Friendliest Category

[NotWise]

I wondered this morning which of the categories gave the highest ratings. I think this question has been asked and answered before, but probably with different methods, and possibly with different results. I devised a comparison then had to go off to work. The results follow. Discussion later.

Rank Category Rating
1 NonHuman 4.93
2 Lesbian Sex 4.92
3 SciFi/Fansasy 4.92
4 Romance 4.91
5 GM 4.9
6 Novels and Novellas 4.9
7 First Time 4.89
8 Trans 4.89
9 Fetish 4.88
10 Mind Control 4.88
11 Non-Erotic 4.88
12 BDSM 4.87
13 Erotic Coupling 4.87
14 Mature 4.87
15 Group Sex 4.85
16 Incest/Taboo 4.85
17 Interracial Love 4.85
18 NonCon 4.84
19 Celebrity 4.82
20 E & V 4.81
21 Loving Wives 4.81
22 Anal 4.8


Some of the results surprised me.

The ranking above is based on the rating of the story ranked at 0.1% of the number of stories in each category on the category's all-time top list. For instance, there are 15.5K stories in Interracial Love, and the score on the right side of the table is the score for the 15th entry on the all-time top list for Interracial love.

This gives a basis for comparing most of the categories and avoids the large spread in ratings at the top of many categories. Some categories are missing: Audio has no top list; several categories have less than 1K stories posted and I ignored them; categories with less than 5K stories (Erotic Horror, for instance) gave undependable results.

Of course, since this is based on the all-time ranking it doesn't represent any recent trends. There probably aren't enough stories on most of the 30-day lists to give meaningful results.

The biggest surprise for me was finding Anal and E&V at the bottom of the list. Comments on AH lead me to believe that both categories were very author-friendly. Non-Human at the top of the list also surprised me. Tentacle sex, anyone? I was surprised to find GM at #5 (I thought it would be lower) and I/T at #16 (I thought it would be higher).

PS: Edited for table format.

 

[alohadave]

I think using the list is a mistake. They represent the very tippy top of the pile, and will always skew much higher than the average story in any category.

I think you need to do some more calculating to find a meaningful number rather than the 15th most popular story.

 

[joy_of_cooking]

(In case any other newbies were confused, GM = Gay Male and E&V = Exhibitionist & Voyeur)

The 99.9th percentile is an odd metric. Why not some of the more accessible quantiles, like 25th, 50th, 75th, 90th, 95th, ...? Or multiple of them? Although of course that means more work.

Does this site have an API?

 

[NotWise]

(In case any other newbies were confused, GM = Gay Male and E&V = Exhibitionist & Voyeur)

The 99.9th percentile is an odd metric. Why not some of the more accessible quantiles, like 25th, 50th, 75th, 90th, 95th, ...? Or multiple of them? Although of course that means more work.

Does this site have an API?

I'm aware of no data available on higher quartiles. There is no "middle list," and frankly, I don't care about medians or 25th quartile, etc. I was expressly interested in how highly readers rated stories, so the top list was good, and it gave me a means of comparing across different categories of widely different sizes.

For what it's worth, the actual rating on the story was just given for documentation. The ranking is the primary product.

 

[EmilyMiller]

Friendliest Category​

Would it make more sense to average all of the 99th centile, rather than take the boundary value?

Also, I think some categories fall off quicker in terms of ratings than others. You really need the entire dataset to draw any meaningful conclusions.

TL-DNR all 99th centiles are equal, but some are more equal than others.

Em

 

[SimonDoom]

I would be cautious about using the term "author-friendly" about categories based on these numbers. E&V may rank near the bottom of the list, but from my experience it's one of the more open-minded and pleasant categories in which to publish. The numbers are just relative and subjective. They have no absolute value.

 

[RejectReality]

You're not going to glean much from the toplists so long as they still include individual chapters. The phenomenon of score rising while all other stats fall as the story goes on is well documented, so you're pulling numbers from a pool that's weighted toward diehard fans. The more chaptered stories a category has in the toplist, the more it's skewed.

 

[joy_of_cooking]

Sorry to pile on @NotWise . I guess I wasn't the only one who fancies himself a numbers kind of person.

Does anyone know how the site's owners feel about people hitting their APIs or scraping their site? It shouldn't be that hard to get ratings for every story. Even an amateur like me could hack something together for that.

