Damn lies?

[Green_Knight]

No, this isn't a thread about Donald Trump. I pulled it out of the Who are the important writers on Lit?, in the full realisation that it will upset those who regard statistics as the work of the devil or insist that the delicate art of writing should never be sullied by such mercenary things. That's a view, of course. I take the view that stats can help my writing – or, at the very least, getting it published and read. Of course, if you write just for the fun of it, that doesn't apply but, if you want people to read and like your work, you need to understand why they do and, more importantly, why they don't.

The New list is certainly important, but I think you're underrating the back catalogue.

When I post a story, I get a lot of views in the first week (i.e. lifetime of the New list). After that it slows down considerably, but it doesn't stop. For instance, my most recent story picked up 2666 views in its first week, and another 2982 (i.e. more than doubling) in the next 11 months. I'm not aware of anything that would be driving readers to that story so I presume they're finding it by exploring the back catalogue. I do have another story on a toplist; it's possible they're finding that one first and then checking out what else I've written.

Some of the others are a bit slower, and it might take a couple of years to pick up as many views as they did in the first week. But they keep going, slow and steady. I still get the occasional comment or e-book sale for stuff I posted 4-5 years ago. I expect authors who post more frequently than I do would see a bit more action in their old stories, since that makes it easier for people to find them.

This chart shows the early progress of a four-chapter recent work of mine. It shows very clearly the 'New list' effect when each chapter is first published. It also shows, in the line for Chapter 1, the secondary boost that the publication of chapters 3 and 4 gave to its views. This work was a little unusual in that the chapters were published in quick succession so that one or more was always in the New list for some three weeks. The views during those three weeks for the four chapters combined amounted to 83 per cent of their current total, getting on for three months later.

In the world of real publishing, the flattening-off of the curve (aka keeping going, slow and steady) would be the signal for a book to disappear from the popular bookshop and supermarket shelves and maybe even for it to be remaindered, though the flattening-off probably wouldn't occur as quickly as in the electronic world of Lit. On Lit, of course, you can't remainder something that was free in the first place and the stuff remains forever, cluttering up the digital shelves. Perhaps we need a second-hand bookshop – an Abebooks for old Lit stories? Or a 'Last chance to read' list on Lit for stories about to be consigned to the ether?

Incidentally, if you're wondering why chapter 2 had so many views compared to the others, it's probably down to the sub-title, which included the words 'daughter' and 'virginity'.

 

[oggbashan]

On Lit, of course, you can't remainder something that was free in the first place and the stuff remains forever, cluttering up the digital shelves. Perhaps we need a second-hand bookshop – an Abebooks for old Lit stories? Or a 'Last chance to read' list on Lit for stories about to be consigned to the ether?

...

Whenever I post a new contest story I see some activity on my older stories too. I even get comments on stories a decade old.

I would be very unhappy if my older stories were to be deleted from Literotica. With the steadily reducing cost of digital storage, 'cluttering up the digital shelves' is not a problem. ALL my stories, posted or incomplete, plus all the earlier versions of them for nearly 20 years fit on one CD as Word 97 and .txt files with space to spare.

Even going back to earlier technology I could fit all my posted stories as .txt files on one 360k 5.25 floppy.

One of the attractions of Literotica is that there are some stories for almost every taste except underage sex or bestiality. Using Google to search Literotica reveals some very weird stories (some of which are mine ).

So far no one else has written about sex inspired by Chairman Mao's Little Red Book.

 

[lovecraft68]

Older stories will always be read providing the author is still active and producing newer works as each new story will get readers to fav you and check out your library of older stories.

Any story that manages a top spot in a category hub will continue to get some reads because some readers use those lists at some point to check out the alleged best in their fav category.

If an author is on the top 250 list, especially in the top 25 or so older stories will keep getting play because people click on the author and see all their stories.

The only way a story here truly vanishes is if the author stops writing and never made any lists. There are stories on here that are close to 20 years old and may have not had a read in forever.

So I feel that after the new lists most of the read stories here come from favoriting authors and top lists.

