Fun With Numbers: Print Sales Bumps and Poking at the Demand Curve

Posted on June 1, 2014 by torisunanohokori

It’s fairly common for manga or light novels getting an anime adaptation to receive some sort of boost in sales after the anime airs. However, both the presence or size of those bumps varies widely. These bumps represent an interesting opportunity for study, since they represent (potentially) an alternative indicator of both the financial impact of anime, as well as a look at the broader-scale demand curve for franchise-related goods. Not everybody can or will easily pay 30,000 yen for a full special-feature-laden set of disks, but such casual fans could still have a big impact on a series if the manga is within their price range.

The real fascinating part of this, though, is that casual fans need not support at all. While one almost certainly has to watch a show to be willing to buy the disks or even the manga, the reverse need not apply. The fact that the relation between disk averages and print boosts is so fragmentary implies a potentially similar disconnect between print bumps and total interest generated by a show.

In theory, sales should be better represented by casual indicators of popularity as costs get lower. In practice, the statistics are pretty garbled, though they do offer a hint as to which sorts of series may end up with bumps at the end of the day. To attempt to better understand the junk described above, I broke down 4 categories potentially indicating no-cost and low-cost popularity, and compared their ratings with the print-bump successes of series which got anime adaptations in Summer and Fall of 2013. Note that while I originally used Torne rankings in the earlier analyses, I discarded the data because of how incomplete they were.

For the purposes of this article, a “significant boost” is one where a series experiences a 20% jump in sales or charts for the first time after the anime. All figures are for a volume’s first 2 weeks of sales, calculated with the average of the 2 most immediate before and after volumes, if 2 volumes of data are available in each case. Be aware that this is not a comprehensive measure of which series got boosts, just one intended to at least catch the biggest ones. 1/3 of all series got such a visible bump, meaning the null-hypothesis accuracy rate these indicators need to beat is 66.6%~67%.

In addition to the previous top 5/10/15 tests used for v1 disk sales, I’ve also included histograms of how successful series were scattered across the indicator. Ideally, they would all be 60th percentile or better, but reality isn’t quite that nice.

Continue reading →

Animetics @ Colossalcon 2014

Posted on May 30, 2014 by torisunanohokori

In news that will probably not matter a ton for people not going to Colossalcon next weekend, June 3-6: we’ll be there in Sandusky, doing a couple of panels (schedule’s here). If you happen to be going and have some time at one of the following spots, feel free to come by and check us out at any of the following times.

Intro to Mecha (3pm Friday)

-Sam, Will and I walk through a number of mecha series we’ve picked out for people who want to get more into the mecha genre but aren’t sure where to start. The focus is either on series that are either unique, flavorful alternative takes on the genre or series that run the formula exceptionally well.

Anime versus Manga (5pm Friday)

-Frame-to-video comparisons of manga and their anime adaptations to show how similar and/or different they can be. We’ll try not to be belligerent in making the point about how 4-koma series are much harder to adapt than winners of major awards, but it’s sort of hard to avoid.

Where’s my Sequel (5pm Saturday)

-We talk about the various income streams (disks, print volumes, merchandise, etc.) that determine whether or not anime are commercially successful, and hence are likely to get sequels. The ultimate answer is disks uber alles, but there are a lot of nuances there that I’ve been able to look at in more detail since posting the super-rough sequel probability equation a year or so ago. Depending on how my writeup goes, I might end up incorporating the results of the casual-interest stuff that’s been boring you for the past 2 months or so.*

Golden Age of Late Night (9pm Saturday)

-A history lesson on how the cutting edge of anime made the transition from TV to OVAs to Late-Night TV, covering all notable series (Dallos, Eva, TWHE) playing a part in that progression and why they were relevant to it.

*Right now, all the numbers I need are on spreadsheets, but I need to sort and plot them to really show what’s going on. Short version is that it’s complicated and cloaked in a thick fog, but there are two indicators that do seem to cut through it just a bit, in that you can build histograms with visible slants and not have most of the series with boosts not in the bottom 10. I’m already dicking around to see if I can find a second indicator that can be used to shave off the wrong guesses for either of those, which means I’ll probably be pulling data for Spring and Winter of the same year to test and see if that holds water.

