8L Stats: Average rating by category and page length

[Britva415]

I posted this in the Ratings thread and am reposting it for those people who didn't see it there.

Ratings are strongly influenced by category and page length
View attachment 2288599
The first number is what percentage of all stories published in that category have that many pages. The second number is the average rating for that category and number of pages. This is for stories published from 8/30 to 10/13, 28 days after they were published. So 55% (39 out of 71) of the stories published in Anal from 8/30 to 10/13 were 1 page in length. Those one-page stories had an average rating of 4.28 on 28 days after they were published.

In my simplistic nerdish brain, I think that authors would be grateful for me providing this kind of information. You want to know how you are doing compared to average? Here you go! But when I think more on it, I can see this kind of information pissing off a lot of authors. "I have a 4.67 rating on my three-page Gay Male story, and you're telling me that's below average? Fuck off, asshole!"

Wow, almost 3/4 of Novels and Novellas are three pages or less.

EDIT: I guess probably because of serialized ones.

 

[SimonDoom]

Wow, almost 3/4 of Novels and Novellas are three pages or less.

EDIT: I guess probably because of serialized ones.

Yes, that's probably because they are publishing novels in the form of many short chapters.

It's a terrible tactical error, but it's common.

 

[Constories]

I think the LW stories are low because some readers go there just to vote low and comment on how the wife should have been drawn, quartered, and her head put on a pike in the town square.

I'm not saying that you're completely wrong... but here is some evidence suggesting otherwise. (Plus I just love when people read this story lol)

Just look at the comments. Most are definitely on the side of the wife.

 

[Sinstories]

Data is fun, good stuff.

However, you must be careful with data because it can lie.

I'm not going to dispute the conclusions based on my observations, but, I will provide some other possible explanations of the variance based on what I have seen.

I have posted a few stories over a long period of time (3 years between chapters is a bit excessive, I know ) but I see more of a correlation in scores to quality of writing (my first stuff was bad and the ratings were as well) and to the amount of sexual activity in the story (more sex better scores).

Assuming the quality of my writing and storytelling is consistent (not counting the first couple that were bad), I receive much higher ratings when the stories are less story and more sex. You could argue that different categories lend themselves to more sex, thus driving the variation OR certain subjects may attract more novice writers thus impacting the quality of the writing. It's also possible that my writing of sexual scenarios is better than my writing of situation and dialog, there are a lot of variables involved and the only way to really know is to collect data on the why of ratings.

Another interesting data question would be the number of views by category, though that would definitely be skewed by whether or not a story is part of a series and if so, where in the series it sits. My experience (3 series) is that there is a 50% drop in views from Ch 1 to Ch 2, the next 2 or 3 are relatively stable, then another big drop. But, for the first chapters, I have seen double the views for incest vs erotic couplings (very small sample size obviously).

 

[Jalibar62]

Loving Wives scores need to be normalized to fit the Gaussian provided by the other categories.

Em

I seem to recall someone coming up with a formula to "normalize" LW scores in relation to other categories. Came across this thread while searching for it, actually. (Yes, I'm really late to the party, sorry!) Just looking at 8letter's spreadsheet, LW certainly does appear anomalous.

 

[astuffedshirt_perv]

I posted this in the Ratings thread and am reposting it for those people who didn't see it there.

Ratings are strongly influenced by category and page length

Can you tell if page length preference has changed over time? As I posted in College kids don't read books?, is there a trend to Lit readers preferring shorter pieces?

 

[SimonDoom]

I seem to recall someone coming up with a formula to "normalize" LW scores in relation to other categories. Came across this thread while searching for it, actually. (Yes, I'm really late to the party, sorry!) Just looking at 8letter's spreadsheet, LW certainly does appear anomalous.

I've advocated for this in a few threads. Here's one: https://forum.literotica.com/threads/one-star-trolls.1613997/page-2#post-99381166.

Scrap the current red H system. Convert a score into a percentile score based on category. Let, say, the top 25% of stories within a category get the red H.