 

[lovecraft68]

So far no one else has written about sex inspired by Chairman Mao's Little Red Book.

Keep saying that and someone is going to take that dare

 

[Green_Knight]

...I would be very unhappy if my older stories were to be deleted from Literotica. With the steadily reducing cost of digital storage, 'cluttering up the digital shelves' is not a problem. ALL my stories, posted or incomplete, plus all the earlier versions of them for nearly 20 years fit on one CD as Word 97 and .txt files with space to spare.

Even going back to earlier technology I could fit all my posted stories as .txt files on one 360k 5.25 floppy...

As a one-time information scientist, my concern is less about the cost of digital storage, more about the difficulties of accessing a specific piece of information in a universe of vast quantities and of assuring the quality of information that you are able to access. For example, used in Lit, the search term "bit his cock" – hopefully not a routine part of most people's sexual repertoire – produces 179 Lit stories, with no indication apart from the red 'H's as to which are worth reading – though I have to point out that 'bit his cock' does seem to produce a lot of red H results! "Camping trip" produces 1423 stories, and so on. At an average of 2 minutes to check out a story, that would take 47 hours – more than a working week – to go through them.

Information in volume becomes unmanageable unless there are extremely good, usable tools to handle it. Such tools don't exist and, in any case, most people don't even know how to use the poor tools that do exist.

 

[oggbashan]

...

Information in volume becomes unmanageable unless there are extremely good, usable tools to handle it. Such tools don't exist and, in any case, most people don't even know how to use the poor tools that do exist.

That's why many use Google tools to search.

http://forum.literotica.com/showpost.php?p=18985333&postcount=1

 

[lovecraft68]

As a one-time information scientist, my concern is less about the cost of digital storage, more about the difficulties of accessing a specific piece of information in a universe of vast quantities and of assuring the quality of information that you are able to access. For example, used in Lit, the search term "bit his cock" – hopefully not a routine part of most people's sexual repertoire – produces 179 Lit stories, with no indication apart from the red 'H's as to which are worth reading – though I have to point out that 'bit his cock' does seem to produce a lot of red H results! "Camping trip" produces 1423 stories, and so on. At an average of 2 minutes to check out a story, that would take 47 hours – more than a working week – to go through them.

Information in volume becomes unmanageable unless there are extremely good, usable tools to handle it. Such tools don't exist and, in any case, most people don't even know how to use the poor tools that do exist.

Hence why the e-book market is increasingly tough because they are on the 'bookshelf' forever and the market now has at this point hundreds of thousands of e-books to choose from.

So same strategy is key...only way to keep selling is keep having something new so people see the old.

 

[sr71plt]

I don't pay a whole lot of attention to statistics here. When I opened up this morning, there were nine "story favorite" notifications--all for old stories, the earliest from 2008. I've had three new entries posted in the last week. I'll leave it to the statisticians to try to make sense out of that. Publishing to Internet Web sites isn't the same as putting print books on the shelves of brick and mortar bookstores.

 

[Green_Knight]

...there were nine "story favorite" notifications--all for old stories, the earliest from 2008. I've had three new entries posted in the last week. I'll leave it to the statisticians to try to make sense out of that.

A statistician friend of mine suggests there are a whole lot of masochists out there reading your back catalogue.

 

[Zeb_Carter]

And I pay so little attention that on most of mine I have voting and comments turned off.

 

[TxRad]

Statistics on Lit are worthless. They are squired nine ways to Sunday by more things than can be counted. There are so many games being played with all the numbers they are meaningless. Even the favorites list is being gamed and has been for the last four years.

 

[HandsInTheDark]

Statistics on Lit are worthless. They are squired nine ways to Sunday by more things than can be counted. There are so many games being played with all the numbers they are meaningless. Even the favorites list is being gamed and has been for the last four years.

This. Even ignoring the gaming, which I see no way to quantify, statistics are as good as they data they come from - and there's no way to scrub the data here. The changes Lit would have to make to fix that would likely put it out of business. I've stopped looking at scores; only emails matter now. And they don't matter as much as they used to.