Fun With Numbers: June 2014 US Amazon Data (Initial Numbers)

Posted on May 28, 2014 by torisunanohokori

Here’s the usual infodump post of initial numbers for the series whose June amazon ranks I’ll be tracking, gathered from this list of upcoming releases. As before, May data is still being collected and will be posted when that’s done in about a week.

Continue reading →

Fun With Numbers: Anime as Light Novel Advertisments in 2013

Posted on May 25, 2014 by torisunanohokori

While light novels work a bit differently from manga in several key ways (stronger second-week showings, lower thresholds, etc.), they similarly often see big boosts after and presumably due to from anime adaptations. I collected the light novel sales history of the series to get anime adaptations in 2013 on this doc, and plotted them on the charts below, to illustrate which series did and didn’t get visible boosts.

This post doesn’t cover series with no post-airing releases (Maoyu, Uchoten Kazoku) or no pre-airing releases (Free/High Speed).

Continue reading →

Fun With Numbers: Anime as Manga Advertisments in 2013

Posted on May 23, 2014 by torisunanohokori

The commercial impact of anime goes well beyond its disk sales. Manga may sell to more people, but anime is extremely visible, airing on TV (albeit often late at night) and propagating around the internet at a very rapid pace. This visibility very often can lead to an increased strength of the franchise in general, propping up sales of print material, figures, and any various other related goods. Sometimes, anyway. 2013 was no exception, and saw a number of manga adaptations have anywhere from minimal to explosive effects on the sales of their source material.

I collected the manga sales history, including thresholds for series which charted sporadically, on this doc, and plotted it below. Note that these sales are not total, but the total number reported in a roughly fixed time period. Comparing sales tail length is a whole other issue, and I’m trying as much as possible to compare like figures.

One important difference from similar breakdowns of 2011/2012 series is that here I’ve opted to use the total sales from a series’ first 2 weeks of release (the highest reported total in that time interval), to attempt to minimize the effects of a bad split in creating artificial variations. It’s still an issue either way, but the difference between 9 and 14 days is a lot less than the difference between 2 and 7 days.

Two important series-specific notes prior to the plots. First, Maoyu is plotted here, in the manga section, because the manga charts more consistently than the light novel did and, more importantly, has available data from both before and after the anime aired (the LN ended just prior to 2013). Second, I can’t parse impact for series that don’t have at least one volume which released after the anime began to air. I thus will not be covering Servant x Service here, though there is data available. I will cover it in an addendum post come September when volume 4 has been out for 2 weeks.

Continue reading →

Fun With Numbers: (Admittedly Arbitrary) Qualifications for Significant Print Sales Boosts

Posted on May 17, 2014 by torisunanohokori

I’m about to dive headfirst into the manga/LN print data I’ve gathered and check how much the series that got adaptations benefited from them. It’ll take a week or so, but when I’m done, I’ll have charts like the 2011–2012 ones and a wealth of organized first-two-weeks data to cross-reference with the casual-indicators checks I’ve been doing, and I should be able to more or less finish that project. Before I do that, though, I just want to outline some guidelines I’ll be using for said project before actually gathering the 2013 print data.

One of the key problems I bumped into early on was how to classify boosts as big/small/significant/what have you. It’s a complicated issue that in practice goes far beyond the Oricon charts, but for the purposes of this project, I’m going to focus primarily on chart-visible boosts. These get split into two categories; one, the type where a series makes the charts for the first time post anime, and two, the type where a series sees a sharp uptick in post-release sales. After looking into the previous year’s record, I decided on different criteria for each.

I’m going to consider first-time charting significant if a previous volume was released under a threshold that would not have prevented the first-time-on-charts total from charting. In other words, most series charting for the first time will make this cut, but it’ll disqualify series like Freezing that charted under these ridiculously friendly circumstances:

I’ll consider boosts for series consistently on the charts significant if a series sees a 20% uptick in first two weeks of release sales from the immediate pre-anime period to the immediate post-anime period (averaged over 2 volumes, if possible). Even if the effects of anime adaptations extend well beyond this limited scope, I hope this will at least be indicative of which series made the very best of their source material boosts.