The result of this would be jarring to some, because in some categories the top 40+ percent are getting red Hs, which is ridiculous. In Loving Wives, on the other hand, getting a red H means you are in the top 10 percent, or something like that. It's pointless to compare the stories.

It would be jarring to many authors because many authors who do NOT publish Loving Wives stories would lose their red Hs.

My take has always been that ratings systems exist for the benefit of readers, not authors. They are meant, like any ratings system, to convey information that the consumer may find useful before choosing to purchase (or in this case read). But authors have gotten attached to them so there's a lot of intertia against change.

 

[alohadave]

I've advocated for this in a few threads. Here's one: https://forum.literotica.com/threads/one-star-trolls.1613997/page-2#post-99381166.

Scrap the current red H system. Convert a score into a percentile score based on category. Let, say, the top 25% of stories within a category get the red H.

The result of this would be jarring to some, because in some categories the top 40+ percent are getting red Hs, which is ridiculous. In Loving Wives, on the other hand, getting a red H means you are in the top 10 percent, or something like that. It's pointless to compare the stories.

It would be jarring to many authors because many authors who do NOT publish Loving Wives stories would lose their red Hs.

My take has always been that ratings systems exist for the benefit of readers, not authors. They are meant, like any ratings system, to convey information that the consumer may find useful before choosing to purchase (or in this case read). But authors have gotten attached to them so there's a lot of intertia against change.

It'd probably be simpler to just show the stars for each story.

 

[SimonDoom]

It'd probably be simpler to just show the stars for each story.

But as a reader, why would I care about that? I want the simplest thing to look at as possible. A single number is far simpler than looking at a set of numbers and then trying to figure out what their average is. Amazon allows you to see the 1-5 grades by distribution, but the point of that is to enable you to read bad reviews as well as good ones. Scores here aren't necessarily tied to reviews in the same way.

A percentile score by category reduces the evaluation system to a single number that provides some more or less reasonable way to compare stories across all categories. It's simple and it's far more meaningful than the system that exists now.

The other big flaw that the site has not addressed yet, but which I think it has said that it will address eventually, is the way long chaptered-stories fill up the top lists and crowd out other stories. This could be corrected by taking the average score of all the chapters and using that to rank the overall story on the list. This might tweek authors who like seeing their chapters fill up the lists, but, again, the point of the list is to provide information to the readers, not to please the authors.

 

[alohadave]

But as a reader, why would I care about that? I want the simplest thing to look at as possible. A single number is far simpler than looking at a set of numbers and then trying to figure out what their average is. Amazon allows you to see the 1-5 grades by distribution, but the point of that is to enable you to read bad reviews as well as good ones. Scores here aren't necessarily tied to reviews in the same way.

A percentile score by category reduces the evaluation system to a single number that provides some more or less reasonable way to compare stories across all categories. It's simple and it's far more meaningful than the system that exists now.

Simpler isn't always better. If I want to see how others rate a story, a distribution that shows better than a simple average. Has nothing to do with comments or reviews.

 

[SimonDoom]

Simpler isn't always better. If I want to see how others rate a story, a distribution that shows better than a simple average. Has nothing to do with comments or reviews.

Fair enough. I would guess, though, that you're in the minority. I wouldn't care about the distribution, myself. It would say more about the type of story and the type of readers than it would about the quality of the story, IMO, and that doesn't help me identify a story to read.

 

[AwkwardlySet]

Coming back to this thread, it's obvious how beneficial it would be to simply raise the word count threshold considerably. There is so much spam in the 750-3k words area. I've also seen so many chaptered stories with ten or more chapters where each chapter is around 1k words. I suppose some authors think it makes their story file look better if they have ten stories rather than one integral 10k words story.

Also, it has often been speculated here that longer stories score better because readers who find they dislike the story give up before reaching the last page, and thus before they end up rating the story with a low score. But what we see from the data is that there is no linear progression of scores with page length. They keep rising, more or less, until a certain point, but when the number of pages reaches four or five, the function changes.

