There exist a question asking about the most connected seiyuu by six degrees. As innocent of a question it may seem, but the foundations the question is based on is flawed and unreasonablely scoped.

Kevin Bacon himself thinks this game was a joke in his TEDx talk: https://youtu.be/n9u-TITxwoM

He didn't invent the game, but it is permanently associated with him thanks to pop culture.

But what's the problem here? The problem is that we are using a fundamentally speculative assumption to pose another about a similar topic. This will only create more misconception if the basis is not first scrutinized.

The 6 Degrees of Separation from Kevin Bacon Theory is based on a well known theory called the “small world theory,” a longstanding serious look at the degrees of separation between people. Some look for fun, some like Stanley Milgram looked to explain how events like the atrocities of WWII could happen given how interconnected we all are.

Many “small world” studies have been done throughout the years, from radio inventor Guglielmo Marconi’s thoughts on the subject in the early 1900’s (he guessed 5), to experimental social physiologist Stanley Milgram’s small world experiments in 1967 (which showed 6 degrees), to Microsoft’s 2008 study (which showed average degree of separation is 6.6), and more.

Although most studies have showed the average degree of separation is about 6 (but not exactly), the Microsoft study showed some people are connected by up to 29 degrees. Others, like a person living alone in the jungle, could be completely unconnected. While it is true that Kevin Bacon is connected to a lot of actors, but the same can be said about other actions that have been around for a while. Popularity and fame will get you a ways to to conbecting people, but it's not an end all be all, especially in this age of the internet. Small world studies typically look at who we know in any way, but each study is different.

With Kevin Bacon game, there are specific rules to narrow the scope, such as one must have worked with Kevin Bacon on a professional or personal level to count.

However, if we count just people we know in any way (including social media) there can be an argument that many people have less than 6 degree today.

To avoid the going on a long winded rant about the finer details of the theory, see this video for a summary: https://www.youtube.com/watch?v=TcxZSmzPw8k

Such a theory was proposed long ago that everyone on earth is separated by 6 or less acquaintance links apart. This idea was first put forth by Frigyes Karinthy in 1929, and popularized by a 1990 play written by John Guare.

The basic concept (behind the Small World theory) is that due to technological advances in communications and travel, friendship networks could grow larger and span greater distances. In particular, Karinthy believed that the modern world was ‘shrinking’ due to this ever-increasing connectedness of human beings.

The small world and 6 degrees of separation theories have truth to them, but they are not true as absolutes (only as general rule-of-thumbs that speak of averages with outliers excluded).

Instead of saying “everyone has 6 degrees of separation,” it is much more correct to say, “the average person has around 6 degrees or less between them and another average person (including Kevin Bacon)…. and with social media considered, those with similar interests may very well have less on average.”

Thus, the 6 degrees of separation rule is a rule-of-thumb, it is generally true for most of us, but doesn’t apply to outliers who live unconnected from modern society.

Trying to apply the same methodology to the Japanese voice acting industry is not a viable approach, in my opinion. Unlike the Hollywood movie industry the Japanese voice acting industry does not command as much respect, attention, or fame. Those that make careers out of it are few and far between, oftentimes the more successful ones move to other more profitable industries, retire, or just fade away into obscurity. Dude to social media the lines have been ever so blurred and hard to distinguish with reasonable objectivity.

The amount of effort to create an viable and objective answer is unreasonable for the average person. You can ask for all the data you want but if the criteria is too boardly, there will be no end to to comparisons. Do we care about the time period? Connected in terms of what? One assumes their voice actors, but voice actors are many different things, some are singers or idols, others are models, some are also stage or movie actors in their own right.

We had a similar type of problem with list requests, but in this case its more of a reverse list. Instead of questions like "how many have X in Y". We have "who is the most A in Z."

There is no reasonable scope or criteria preventing this question from being overly broad. If we limit it to a certsin period, we have limited data available. We can look at work history based on credits, but voice actors also do networking out sude of their roles? How can we talk factors as such into account?

It us in my belief thatanswers are not an appropriate place to be a plase to do a new small world study and we should not be encouraging questions where the answer may as well be a thesis on a bigger topic.

