Mathematicians mapped “Game of Thrones” relationship to find the main character | Hacker News Reader

Mathematicians mapped “Game of Thrones” relationship to find the main character

qz.com

94 pointsCarolineW61 comments

This is only a popularity contest among the known characters and their known interactions. For instance, Littlefinger started out seemingly as a small player but then we learn later he was actually a rather significant player in almost all aspects of what has been happening so far. Granted his obvious interactions are with rather important people but by appearances the number of interactions seem low while actually being quite large.

As for Tyrion; he has a high position in life, hangs out with a large number of similar characters, and even crosses an ocean to meet up with more of that group that the others have not met yet. Regardless, I would say he's not so significant to what's going on simply because most of the others don't listen to a thing he says, even though he's right most of the time, so he doesn't have as much impact on events as he should. Just because he's present at significant events doesn't make him significant. Although, I admit those reasons are why I find him one of the more interesting characters.

Seems fun but I don't see how it can answer the question without it just being about today's favorite. It cannot be fully answered until the end.

> [Tyrion is] not so significant

He was essentially responsible for the successful defense of King's Landing at the Battle of the Blackwater. If he hadn't been there, Stannis would be on the Iron Throne -- I'd call that high-impact.

I agree that Littlefinger's impact is under-represented by this measure, though. And obviously Arya has had no impact on anything important at all ti date...

Not to mention his kidnapping was responsible for the war that broke out in the first book and set the stage for every major event in the series. He also put a cross bow bolt through the gut of the guy who had been running the 7 kingdoms.

The kidnapping was not the reason, it was the excuse.

He was also accused of killing the King, which is more important than killing his father in the scheme of things. But notice he didn't do it and not many characters care anymore.

Tywin wasn't running the 7 kingdoms and never did. Once he came into that level of power there was no unity in the kingdoms, half of them or so in open revolt. He controlled King's Landing and related lands through his grandchild. His family was broke and the only thing they had left was family on the throne (which is not a rightful claim), which is quickly dwindling away and under direct threat from multiple fronts. Notice that once he was dead, it was swept away to something not to worry over just like much of anything involving Tyrion.

Tyrion has been present at many significant events but in the end has little impact on their outcomes. I suspect he'll continue this trend and then one day people may actually start listening to him, which will make things really interesting.

This is one of the things that make this story interesting; that anybody can be made insignificant regardless of their position and power. If there is to be a main character named, I would say it's the "game" because all the characters are just pieces on the board.

Defense of King's Landing? You mean the event that everyone immediately dismissed and took credit away from him? How many characters even recall it that way? It was significant at that moment, but dwindled away to next to nothing almost immediately. Other than that one thing, Tyrion has mostly gone along for the ride, which fits his character.

As for Stannis being on the Iron Throne, it would makes some things different but it would not make a huge difference in the overall storyline so far as to what's coming. We wouldn't have gotten all the fun stuff with Tyrion's family true, but most of everything else likely would have continued on the same. Just because Stannis would have taken the throne wouldn't mean everything else would suddenly stop or change.

But remember what they are saying they are trying to find here is the main character, not the main character in the universe. You actually argue against your own point. You say this is a popularity contest, and then say no one in universe would remember it that way. But if his impact, and hence relationship to the actual story, is what this measures.

But I am talking about the story, not the universe. There is no main character to a universe.

But I'm not understanding how I'm arguing against my own point, could you elaborate?

I'm saying that the chart is an indication of a popularity contest as defined by the rules of the chart. But the rules of the chart does not necessarily follow the rules defined by the story.

As for the events that I'm saying the characters don't recall the same as it happened, that's my point that the character did not have that much impact to the overall story because their involvement is forgotten or has been deemed unimportant by the other characters. Therefore, Tyrion's impact has been negligible to the story overall. For the most part, almost all the characters have had very little impact to the story overall; which I find really interesting. They have significant moments that's a huge impact to the characters individually, but may not have much significance to the story as a whole.

So far I would say that Littlefinger is the closest to a main character we have with this type of thing in mind. Simply because he has influenced almost everything that has happened so far in some way.

Almost all the deaths in the story have had an effect on readers simply because almost all the characters have the same weight (or whatever term) as any other despite their position in the perceived power structure. The story taught us very early that no one is safe, therefore there is no easy choice to be made as to who the hero or villain of the story is. Because so far neither really exists. This is one of the many reasons I find this story so interesting, it doesn't read as your typical hero-vs-villain fantasy story where you can guess the outcome before the end of the second act.