Also, I’m going to avoid analyzing series that lack post-anime releases (Servant x Service) or pre-anime releases (High Speed). For obvious reasons, both before and after data are necessary to identify a boost.

Weekly Light Novel Sales Charts for 2014

Posted on May 14, 2014 by torisunanohokori

These are the weekly light novel sales charts for the first four months in 2014, via myanimelist news, continued from the 2013 post. I’ll be doing one of these updates every 4 months; if you want more recent data, there’s plenty of places where charts are available (eg. ann, the mal news forum I get them from).

Edit: Added data through August 2014.

Edit 2: Added data through December 2014.

Continue reading →

Weekly Manga Sales Charts for 2014

Posted on May 14, 2014 by torisunanohokori

These are the weekly manga sales charts for the first four months in 2014, via myanimelist news, continued from the 2013 post. I’ll be doing one of these updates every 4 months; if you want more recent data, there’s plenty of places where charts are available (eg. ann, the mal news forum I get them from).

Edit: Added data through August 2014.

Edit 2: Added data through December 2014.

Continue reading →

Fun With Numbers: Myanimelist Stats’ Correspondence with the Summer 2013 Disk Over/Unders

Posted on May 14, 2014 by torisunanohokori

Quick recap: I’m taking a look at various no-cost indicators of popularity for anime and their related goods. First, I’m checking how well they correspond to disk sales by checking whether different applications of that statistic beat the null “every v1 will sell less than 4000 disks” accuracy criterion for a given season (66% for Fall, 59% for Summer). Later I’ll check how well these indicators corresponded to boosts in manga/LN source popularity (for works that were originally LN/manga), to contrast their predictive abilities at both high-cost and low-cost levels of interest.

This entry covers the myanimelist rankings and popularity for the 34 Summer 2013 shows I’ve been looking at, with the title tweaked to reflect that these are the series rankings from several months after the series have ended. It shouldn’t be a huge factor; positions don’t shift radically that often, especially after a series is done airing (more research on that forthcoming). The data I gathered can be found here, and the results are shown below.

Compared to Fall, Summer showed a relatively weak performance by the mal metrics. The popularity component still does alright, but the the rankings consistently underperform the accuracy of the null hypothesis. This is more in line with what I would have expected going in; there’s a lot of evidence to suggest that casual audiences aren’t that different inside and outside of Japan, but there’s a very relevant degree of separation between casual audiences and disk-buying ones. Too, a series’ aggregate score being less important to sales is unsurprising, as the series profile of “publicly lambasted but successfully appealing to its target audience” is exceedingly common in any medium.*

*It’s a principle potentially less applicable with nico rankings because they represent the subset of people watching via nico streams, which can be a qualitatively different audience from one that DVRs episodes and/or stream through other services.

Fun With Numbers: Nico Stream Rankings Predicting Summer 2013 Disk Over/Unders

Posted on May 14, 2014 by torisunanohokori

Nico stream rankings are often discussed in the context of early-season indicators of disk sales, and my analysis of the Fall 2013 shows for which those numbers were available did show a bunch of different versions of the data proving more accurate than the null hypothesis. Adding another season should help control for any peculiarities with the Fall season.

The accuracy rates for the top 5/10/15 models using only the 25 series Summer series with Nico stream data available, split into 3 averaged samples corresponding to the beginning, middle, and end (labeled 9-12, but occasionally 9-13 or 9-10) of the series. The individual rank data was compiled here, and the sorted version can be found here. The bare-bones summary of the results is below (green means an accuracy greater than that of the null hypothesis).

Note: the chart format groups by period, rather than by metric, as the Fall chart did.

That’s a lot of green, corresponding to a fairly useful indicator. While there are notably some top-tier performances that failed to lead to disk sales (Watamote was top 5 in half the metrics, top 10 in all of them), the general results point to an indicator that can be fairly discussed early on as a broad-basis indicator of disk-buyer popularity.

Peculiarly, the max rating percentage becomes a less accurate indicator for the later periods. This appears to be due to the rise of specific shows (Genshiken Nidaime, Rozen Maiden’s reboot) that lost viewers while maintaining the number of viewers giving the max scores; most likely, many of the people who had middling opinions of them checked out after the first month, leading to somewhat inflated scores.