We need some sort of policy in place to keep questions within reasonable scope or it its founded on a controversial theory or topic. Such a question about voice actors cannot be properly answered without sorting and refercing large amounts of data. Furthermore fact checking such answer for bias is an equally cumbersome ordeal.

So I ask, what can we do about this issue.

  • Given the technical literacy of people who watch anime, if the data is out there, it would be a minor hurdle to query it to product this graph. Presumably, someone(s) has done the basic code for the 6 degrees of Kevin Bacon game, so there hard work is done. Therefore, your assumption that this is too difficult is as flawed as you accuse the Kevin Bacon to be. When mining data like this, vast sums of data make better datasets than small sets. Commented Nov 10, 2018 at 21:27
  • Who will mine this data? Its unreasonable to ask someone to mine vast amounts to answer a question. Given the attention span of the average user its unreasonable to ask them to do such work for next to nothing. Unless the OP volunteers themselves to take on the burden
    – кяαzєя Mod
    Commented Nov 10, 2018 at 22:03
  • Someone here or on the OP mentioned that it's already available one or more websites (Anime Planet, MAL, IMDB, etc). It would not be terribly difficult to build a website to showcase this. Commented Nov 10, 2018 at 22:04
  • We presume it's doable, but rather than do it ourselves we wait for someone else to do it. I do not believe we should allow for the proliferation of such questions with putting guidelines in place. We dont want to make people have to spend time crunching numbers and building websites. We're not a technical recreation site like code golf. It's unreasonable to expect such efforts from the average users. We're a recreation site about Japanese cartoon ans comics. There might be some overlap, but its unreasonable to expect such a level of effort from the average user.
    – кяαzєя Mod
    Commented Nov 10, 2018 at 22:28
  • Such as but is not limited to: if you put forth a question based on flimsy, questionable, or misrepresented theories or ideas. One must first must provide a reasonable scope for the answer. This will be a minimum criteria, anyone that wants to go beyond it are free to do so. If the question requires working on large data sets. The OP should make an effort to provide resources if they can't attempt an answer themselves.
    – кяαzєя Mod
    Commented Nov 10, 2018 at 22:28
  • The question in answerable. Just you don't want to put forth the effort to answer it, doesn't mean it should be discouraged from the site. That's what Stack Overflow does, and you'll find no shortage of people, myself included, that don't go there for that very reason. Commented Nov 10, 2018 at 22:42
  • Every question is answerable, with some level of effort. Just no one wants to put forth the effort to answer it, does mean something should be done to make it more answerable. If less and less people are interested in answering, it's a clear sign that people don't like this type of question. We're not Stack Overflow and Stack Overflow as no influence over us. We're here to discuss the appropriateness of such questions on this site. I'm concerned that such type questions will follow the same path as id-requests.
    – кяαzєя Mod
    Commented Nov 11, 2018 at 0:52
  • Not every question is answerable, and not every question is answerable in the format of a Q&A site. For example, list questions are answerable, but banned across the entire SE eco system. Even if this SE is different than SO, there are users that use multiple SEs, and their mannerisms follow them. I'm on several other SEs that have the same problems as SO, much to my chagrin. Commented Nov 11, 2018 at 0:54
  • Also, are there more than 1 of these types of questions? Commented Nov 11, 2018 at 0:54
  • The similar arguments have been made for id-request as they have be question such as this. It take way more effort to ask such a question and not enough context is given, making the answer put extra work in order to find something that fits. These I do not believe these types of question are good questions without first setting some guidelines for content and scope.
    – кяαzєя Mod
    Commented Nov 11, 2018 at 0:55
  • No, I mean, are there more than 1 of these too-broad questions? or is this the only one that you found? Commented Nov 11, 2018 at 0:56
  • Let us continue this discussion in chat.
    – кяαzєя Mod
    Commented Nov 11, 2018 at 0:57

2 Answers 2


I honestly don't see the problem with the question. It doesn't ask the nebulously vague question "who is the most-connected seiyuu?", which would indeed be off-topic for being too broad/primarily opinion-based. Rather, it asks "who is the most-connected seiyuu in the 'six-degrees sense'?", which has a clear scope and a clear set of answerable criteria.