> would not make a huge difference in the overall storyline so far as to what's coming

If you mean Daenerys I mostly agree. Stannis would not recognise her claim, and would fight her invading forces like whoever's left will do. On the other hand, he'd likely have rather more resources to do it with -- an earlier end to the fighting, possibly ending the war with Robb etc etc.

If you mean the Walkers, though, I rather disagree. Stannis is about duty, and he knows that defending the wall is a big and important job.

Essentially I think the point of most of the action in the books is to leave Westeros as unprepared as possible for the invasions of Dani and the Walkers, and Stannis quickly on the throne means less fighting and dying for everyone.

One of my favorite, subtle parts of the series is that the prisoner escape Arya engineered for Robett Glover and his men, was actually already planned. Glover had struck a deal with Vargo Hoat to release them once they were in prison to take over Harrenhal. Which means, once again Arya wasted her wish.

Further analysis here: https://racefortheironthrone.wordpress.com/2015/10/14/chapte...

Yes, I absolutely love Arya's character because of things like that. She is a major factor of teaching the reader how the world works because we learn through her realizations of her failures and disappointments just as she does.

> For instance, Littlefinger started out seemingly as a small player but then we learn later he was actually a rather significant player in almost all aspects of what has been happening so far. Granted his obvious interactions are with rather important people but by appearances the number of interactions seem low while actually being quite large.

i would argue that that makes him a significant plot device, rather than a significant character. tyrion is a major character not because he influences events the most, but because a lot of the book focuses on how events relate to him and vice versa.

But I see that as a difference of a character being significant to the reader versus significant to the story.

The graph is interesting. But maybe Game of Thrones doesn't have a "main" character. One of the things fans love about the books is that they subvert genre expectations at every turn. Given how things are going, in fact, I won't be surprised if the next literary convention GRRM eschews is "conclusion".

Exactly. Instead of a lot of plotting and planning, people just kill and die. It's very, very refreshing.

But there's lots of plotting and planning for people to just kill and die as well.

I've only read three fantasy series besides ASOIAF but all of the have similar character development.

I think the difference in ASOIAF is that there isn't a clear set of protagonists and antagonists. Or maybe the story so far only talked about the protagonists and the antagonists are yet to be revealed.

I think this is one way to look at it, at least for the books. For the TV show, over all the seasons it seems like three main characters have emerged (based on longevity, screentime, and plot importance): Daenerys, Tyrion, and Jon Snow. You could perhaps argue Cersei as well - iirc she has more POV chapters and page-time in the books than any other character.

An alternative view is that the main character(s) change from book to book and season to season. Ned was probably the main character in the first book & season one, Robb replaced him in book/season two, with similar evolutions for other "temporary-mains." Though really the first explanation seems better: a "main character" just doesn't fit the structure of ASOIAF very well.

Robb was never a POV character

Didn't actually say that he was, though it may be the TV show has changed my memory of his place in the second and third books. After a few years it gets hard to keep track of exactly what happened in a ~5000 page series.

Ah interesting. Personally I don't think so. He is not a postmodern author in that sense. Subversion of tropes in the fantasy genre is one thing, getting rid of literary/narrative conventions is quite another leap.

spoiler alert I wouldn't say this necessarily points to the "main" character, as its just one measure of a character's significance. Littlefinger (Petyr on this diagram) is rather small and on the edge - but he's ultimately responsible for a huge amount of what goes on.

you are right but the aim is to find who the writers consider the "main" character and not who had the most impact on the GoT universe. If we take it in that sense the white walkers are probably the most significant.

I would say there is no single main character to the story.

that's also the most likely reason the story (book or TV) is so popular. the fact that there is no distinct good or evil, no main character, no main villain etc. everyone is grey, everyone can die etc

There are obviously evil characters, obviously good characters, and ambiguous characters. Mostly ambiguous.

This is an analysis of the books, who have just one author.

Some of the characters who have had the most effect on the events as they are taking place in the books are dead before they start and get very little page time.

> Whenever two characters appeared within 15 words of one another, a link (or “edge”) was added between them. The links are weighted based on how often the two characters appeared in close proximity.

The interesting part here is that the algorithm needs no upfront understanding about the relationships nor about the type of interactions between characters. You can construct the graph from just looking at the locations of the words associated with each character.