 

[GoldenCojones]

This. Even ignoring the gaming, which I see no way to quantify, statistics are as good as they data they come from - and there's no way to scrub the data here. The changes Lit would have to make to fix that would likely put it out of business. I've stopped looking at scores; only emails matter now. And they don't matter as much as they used to.

Unfortunately, statistics are only as good as the person manipulating them. In one of my advanced statistics classes (years ago) the class split into two groups (6 in each group) and given the same data. We were then instructed to prove two opposite conclusions using the data given. Both groups were successful. The entire exercise was a lesson in not trusting statistics.

That is not to say that statistics are not useful. They are. But, when used to draw a conclusion, they are not to be trusted.

 

[sr71plt]

A statistician friend of mine suggests there are a whole lot of masochists out there reading your back catalogue.

I like to think it's folks who aren't spinning their wheels trying to make numbers mean something. I never think about stats when I write and submit a story. Stats aren't going to make me change what I'm doing. And it saves me time and effort I can put to writing another story. I think all of the stats business is just an aspect of procrastination to avoid actually writing.

 

[lovecraft68]

I think statistics are more accurate for the majority of authors than the ones here. Want to get trolled, gamed, have the authors around you gamed to pass you? Come to the boards and post something unpopular and your now the target of all manner of BS.

For those who simply write here and avoid the boards, the contests especially and aren't creeping up on the hall of fame list(another sire way to get trolled and sniped) I think the numbers are far less skewed.

I agree with pilot in the sense of you can't let stats effect what you write, that is if you strive to be your own writer. The stat junkies here are obviously writing based more on what readers want rather than any actual muse.

 

[MindsMirror]

We're stat junkies, but only stumbled upon the boards when researching the plagiarism of one of our stories. As we've posted here, we have seen the odd corresponding down votes, but don't have proof positive that there is a linkage. Hopefully Lit will come to their senses and revamp the way stories and authors get exposure. While they are at it, they should take some input from the writers that provide for their income (via traffic to the site). There were rumors of better trending of author statistics, but they haven't appeared as yet.

-MM

 

[Bramblethorn]

This chart shows the early progress of a four-chapter recent work of mine. It shows very clearly the 'New list' effect when each chapter is first published. It also shows, in the line for Chapter 1, the secondary boost that the publication of chapters 3 and 4 gave to its views. This work was a little unusual in that the chapters were published in quick succession so that one or more was always in the New list for some three weeks. The views during those three weeks for the four chapters combined amounted to 83 per cent of their current total, getting on for three months later.

For a larger/longer dataset, I've attached a couple of mine. First shows views against date; second shows views against time since posting. (Unfortunately I didn't think to start keeping stats until several months after I started posting, so it doesn't show the very earliest stuff).

The first 14 "SI" stories are a single series in Lesbian Sex; "Counting to Eleven" is also in LS. MI and WSJ are Horror, Red Callum & Sweet Cate is in Text With Audio, and Riddle is F/SF.

As you noted, every successive chapter in a series gives a bump to the earlier chapters - following the "SI 1" line you can clearly see that happen. There doesn't seem to be much a bump from stories in a different category, e.g. RC/SC didn't have any visible effect on SI stats. From that I'm assuming that

Most stories show a "hockey stick" shape: they get a burst of views in the first week, then slow down sharply around the time they'd fall off the New Stories list, but continue to accumulate views at a more or less linear rate. Chapter 1 of SI gets more views than the others, presumably from people who check out the first chapter and decide it's not their thing.

Some of the chapters of SI have been in and out of the relevant toplist. You can see this by sudden changes of slope: e.g. chapter 13 made it onto the toplist around day 280, with flow-on benefits for chapter 1, before falling off again around day 500.

"Counting to Eleven" does very well on views despite having the lowest ratings of any of my stories. I think this is because the blurb ended up more stroke-y than the story, so it may be drawing in a lot of readers who are hoping for something that it doesn't give them.