 

[NotWise]

Sorry to pile on @NotWise . I guess I wasn't the only one who fancies himself a numbers kind of person.

Does anyone know how the site's owners feel about people hitting their APIs or scraping their site? It shouldn't be that hard to get ratings for every story. Even an amateur like me could hack something together for that.

8letters scrapes things regularly, and I haven't heard him talk about kickback from the site.

 

[ronde]

This isn't kickback. It's just an observation by one who has done statistical analysis of manufacturing processes for the last thirty years. What you've attempted to do is a cause and effect analysis but your hypothesis is that the category with the highest rating for the story ranked at 0.1% of the total number of stories in that category is representative of overall reader satisfaction for all the stories on the site. I believe you are also assuming the distribution of the various ranking is normal, and I doubt it is. The reason is that most people who do not like a story simply do not vote, so the distribution of rankings is skewed toward the high side of the distribution.

That said, I don't think it's possible to do a meaningful analysis of the relative reader satisfaction of any category given the fact that many people read or at least click on a story but in my experience anyway, only about 2 "readers" per 100 actually vote. Those readers who vote either really like the story or hate it. Thankfully, the haters are a small number or else the ratings you state would be much lower.

 

[ElectricBlue]

8letters scrapes things regularly, and I haven't heard him talk about kickback from the site.

The metadata's in the public view, so providing a scrape is passive, read only, then it shouldn't be too much of a problem. The polite thing to do would be to ask. And to definitely share any analysis with the site, before publishing results.

Remember years ago, someone studying anthropology did an interesting analysis - not on the OP's topic as such, but on the Literotica readership as it was back whenever. @Bramblethorn knew where it was.

 

[JGittes]

Last year (2022), I gathered detailed, daily data on every story in the Valentine's Day and Nude Day contests. I'll have to dig it up and post the results here. While not every category had submissions, there were enough stories in the major categories to draw some conclusions.

I ranked categories with the highest scores, and NW's list is close to what I remember. The clear winners in higher scores were Romance, Novels & Novellas, and Lesbian Sex. My interpretation was not that voters in those categories had lower standards, but rather they didn't give deliberately low-ball scores.

 

[HectorBidon]

I wondered this morning which of the categories gave the highest ratings.
. . .
This gives a basis for comparing most of the categories and avoids the large spread in ratings at the top of many categories.

You're right that different categories have different voting curves. However, I have to agree with alohadave that a single number, especially way up at the tip-top percentile, probably doesn't tell the whole story.

You can also go at it the other way around and figure out where a particular story ranks in its particular category. For example, I see that you have a story in Erotic Horror with a rating of 4.69. A little work on the Search Stories page reveals that this story is ranked 627 out of 4842 stories in the category, i.e., in the 13th percentile. You have another story in Erotic Couplings with the same rating of 4.69. This one is ranked 6300 out of 71932, or in the 9th percentile.

So even though the stories have the same numeric rating, the one in EC has a better percentile ranking. This presumably means that scores in EH are a little inflated in comparison with scores in EC.

There was a thread about this a while back (unfortunately the tables no longer display correctly). The way to determine a story's ranking is to go to the Search Stories page, enter a single period (".") in the "Search For" box, click the gear icon to select advanced search, select the category (Erotic Horror, say), select "Sort results by vote score in descending order," and click "Search." In this case the page finds 4842 stories. Then you have to look through them to find the one you're looking for. In this case it turns out to be the 27th story on the 13th page of the results, which works out to be the 627th story overall.

This procedure is pretty inefficient because you have to page through the results looking for the story of interest. So you wouldn't want to do it all the time. But once in a while it's interesting to see how your stories stack up in comparison with the other stories in the same category.

 

[Bramblethorn]

I'm aware of no data available on higher quartiles. There is no "middle list," and frankly, I don't care about medians or 25th quartile, etc. I was expressly interested in how highly readers rated stories, so the top list was good, and it gave me a means of comparing across different categories of widely different sizes.

For those who do want quantiles, it's not too hard to get them; it's basically what Hector posted, but without the trial-and-error step.