Hell, I'll be the first one to speculate differently. Maybe these long stories are actually better than those 1-2k words scenes? Outrageous, I know.

 

[SouthernCrossfire]

I don't care to see the distribution of stars but we should be able to see the number of readers who've awarded a rating to older stories without having to go to the top list for the story category and hope that the story is included on it (since that number is shown there). It's great to see a story that's rated, say 4.85, but then realize it's part forty-leven in a series and that a total of 11 readers (who must love the story to be reading Chapter forty-leven in the first place) have rated it. This wouldn't show up in a top list anyway since it doesn't meet the minimum number of votes, but why should we have to search for it beyond the basic story info line?

The other big flaw that the site has not addressed yet, but which I think it has said that it will address eventually, is the way long chaptered-stories fill up the top lists and crowd out other stories. This could be corrected by taking the average score of all the chapters and using that to rank the overall story on the list. This might tweek authors who like seeing their chapters fill up the lists, but, again, the point of the list is to provide information to the readers, not to please the authors.

I second this. Some writers develop large followings this way, but many of the long chapter story listings have relatively few votes in comparison to other stories rated just a bit lower that may have received thousands of votes.

 

[Gamblnluck]

A percentile score by category reduces the evaluation system to a single number that provides some more or less reasonable way to compare stories across all categories. It's simple and it's far more meaningful than the system that exists now.

A percentile will not help a reader select a better story to read from the current list if it is compared to history of the whole category. Or it would be totally skewed if the first ten ratings were a 5 (because of immediate support by followers) it would appear as 100% for a small time then drop later in the day when other readers pan the piece.
I occasionally look at the "Popular' listing and then go to the 7 and 30 day filters for stories I may have missed. I especially do this when I've been away for weeks. I go on cruises and do not have internet. (Lit is banned on those servers anyway)

 

[Gamblnluck]

Simpler isn't always better. If I want to see how others rate a story, a distribution that shows better than a simple average. Has nothing to do with comments or reviews.

I tend to agree that would be most helpful to the author. He'd at least see how the scoring was affected by a couple outlying low scores (One-bombing) or if the overall readership did not care for the piece.

 

[8letters]

Can you tell if page length preference has changed over time? As I posted in College kids don't read books?, is there a trend to Lit readers preferring shorter pieces?

In 2021, I pulled down all the Lesbian Sex stories. Here is how the rating changed over the years by page length:

One and two-page stories have an upward trend in their ratings.

That being said, the biggest thing that jumps out to me is the downward trend in one-page stories. Started at 79% and almost dropped to in half. So LS authors have preferred to write longer stories over time.

 

[SimonDoom]

Coming back to this thread, it's obvious how beneficial it would be to simply raise the word count threshold considerably. There is so much spam in the 750-3k words area. I've also seen so many chaptered stories with ten or more chapters where each chapter is around 1k words. I suppose some authors think it makes their story file look better if they have ten stories rather than one integral 10k words story.

Also, it has often been speculated here that longer stories score better because readers who find they dislike the story give up before reaching the last page, and thus before they end up rating the story with a low score. But what we see from the data is that there is no linear progression of scores with page length. They keep rising, more or less, until a certain point, but when the number of pages reaches four or five, the function changes.

Hell, I'll be the first one to speculate differently. Maybe these long stories are actually better than those 1-2k words scenes? Outrageous, I know.

My tentative theory is not that longer stories are objectively "better" but that in the case of EROTIC stories, there's a minimum sweet spot that delivers a satisfying erotic experience to a plurality of readers, and that minimum sweet spot is over 7000 words.

By the standards of classic, published, well-regarded short stories, Lit stories are rather long. They're longer than the average short story published in The New Yorker or other well known platforms for short stories. The average O Henry story is about 3500 words. That's shorter than one Lit page. So I think it's hard to say longer is better. But there's something going on in terms of reader satisfaction.