I think it's pretty clear from the question text that the asker is referring to the Six Degrees of Kevin Bacon, which has a clear set of criteria for calculating Bacon numbers:

  1. Kevin Bacon himself has a Bacon number of 0.
  2. Those actors who have worked directly with Kevin Bacon have a Bacon number of 1.
  3. If the lowest Bacon number of any actor with whom X has appeared in any movie is N, X's Bacon number is N+1.

This is really just a movie industry-specific version of a collaboration graph. There are lots of similar numbers in other fields that are also well-known in the interwebs:

  • The Erdős number, which describes the collaborative distance between mathematician Paul Erdős and another person, measured by authorship of mathematical papers.
  • The Erdős–Bacon number, which is the sum of someone's Erdős number and Bacon number.
  • The Morphy number, which describes the minimum number of games that link someone to the chess player Paul Morphy.
  • The Shusaku number, which describes the minimum number of games that link someone to the Go player Shusaku.

Sure, the "six-degrees" metric may not necessarily be the most accurate indicator for how "connected" a seiyuu is for some reasonably meaningful meaning of "connected", but it's still a fun piece of trivia that's interesting to know and can be calculated.

In any case, there are lots of algorithms for calculating centrality of a node in a graph, but since Bacon numbers are mentioned, a reasonable metric for centrality (as used by the Oracle of Bacon site) could just be to find the weighted average of the degree of separation of all the people who link to that particular person. And then just find the person with the lowest value. There, a clear and objectively determined answer!

Kevin Bacon himself thinks this game was a joke

I don't think anyone takes Bacon numbers seriously. It's just for fun!

The problem is that we are using a fundamentally speculative assumption to pose another about a similar topic. This will only create more misconception if the basis is not first scrutinized.

So I think you're saying that this may be an XY problem, and you're arguing that finding the best center in a collaboration graph is not a good indicator of how well-connected a seiyuu is. Hmm, maybe. But even if the resulting scores are completely meaningless, it's still trivia that people would be interested in knowing. What's the harm?

However, if we count just people we know in any way (including social media) there can be an argument that many people have less than 6 degree today.

Well yes, but the question's not asking for that. It's just asking for the Bacon-style calculation metric, which only includes professional collaboration.

The criteria is too broad.

I think any reasonable answerer can scope the answer just out of pragmatism, based on which databases are available. I think it'd be rather straightforward to throw together a script to crawl through all of the seiyuu entries in various databases (e.g. IMDB, AniDB, MAL, ANN, etc.), build the collaboration graph, and perform the necessary computations. (This may take a while because you don't want to run afoul of any rate limits.)

I came across the question a while ago and I actually already looked up some of the suggested answers on the Bacon site (which uses data from IMDB); it's just that there's no easy way to query the Bacon site for the actor with the lowest average number, so that's why I haven't posted an answer yet.

Here are some seiyuu that I've looked up so far on the Bacon site:

  • Average Kôichi Yamadera number: 3.624
  • Average Mamoru Miyano number: 3.675
  • Average Norio Wakamoto number: 3.751
  • Average Masako Nozawa number: 3.768
  • Average Hôchû Ôtsuka number: 3.797
  • Average Tomokazu Sugita number: 3.837
  • Average Keiji Fujiwara number: 4.008
  • Average Aoi Yûki number: 4.066
  • Average Nobuyo Ôyama number: 4.216
  • Average Kana Hanazawa number: 4.222

idk, it seems like a fun anime/graph theory crossover question, and I'd like to see it stay open.

  • 1
    Comments are not for extended discussion; this conversation has been moved to chat.
    – Dimitri mx Mod
    Commented Nov 7, 2018 at 10:03

Hmm. I see that I upvoted that question. I admit that and I thus feel some shame.

Anyway...I feel like a question like this isn't infinitely useful knowledge. Sure, it's pop-culture-esque, but hardly useful and hardly anything that merits long interpretation over.

If we're looking for a way to close it, it's objectively too broad; voice actors and actresses can continue to perform roles or work in the industry so there is no definitive way to define this expansion until a seiyuu has retired or passed away, which isn't the point of the pop-culture reference.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .