Analyzing the Top Five Longest Stories on Fimfiction, and Understanding Rate of Decay (RoD)

Let's face it; I have far too much time on my hands. And often, this time is spent rather poorly. So in this case, I had a rather strange idea. Has anyone actually looked at the top longest stories, lately? Has anyone wondered about their statistics? And, most interestingly of all, has anyone wondered how many people actually stick with a story through to the end? Well, even though some people probably didn't (and don't), I ended up wondering this myself. And since I like sharing my stupid ideas with people I barely know, I decided to post my findings here. Along the way, I also came up with a means of measuring the "reader loyalty” on longer, multi-chapter stories, which I will elaborate on in detail further down the post.

So, without further stalling for time, here's the (current) list of the longest stories on Fimfiction, accurate as of two or three days ago. (Granted, this is also with the mature tag on, so I can't link the mature-rated stories. It's not hard to find them anyway, you only have to click the bar up top and search by "longest story"...)

1. Diaries of a Madman, by whatmustido - posted on 19th Aug 2012 - 1,730,229 words total
2. MLP Time Loops, by Saphroneth - posted on 19th Apr 2013 - 1,339,952 words total
3. The Chase, by kudzuhaiku - posted on 25th Apr 2014 - 1,082,847 words total
4. Man of War, by tallestbrony - posted on 7th Mar 2012 - 1,046,509 words total
5. Glory Be, by BlackRoseRaven - posted on 7th Sep 2013 - 1,039,124 words total

Yep, all of those stories are over a million words each. Diaries of a Madman is still the longest story on the site, and interestingly, The Chase has only recently entered the longest story list—which I suppose can be owed to the DAILY updates. (And it's only eight months old at this point!) Good look trying to catch up on them all, guys. But, just how WELL have these stories been received by the public thus far? Well, thanks to a few quick calculations (and dividing), we can figure out the overall vote ratio of these stories.

1. 2900:201 = 14.4:1
2. 902:38 = 23.7:1
3. 1126:182 = 6.2:1
4. 1892:200 = 9.46:1
5. 102:5 = 20.4:1

So, it's quite apparent that the vote ratios are reasonably spread out. MLP Time Loops holds the highest ratio of 23.7 upvotes for every 1 downvote, and The Chase holds the lowest ratio, with a measly 6.2 upvotes for every 1 downvote. Glory Be has the fewest overall votes, but manages to retain a ratio of 20.4 upvotes for every 1 downvote, which is still a good ratio, but still a bit oddly-low, considering the length and age of the story.

Moving right along in the tally, the number of overall comments on the respective stories, along with the total number of chapters:

--Comments--
1. 12,836
2. 3,459
3. 19,690 - Most comments on site(!)
4. 7,392
5. 331

--Chapters--
1. 128
2. 137
3. 428
4. 227
5. 103

Searching the site by "most comments" also confirms that, yes, The Chase now has the most comments out of any story on the site. Such a thing can be owed to the sheer number of chapters, in addition to the dedicated nature of its few fans, who are often repeat commenters on the story. The Chase also holds the highest number of chapters of any story on the site, with over 428 as of this writing and growing daily. (This also makes the story a statistical nightmare, as the numbers change so regularly, information is often out of date within mere days. It's probably already outdated by the time I posted this, even! (Post-posting edit: Yep, it's at 436 now. And a few other values have changed. Not changing them, though; too lazy.)

Now, we move on to the views. The first number is the highest number of views that any single chapter out of the other chapters have recieved, and the numbers in brackets are the total number of views across all chapters.

1. 21,038 (996,175)
2. 17,192 (608,042)
3. 7,826 (661,115)
4. 15,656 (896,875)
5. 817 (39,951)

More often than not (with an odd exception being Glory Be, as the highest viewcount belongs to the epilogue), the first or second chapter of a given story is the one with the highest viewcount. Which only makes sense, of course, given that these chapters will be the first thing people will read before deciding to read on or give it up. I'll also be looking at a formula that determines how "hooked" a reader is on a certain story as it goes on, which is often quite indicative of general interest. But, more on that later. Before then, let's have a look at some chapter-by-chapter averages.

--Avg. comments per chapter--
1. 100
2. 25
3. 46
4. 33
5. 3

--Avg. words per chapter--
1. 13,517
2. 9,781
3. 2,530
4. 4,610
5. 10,089

--Avg. views per chapter--
1. 7,783
2. 4,438
3. 1,545
4. 3,951
5. 388

Going by the averages presented here, Diaries of a Madman has the most comments per individual chapter, where Glory Be has the fewest. Diaries of a Madman also has the most average views per chapter, where again, Glory Be has the fewest. The Chase clearly has the shortest chapters on average, while Diaries beats it out by having the longest. The Chase is also an interesting case, as the total views tally up to over 600k, and the views on the first chapter alone is 7.8k... but then the average views are straight down to 1.5k, which is a massive 81% loss of average views from initial readers. The only story that comes close is Man of War, with an average loss of around 75% of initial readers.

These losses are thanks in part to something I like to call the Rate of Decay, or "RoD" ratio. It's a fairly simple formula that I've tweaked and refined while taking a look at these stories, and the basic form of it is as such:

x = Chapter Views / Any Corresponding Chapter Views x > 1

Pretty simple, right? Measuring the first and last chapters is a "Total RoD" (TRoD), where measuring any two chapters in-between is a "Partial RoD" (PRoD). Any RoD is simply a ratio that expresses, in fraction form, the average trend of reader loyalty; how many people on average start out the story, as opposed to how many finish it. Here's a few examples of the Total RoD from the five longest stories:

1. 11 > 1
2. 8 > 1
3. 12 > 1
4. 53 > 1
5. 1.15 > 1

As can be clearly shown, Man of War currently has the worst Total Rate of Decay, with a pitiful 1/53rd of readers that started the story reading to the end. Conversely, and quite interestingly, Glory Be has the best overall ratio, with nearly all of starting readers finishing the story on average. Diaries of a Madman and The Chase are fairly close in ratio, with The Chase just barely doing less well than Diaries as far as ratio is concerned. But, when the dates and ages of the stories are considered, it seems a little odd that The Chase is already starting to do a little worse in overall ratio in eight months than Diaries of a Madman is doing in over two years. Conversely, from a quick ratio calculation, the first eight months of Diaries of a Madman's history were significantly better in terms of RoD, with story managing to hold an average of 1/3rd of initial readers by chapter eighty.

But, that only begs the question... why would such a thing be happening to a story so soon? It often updates enough, doesn't it? Or... perhaps it updates too often? Hell, The Chase is update at least once if not TWICE on a DAILY basis. Diaries follows a far slower but fairly-steady update schedule, and has remained relatively consistent thus far. Man of War updates about once or twice a month at this point, though it's somewhat unplanned and sporadic. It's possible that if a story updates too soon, most readers will feel overwhelmed and can't catch up with the story after a certain point, and end up dropping it (keep in mind, many of these readers also have plenty of other things to be doing and stories to be reading). If it updates too slowly, people will just forget about it, and not bother keeping up—heck, some people might have even moved on from the site in that time period. There could very well be a "sweet spot" in update frequency that's almost essential for a steady viewership, but there's no obvious signs pointing to an exact value just yet. Chapter length is also an important factor to consider, as not everyone will want to read a series of long 20k chapter as opposed to smaller, 2 or 3k chapters.

Many things seem to factor into a Rate of Decay and/or interfere with results, which can make accurate calculation a tad spotty at best, but such trends seem integral in determining how many readers stick with a story; update frequency, story quality, genre interest/appeal, and plenty of other factors are critical. Finding the trend of how many people are reading the story and actually STILL read it is tedious to determine, as there is no easy way of doing so other than manually checking each individual chapter. It can perhaps be best shown and explained in graph form, as I've already taken the liberty of creating and demonstrating below.

As you can easily see, there's often a “downward trend” of some sort that accompanies Rate of Decay, and how fast the decay occurs is extremely dependent on several factors. In the graphs above, you can easily see the initial read drop-off within the first few chapters of whatever story is analyzed; that's what I refer to as the “sampling slope”, in which curious readers 'try out' the story before committing to it. Obviously, not ALL of readers are going to find the time nor interest to read the whole thing, and not everyone will like it in the place, hence why the sampling slope occurs.

The middle portion of the graph is what I like referring to as the “loyalty marathon”, where a steady readership is achieved, and the graph starts to balance out. After the initial drop-off in readers from the first few chapters, the ones that follow tend to be a bit more interested in reading more of the story. However, as the trend is easily seen, this is not always consistent. Were the graph more detailed in showing the views per-chapter instead of every ten chapters it would be easier to see, but I think it's still obvious that there's some odd downward fluctuation in views as the chapters go on. Not everyone is going to hold on to a story forever, so these losses in views are to be expected. It happens sooner for some stories, as can be seen in the case of The Chase, and sometimes it's not quite as obvious right off the bat. But trust me when I say that, well, it happens on one level or another all the time.

Glory Be is a bit unusual in that the RoD is rather high; there's an obvious dip in viewership around the middle (though it remains quite consistent), but the ending views around the last chapter and the epilogue take a sudden rise back up again. Why is this? Well, from what I can find, the answer is rather simple: This is actually a story that's the latest in a series. The tenth story in said series. Starting to make a little sense, now? People who would have read up to this point are already dedicated, "balanced out" fans, while any extra views are likely from curious spectators. As a consequence, the viewership is very low (only a few hundred), but dedicated. I'm not too sure whether or not this is definitely the last story, but, who knows. I can only determine that the minute fanbase will slowly continue to atrophy as time goes on.

Overall, it's clear to see the states of these stories. The ones that are still ongoing have obviously lost a ton of their initial readers, some faster than others, and for whatever multitude of reasons and variables. Man of War seems to be faring the worst in terms of viewership by ratio, with The Chase coming close behind that. Diaries of a Madman still seems to hold on to a good majority of readers still, both in overall size and in ratio. Glory Be is something of an underdog here, having the lowest overall numbers across the board, but still having good ratios likely because of the lower and more dedicated readership. Fascinating stuff, to be sure.

So, what have I gleaned from all this information? Plenty of things. I know there are a ton of other things I could check and cross-reference and whatnot, but I've already spent too much time on this already, so I think I'll call that a day. Hopefully I'm not the only person that finds this sort of thing mildly interesting, though; hell, you can probably even check your own stories, too. As a matter of fact, I think I oughta check mine out, as well...

...Huh, not too bad, actually. A roughly 3 > 1 ratio is pretty decent, all things considered. XD

Anyway, that's enough of my rambling for tonight. Hope you all have a pleasant New Year's, guys! And I happen to have a few new stories coming pretty shortly, along with a badly-needed update for a prior story. That'll be something to look forward to, right? But until then, good night, and have an awesome New Year's Eve.

From the desk of die Kanzlerin,
~Maud

Comments

[Csquared08]

I'm pretty sure that BlackRoseRaven has written more words than anyone else on the site.

[Chained Marionette]

Slow day, huh?

[Glimglam]

2691640
Ja...

[StardustSurvivor]

You know you're a boring person when you take the time to read something like this and somewhat enjoy it. I am a very boring person.

[Chained Marionette]

2691647

Aww... I could always try entertaining you sometime.

[Roarin Thunder]

2691647 I guess the Chase is really the worst long fic on site.

[Cerulean Voice]

This is actually quite fascinating. I wondered about the rate of decay on my own stories, and found it to be fairly low. I did something a little different some weeks ago, where I compared the total views of a chaptered fic vs the amount of chapters, to determine the average amount of story a user read.
I also made a blog about the top five most followed users once, back in December 2013. It involved their total upvotes/downvotes, their total views, and their total upvote:downvote ratio vs total views in comparison to my own. I actually found that in regards to u:d ratio vs total story views, I was beating them all.
Go figure.
Of course, that was over a year ago. I wonder how times have changed...

[FOXTrot2]

I love this chart.