In the world of real publishing, the flattening-off of the curve (aka keeping going, slow and steady) would be the signal for a book to disappear from the popular bookshop and supermarket shelves and maybe even for it to be remaindered, though the flattening-off probably wouldn't occur as quickly as in the electronic world of Lit.

Yep, because keeping physical stock costs money. But that's a drawback of physical media, not a "feature" to be emulated in digital publishing.

On Lit, of course, you can't remainder something that was free in the first place and the stuff remains forever, cluttering up the digital shelves. Perhaps we need a second-hand bookshop – an Abebooks for old Lit stories? Or a 'Last chance to read' list on Lit for stories about to be consigned to the ether?

I think you're trying to solve a problem that doesn't exist. The search function already has an option to restrict searches to the last 1/3/6/12 months, and stories are displayed with their posting date. Anybody who doesn't care to read older stories can already achieve that pretty easily without having to enforce that restriction on other readers.

Incidentally, if you're wondering why chapter 2 had so many views compared to the others, it's probably down to the sub-title, which included the words 'daughter' and 'virginity'.

Could be; I suspect the blurb does have a big effect on views. But did you happen to notice where it appeared on the New Stories view?

One chapter of my SI series was lucky enough to appear right at the top of page 1 of the New Stories view, and it sat there for most of a day. Another one came in at the bottom of a big batch of updates, so it was shunted to page 2 as soon as it appeared. That made a BIG difference to the view numbers.

 

[Green_Knight]

I think statistics are more accurate for the majority of authors than the ones here. Want to get trolled, gamed, have the authors around you gamed to pass you? Come to the boards and post something unpopular and your now the target of all manner of BS.

For those who simply write here and avoid the boards, the contests especially and aren't creeping up on the hall of fame list(another sire way to get trolled and sniped) I think the numbers are far less skewed.

Oh, yes, I've seen plenty of evidence of that.

I agree with pilot in the sense of you can't let stats effect what you write, that is if you strive to be your own writer. The stat junkies here are obviously writing based more on what readers want rather than any actual muse.

I take your point and that's fine if you're writing for fun but, if you're writing for profit or for a publisher, what the market/publisher wants – which is a mixture of economics and what the reader demands – is paramount. It's no good your muse telling you to write a 500-page modern day romance when your publisher is expecting a 400-page historical drama. Of course, you have every right to write a work that no-one wants to read but, secretly, at least, most authors write what they think readers want to read, even if they deny it. Otherwise, you might as well write the story for your own satisfaction and then press 'delete' instead of the upload button.

 

[TxRad]

I take your point and that's fine if you're writing for fun but, if you're writing for profit or for a publisher, what the market/publisher wants – which is a mixture of economics and what the reader demands – is paramount. It's no good your muse telling you to write a 500-page modern day romance when your publisher is expecting a 400-page historical drama. Of course, you have every right to write a work that no-one wants to read but, secretly, at least, most authors write what they think readers want to read, even if they deny it. Otherwise, you might as well write the story for your own satisfaction and then press 'delete' instead of the upload button.

I don't know who your publisher is but making demands beyond what their writers normally write is way too far out there. They may make suggestions on what they think they are going to want in the future but that changes all the time. Most publishing houses are broke up into section, each with their own style and subjects.

New writer getting their foot in the door submit what they have unless someone within the publishing house sees something that they are looking for. That was what happened in my case. They needed some stories on a certain subject, read a non-erotic story i had posted here. They wanted the story in a hurry. Three months later, I delivered the first of three novels.

Since then, I have written stories in two other genre but the stories were of my own choosing up to a point. Most difference were hashed out between me and my editor. She also has a list of what is wanted or needed. I look through it and see what I think i can do well.

If your publisher makes such demands then you need a new publisher among other things.

 

[sr71plt]

There isn't one market in e-book erotica publishing (or Lit. publishing, for that matter). It's an expansive group of niches, each with its own target audience. I have no problem linking with a publisher for a niche something I've written because it was something I wanted to write without any convoluted statistical study followed by hammering a square peg in a round hole. Just sounds to me that someone loves to play with (mostly meaningless) numbers, which is OK, but . . .

