as-light-novel-advertisments-in-2012/

Initially, the sample I was analyzing contained 136 print adaptations made over 2010-2012, but I had to eliminate several of them from the final analysis. Specifically, I excluded from analysis series that had a definite split cour sequel (less than a 1 year gap between season airdates) and those that did not have releases both before and after the anime aired. Either of these conditions is represented by a dash on the source spreadsheet. The final sample contains 113 show-source combinations. I counted shows as having gotten a sequel if they got a second TV season that aired one year or more after the original, or if they got a movie.

After splitting the shows between those that did and did not get print boosts and sequels, respectively, I further split them up into several subcategories based on disk sales. My basic reasoning in creating said follows from the following 2 assumptions:

1) Given other sources of revenue, a disk average of 3000 copies is generally enough for a series to be considered break-even.

2) Sequels tend to sell an average of 20% fewer disks, with a 1-sigma range of 0%-50%.

This led to the creation of the following categories:

sub-0% (Would require an increased average to profit via disk sales) [avg<3000]

0%-20% (Would require a a better-than-average, but within 1σ, dropoff in disk sales) [3000<avg<3750]

20%-50% (Would require a below-average dropoff in disk sales) [3750<avg<6000]

50%-10k (Could survive a 50% dropoff, but not necessarily 100% safe) [6000<avg<10,000]

10k+ (Averaged over 10k in disk sales, money almost certainly not an issue here) [avg>10,000]

Dividing up the shows into these categories by disk and print sales, and measuring the frequency of sequels in each group led to the following results:

print-sequel

Essentially, at lower levels of disk sales (20% or less dropoff required), everything that did get a sequel got some sort of significant print boost, though the general rate of continuation is fairly low. The series in this range that got sequels were Bakuman, Chihayafuru, Space Brothers, Kore wa Zombie Desu Ka, and Maken-ki. When you get to series with really high disk sales, how the print source did ceases to matter and the sequel odds are just very high regardless. It should be noted, though, that only 5 series out of 36 total with 3750 or higher averages did not get some sort of significant print boost (these 5 were Mouretsu Pirates, Seikon no Qwaser, Jojo’s Bizarre Adventure, Ladies Versus Butlers, and Freezing).

What’s a little odd is the 20%-50% dropoff category. If they were getting print boosts similar to those of series with lesser disk totals, one would expect them to get sequels more frequently. The answer to that quandry comes from the fact that not all print boosts are created equal; a 30% boost to a light novel that comes out 3 or fewer times a year that bumps it up from 10,000 copies to 13,000 copies is not economically equivalent to a 30% boost to a a mainsteam, 5-volumes-per-year weekly manga that takes it from under 300,000 copies to almost 400,000. This particular interval contained a bunch of series like Joshiraku, Mayoi Neko Overrun, and Humanity Has Declined; series which, while boosted, don’t carry enough raw volume in their totals to move the sequel needle. For comparison, Maken-ki and Kore wa Zombie were both selling at least 40,000 copies post-bump.

Final word seems to be that high-volume print boosts matter somewhat in determining whether low-selling series end up with a season 2, though they’re nowhere near as important as disks in most cases.

*They almost certainly matter in determining whether a manga publisher wants to be a part of production committees and offering up new sources for adaptation long-term, just not necessarily for a particular show being continued.

7 thoughts on “Fun With Numbers: Print Boosts’ Effect on Sequel Odds

  1. If I read the article correctly, one of your key finding is that source sales boosts are irrelevant to sequel chances for anime best sellers. Is the inverse also true? That is, the anime flopping is irrelevant as long as source saw very high boosts?

    If so, aren’t these factors effectively mutually independent in-so-far as to the chances of an anime adaption getting a sequel or not?

    • The inverse is not also true – Giant Killing (130k->250k) and Kamisama Hajimemashita (40k->120k) picked up Chihayafuru-tier (220k->300k) boosts volume wise, are still running, and still have no sequel, likely due to poor disk sales. By contrast, the only things without sequels in the 6k-and-above sample are Accel World/HSotD/IxB SS, where non-monetary reasons seem to be the chief cause (insufficient source material for Accel/HSotD, an occupied staff for IxB SS).

      Not sure what you’re getting at with the mutual independence thing – yes, a boost and disk sales are very weakly correlated for manga adaptations (though it’s a bit more robust for LNs). Do you mean that they can’t both affect sequel odds for the same show? Zombie/Maken-ki are pretty solid counterexamples to that, but I’m not sure exactly what you’re getting at.

  2. Great. Now can you do an analysis of light novels/manga which ended shortly after their anime adaptations ended (probably by defining shortly as one year or less after their anime’s end)? Basically, of the manga/LNs which ended shortly after their anime, which ones had sales boosts, and which ones did not? Based on your article, Nichibros was most likely not cancelled and the information that it ended on the author’s note is correct because Nichibros received a sales boost. It could be interesting to speculate which series were most definitely cancelled and which ones were simply coming to an end when the anime came out.

    • There are too many Seikon no Qwaser/Wandering Son/Mysterious Girlfriend X/Upotte types that never chart but continue for years after the anime for the evidence in favor of a “bad sales cancellation” theory to be particularly significant, but I suppose it would be worth it, if only to prove the point definitively.

      I suspect the more relevant number to check against would be magazine sales number versus volume, taking the anime question out of it. Once an anime airs, it’s pretty much been paid for. The questions then become “Is this series selling more than what we would, on average, expect a series by a new author to do given the popularity of the magazine?” and “Are we Shonen Sunday?”. Your probable cases for cancellation are when the answers to both are no. It’s why Mx0 got cancelled after 10 volumes in Weekly Shonen Jump with constant top-10 appearances while WS ran 15 volumes in Comic Beam with no chart appearances whatsoever; one magazine has a serialization of 2.8 million, the other a serialization of 25,000.

      Source: http://www.imagebam.com/image/aa55ce327992839 (via Mangahelpers)

      In any event, I’d probably look at both in an analysis of this sort of thing. It’s interesting enough to do, so it’ll probably happen before the god-awful project redoing the sequel equation is going to be.

  3. Pingback: Fun With Numbers: The Non-Cancellation of Popular Manga Post-Anime | Animetics

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

Connecting to %s