Can they map postmodern literature to find the plot?

http://www.critical-theory.com/this-is-what-happens-when-you...

This would be interesting to run on the HN community, or any popular forum.

Reminds me of Six Degrees of Harry Lewis[0], where Mark Zuckerberg set up a graph to find out the most popular person in Harvard, via cross-referencing articles in the crimson. That project may or may not have been the origin of Facebook.

0 - http://bostinno.streetwise.co/all-series/the-harvard-profess...

This graph only covers the third book, A Storm of Swords.

I'm surprised that they haven't gone back and generated a graph for all of the released books yet.

GRRM said Tyrion is his favorite character so not that surprising.

Is it so hard to imagine that a popular series could represent a non-traditional narrative structure in which there is no, one, main character?

"In spite of his small size, Tyrion is mathematically the most important character in Game of Thrones."

... followed by Jon Snow.

I want to see this animated.

With the 3D clockwork graphics and the musical theme score!

The dragon has three heads, so we're told. Tyrion, Jon, Daenerys?

This is a popular fan theory (Jon's parentage is very ambiguous, and it can be argued that Tyrion's parentage is not completely clear, leading to speculation that they both have Targaryen blood).

Yep. That's the general theory so popular and makes enough sense that in a way, I'll be both pleased (for picking up on it) and disappointed (for lack of surprise) if it really turns out that way.

At the same time, it would be sort of lame if GRRM decided "well, everyone figured out R+L=J and the three headed dragon and all so I guess I better change some things around".

Then I just get a little bit bummed out (and guilty for feeling entitled) that the next book probably won't be out for a while still.

tl;dr for those with short attention spans like me: Tyrion, Jon, and Sansa are identified as the central characters.

Mathematicians take the fun out of everything and turn it into a graph

you need a mathematician to come up with that diagram?

Certainly not to record the relations or visualize them, but social network analysis certainly falls into the realm of mathematics.

In this case the mathematician decided on his own to make the diagram, it isn't an answer to some external need.

Graph centrality measures are a pretty interesting and ongoing topic of research in mathematics, yes. I'm familiar with methods based on eigenspectra of the matrix exponential (and some relevance to Krylov subspace methods), but that's really it.

But I guess there will be some arbitrariness involved anyway, even when mathematicians look at the problem. E.g. why would one measure be better than another?

Shocker: POV characters are probably main characters.

http://awoiaf.westeros.org/index.php/POV_character says there are 31 POV characters, many of whom aren't main characters.

Readers of the books know that is definitely not the case.

For instance Areo Hotah is a repeated point of view character who is merely there to provide prospective on the court of Dorne as a silent body guard. Fans of the series call him "the camera who rides."

On the other hand GRRM has said characters like Littlefinger and Varys will never have chapters because their plans are too revealing of the end of the book. They are too central to be PoV characters.

His chapters is called "The captain of guards" and not "Areo" as is the way chapters of the main POV characters is named.

Numerous characters both major and minor get chapter names like that. Asha gets a chapter where she's "the sacrifice", Victarion gets a chapter where he's "the iron suitor" and chapters that just go by their name as well. Arya has been "the ugly little girl" in the chapter title. Theon has been "the ghost of winterfel".

Regardless, he's still a point of view character that isn't central to the plot.

Or you know... you could count the number of chapters that every character has and come up with the one that has the highest number of chapters as the main character?

Which, if your RTFA, they have done and noted. They show that this is not quite as simple as it seems since, for example, Arya has more POV chapters but Sansa is more important to the story.

Is she though? What measure are you using to say that Sansa is more important than Arya?

Or you could just count the number of times their names are mentioned... Except neither approach really represents character importance accurately. Why just go the naive route when you can extract more information from your data by using a more complex approach?

The article actually counters your point exactly:

"For example, Arya Stark, Sansa’s younger sister, has the third most chapters with 34, but ranks behind Sansa in terms of network importance."

I don't think the naive approach is too far off in this case. Arya's importance is reflected in the sheer amount of exposition devoted to her. She's pretty clearly being set up to make a major return at some point, just as it is with Daenerys. The characters with a lot of network connections, on the other hand, are easy to kill off because they're ultimately redundant.

> Or you could just count the number of times their names are mentioned

"Hodor"

Reply on news.ycombinator.com