It also sounds a bit like someone constructing rules and theories for something they don't actually do.

 

[lovecraft68]

Oh, yes, I've seen plenty of evidence of that.



I take your point and that's fine if you're writing for fun but, if you're writing for profit or for a publisher, what the market/publisher wants – which is a mixture of economics and what the reader demands – is paramount. It's no good your muse telling you to write a 500-page modern day romance when your publisher is expecting a 400-page historical drama. Of course, you have every right to write a work that no-one wants to read but, secretly, at least, most authors write what they think readers want to read, even if they deny it. Otherwise, you might as well write the story for your own satisfaction and then press 'delete' instead of the upload button.

All very true, and so you know, I am speaking solely about lit. Not the outside market.

They do have a some thing in common though. Biggest of which that I have noticed in erotica is, like here, you can write the same trope fifty times and they will keep buying it just as they keep reading it here. Original still is odd man out for erotica free and paid.

 

[HandsInTheDark]

Unfortunately, statistics are only as good as the person manipulating them. In one of my advanced statistics classes (years ago) the class split into two groups (6 in each group) and given the same data. We were then instructed to prove two opposite conclusions using the data given. Both groups were successful. The entire exercise was a lesson in not trusting statistics.

That is not to say that statistics are not useful. They are. But, when used to draw a conclusion, they are not to be trusted.

Depends on the discipline. In a hard engineering realm, where events are measurable and truly quantifiable, statistics works fine. It's possible to cherry-pick data of course, but that's detectable. In other realms, events are not measurable in a meaningful way, the data is inherently fuzzy, and p-values start to wander. Any attempt to clean such data becomes cherry-picking, and you lead the numbers rather than the numbers leading you. It can even be unconscious. Your class proved it.

That's why stats at Lit are pointless. The event is a person clicking 1 to 5 stars, and there's no measurable meaning to what 5 means or what 1 means. Around here 1 often means "your story rocked and it started to rise into the toplist and we can't have that now can we." Or "you mentioned red stockings. Only nude stockings are hot." Or "I just got dumped by my boyfriend and your character is just like him, you bastard!"

No context, no meaning.

 

[Bramblethorn]

Unfortunately, statistics are only as good as the person manipulating them. In one of my advanced statistics classes (years ago) the class split into two groups (6 in each group) and given the same data. We were then instructed to prove two opposite conclusions using the data given. Both groups were successful. The entire exercise was a lesson in not trusting statistics.

That is not to say that statistics are not useful. They are. But, when used to draw a conclusion, they are not to be trusted.

Yeah, this is why best practice is to decide how you're going to analyse the data BEFORE seeing the data, or to split it into 'training' and 'validation' sets. Unfortunately an awful lot of professional researchers seem to have missed that memo.

It's no good your muse telling you to write a 500-page modern day romance when your publisher is expecting a 400-page historical drama.

If the only thing your publisher wants is historical drama and you don't feel like writing historical drama... you're probably better off finding a different publisher, or finding another job entirely.

By and large, writing pays very badly. A few superstars manage to get rich off it, but most professional writers are barely scraping by unless they have some other means of support. So why bother writing something you don't enjoy writing? Might as well get a job stacking shelves somewhere.

 

[Bramblethorn]

Depends on the discipline. In a hard engineering realm, where events are measurable and truly quantifiable, statistics works fine. It's possible to cherry-pick data of course, but that's detectable.

I've seen plenty of stats abuse even in hard engineering. Occasionally cherry-picking of data, but more often cherry-picking of methods. If you get to play with the data before deciding how to analyse it, there's plenty of room to bias the findings towards the results you want to see. cf. Simpson's paradox: simply by choosing what variables to include in your analysis, you can reverse the direction of the findings.

 

[ChloeTzang]

So far no one else has written about sex inspired by Chairman Mao's Little Red Book.

Oh no. Don't say things like thst. I have enough ideas now.
."Power comes from the barrel of a gun," Ming said.
"Target practice time," I whispered, jerking his grey Chairman Mao pants down. He certainly seemed close to shooting as I reached for him.
Etc.