 

[8letters]

I seem to recall someone coming up with a formula to "normalize" LW scores in relation to other categories. Came across this thread while searching for it, actually. (Yes, I'm really late to the party, sorry!) Just looking at 8letter's spreadsheet, LW certainly does appear anomalous.

I'm not a LW reader. Occasionally, I check in on the category to try to understand the madness. What I found is that the majority of LW stories follow this plot line:
1. The husband-narrator is in a happy marriage
2. (Optional) Some ominous changes results in the wife pulling away while the husband continues to love her faithfully
3. Wife cheats on husband
4. Husband finds out wife has cheated on him (can happen at the same time as #3)
5. Resolution, which can be of four kinds:
5a. The husband takes the wife back. Restoration At All Cost (RAAC)
5b. The husband refuses to have anything more to do with his wife. He may try to destroy her life and make her miserable for cheating on him. Burn The Bitch (BTB)
5c. The husband accepts that his wife is going to cheat on him. Cuckold
5d. Story ends before the husband decides what to do about his cheating wife

If the story has a RAAC resolution, the BTB crowd will leave angry comments. If the story has a BTB resolution, the RAAC crowd will leave angry comments. If the story has a Cuckold resolution, everyone will leave angry comments. If the story ends before the husband decides what to do, the RAAC crowd and the BTB crowed will argue over what the husband should do.

Stories with such a plot line invariably have a rating around 3 as the people who are pissed off at the ending give the story a 1.

These types of stories are therapy for many men who were cheated on, and Literotica is the only place on the Internet where they can read these stories to their heart's content.

Another common type of LW story is where the wife-narrator cheats on her husband and gets away with it. Those stories typically have a rating of less than 3. LW readers leave lots of nasty comments along the line of "So evil wins, huh?"

Now, there are plenty of stories published in LW that don't follow the above plot lines. From what I've seen, their ratings are in the 4's and are not much different than other categories. But the majority of the LW readership doesn't care about them, and they don't get the extraordinary number of comments that stories that follow the above plot lines get.

So it's not all LW stories get low ratings. Is that there are rating ranges in LW for certain plot lines. "February Sucks" is considered the best of the first type of plot line, and it has a 3.93 rating.

 

[yowser]

Well it's always an endless (and endlessly entertaining) business to speculate/analyse LW reader behaviors, and I mostly agree with the observations above, but like everything in life, the variations and exceptions can prove instructive.

My 8 stories in LW (two series with a LW focus also have chapters with different labels (GS is one) which [surprise] get much higher ratings than the LW chapters, same characters, same story, but you all know how this works) do not fit the outlines above. All but one (a 'boring' no cheating tale with a sub 4 score) are above the 4 threshold, one 'almost' earning a Scarlet H (4.48), so it's not impossible to score relatively well when writing outside the polarised parameters.

My surprise discovery was just how salient the 'cheating' bit is central to LW readership. The closest I got to a 'revenge' tale was told from the wife's perspective, and the revenge was neither dastardly and even had some humor attached to it, but the voting (1807) was an order of magnitude greater than my next most-voted story (almost 800 votes.)

More evidence: in my 'boring monogamous' story (comment: 'A good story and no cheating.') there was some built-in ambiguity about faithfulness, which attracted the 'cheater-antennae' readership like a red flag to a bull:, ie. a comment 'No cheating, huh, you're sure about that ?'

I think to most of the LW readership, a LW tale without cheating is bread without butter.

 

[Plathfan]

Coming back to this thread, it's obvious how beneficial it would be to simply raise the word count threshold considerably. There is so much spam in the 750-3k words area. I've also seen so many chaptered stories with ten or more chapters where each chapter is around 1k words. I suppose some authors think it makes their story file look better if they have ten stories rather than one integral 10k words story.

Also, it has often been speculated here that longer stories score better because readers who find they dislike the story give up before reaching the last page, and thus before they end up rating the story with a low score. But what we see from the data is that there is no linear progression of scores with page length. They keep rising, more or less, until a certain point, but when the number of pages reaches four or five, the function changes.

Hell, I'll be the first one to speculate differently. Maybe these long stories are actually better than those 1-2k words scenes? Outrageous, I know.

There’s a clear correlation between a story’s length and its score, though there shouldn’t be. Quality shouldn’t be measured by word count.

There’s also a link between those who vote generously and the writing style. Character development and slow burn often serve as excuses for endless rambling.

Instead of “He burst into the room, excited,” you’re more likely to get: “He burst into the room, agitated, heart pounding in his ears, sweat beading on his temples, hands clenched, eyes darting around in a frantic blur…” blah, blah.

While some pivotal events deserve detailed description, inflating every sentence makes reading tedious, at least for me.

In first-person, it’s even worse: “I lay on the old, battered purple sofa I bought for my thirtieth birthday at Zagli’s furniture store downtown during the holiday sale. I should’ve gotten rid of it years ago, but it was so comfortable, and I’ve grown used to its ugliness in the living room...” And the rambling, about an effing sofa that has nothing to do with the story, goes on and on.

Voters must enjoy wasting their time, or there’s no other explanation for the inflated scores such stories get.

Stories over 30K words rated 4.85+ are likely well-written, with an engaging plot, a few steamy scenes, and some heartwarming moments. But I’ll probably never know, as there’s no way I’m slogging through 70% fluff and risking brain-damaging irritation.

And to make matters worse, you get awkward repetitions, tedious dialogue that doesn’t move the plot forward, and unnecessary speech tags stuffed with idle descriptions: “she sighed, hand on her hip,” “she gasped, soft disappointment in her eyes,” “she scolded, a stern tone in her voice,” and so on.

I have no problem with long stories that earn their length. But those are exceptionally rare.

BTW, did I mention that voters aren’t exactly the sharpest tools in the shed? Yet some consider pleasing them “success.”

 

[pink_silk_glove]

There’s a clear correlation between a story’s length and its score, though there shouldn’t be. Quality shouldn’t be measured by word count.

As if there's any measurable correlation between score and quality. : P

 

[MillieDynamite]

On this point, @pink_silk_glove and I are in lock-step. Scores are arbitrary, given by people with their own reason well beyond the quality of the text that weighs into the end result.

 

[AwkwardlySet]

BTW, did I mention that voters aren’t exactly the sharpest tools in the shed? Yet some consider pleasing them “success.”

I remember you calling LW readers intelligent. That's quite a change there.

Anyway, I agree that word count shouldn't be the measure of story quality. But it's not so much about word count as it is about who writes all those short stories and who writes longer ones. The impression I got from browsing through the new stories almost every day is that, setting aside for a moment all those silly 750-word challenges that some AH-ers liked to put forward, a big majority of those 1k-word stories were written by new authors, authors who are likely only trying to write for the first time and who are either reluctant or not skilled enough to produce longer, more ambitious stories rather than those single-scene ones.
So by the nature of things, the average skill and experience of those who write 5k-word+ stories for example, and those who write all those numerous 1k-word stories differs by a considerable margin.

 

[ElectricBlue]

My tentative theory is not that longer stories are objectively "better" but that in the case of EROTIC stories, there's a minimum sweet spot that delivers a satisfying erotic experience to a plurality of readers, and that minimum sweet spot is over 7000 words.

This is a point people repeatedly miss - one for the key points of erotica is arousal (I'd have thought), to have a bit of a personal play, a good time, a happy ending.

Judging by the number of short chapter, short stories, there seems to be a whole bunch of people with a very quick release. They're not strokers, they're two strokers, and I thought that was for lawn mowers.

 

[SimonDoom]

This is a point people repeatedly miss - one for the key points of erotica is arousal (I'd have thought), to have a bit of a personal play, a good time, a happy ending.

Judging by the number of short chapter, short stories, there seems to be a whole bunch of people with a very quick release. They're not strokers, they're two strokers, and I thought that was for lawn mowers.

Let's just say that when I was a young man, I might have preferred a shorter erotic story. At this age, I like more buildup, and a longer happy ending.