Say I want to find the 90th percentile for stories in Lesbian Sex:

• Go to "Advanced Search" page
• Enter "the OR a" as search term (basically a dummy, because this field can't be blank; almost all stories will contain one of those words)
• Select category
• Select "sort results by Vote Score", in descending order.
• Click the "Search" button.

https://search.literotica.com/?query=the OR a&categories=11&sort=vote&page=1

The search results tell me that there are 21960 stories fitting these criteria. (The main stories page says 21.9k stories in this category, so that looks right.) Looking for the 90th percentile, i.e. 10th from the top, means I'm looking for story # (21960 * (100-90)/100) = story # 2196.

There are 440 pages of search results, i.e. fifty stories per page. Dividing 2196 by 50 gives me 43, remainder 46, which tells me that the 90th percentile will be story #46 on page 44 of the search results. There's a text box that lets me jump straight to page 44:



And the 46th story on that page has a rating of 4.75. In fact, every story on that page does, so I don't need to bother counting to find it.

Stories without a rating are sorted as if they had score zero, so if you wanted to be absolutely precise you'd need to figure out how many of those there are and exclude them from the story count. For LS that number is very small, only about 135 out of all 22k stories in the category, so I'd actually be looking for story #(21825 * (100-90)/100) = 2182.5, i.e. midway between 2182 and 2183. That's still on page 44 of the results, still 4.75.

For the very low percentiles, and for categories where people are more prone to switch off voting, it might be more of an issue. But even in LW, where I might expect more people to disable voting, it looks like only about 1.4% of stories have no score.

Odd thing, if I sort in ascending order and look at the very lowest scores just after the last scoreless story, the first few results are not displaying in score order. The exceptions are all new stories. I suspect what's happening here is that these stories all had a 1.0 on their first and then-only vote, so they got sorted accordingly, and then by the time I actually got to this page in the search results their scores had updated with new votes, but the positions hadn't. So might want to be careful of new stories when doing this.



For the very high/low quantiles, I'd probably want to add an "older than 1 month" restriction to the query, otherwise the results are going to be dominated by new stories which have only received a single vote.

 

[lovecraft68]

Part of why the scores in some of the cats is so high is not so much friendly, but they're flooded with endless chapter series and those feature a lot of 4.95 scores on 30 votes because its chapter 55. I'd dub them "The Milked to death, never see the words The End" category more than friendly.

Off the top of my head, going by comments and general attitude of readership, I'd say group seems to be pretty laid back(minus the LW shit stains the run around yelling its cheating) and TBH I/T is friendly, they just want their fix and unless you mix in Non con or don't have "real" incest they're generally not ones to troll.

 

[Bramblethorn]

@NotWise You have "BDSM" in the table twice.

You're not going to glean much from the toplists so long as they still include individual chapters. The phenomenon of score rising while all other stats fall as the story goes on is well documented, so you're pulling numbers from a pool that's weighted toward diehard fans. The more chaptered stories a category has in the toplist, the more it's skewed.

I didn't expect this to work, but it does: if you set your search terms to something like '-"Ch." -"Pt." -"Vol." -"Ep." -"Bk."', applied to titles only, you can get all stories which don't have those terms in the title. This doesn't filter out all the chapter stories, but it gets a lot of them. Adding '-"0" -"1"' cuts it further; it will also exclude some non-chapter stories, but I wouldn't expect that to affect the quantiles too much.

Without that filtering, SF/F has about 27744 stories. 99th percentile is 4.91, 90th is 4.83, median is 4.6. But if I filter on those terms I get 8395 hits, and the 99/90/50 percentiles are 4.89/4.74/4.38.

Same filtering for a few other categories:

NH: 4.85/4.69/4.35
EC: 4.79/4.61/4.28
Anal: 4.75/4.58/4.22
LW: 4.67/4.4/4.01

So, yes the chaptered stories do affect the scoring, but even with most of them removed from the picture it still looks like SF/F and NH are much more generous at the top end than EC or Anal, and with those filtered LW is now tougher than Anal. I'm not going to do all the other categories, but if anybody's curious, feel free.

 

[AwkwardlySet]

Part of why the scores in some of the cats is so high is not so much friendly, but they're flooded with endless chapter series and those feature a lot of 4.95 scores on 30 votes because its chapter 55. I'd dub them "The Milked to death, never see the words The End" category more than friendly.

He is taking the stats from all-time top list where you need 100 votes minimum for your story to get listed, so series that are too niche aren't included. Arguably, the threshold should be more than 100 and probably adjusted for each category due to huge differences in popularity.

 

[NotWise]

@NotWise You have "BDSM" in the table twice.

Thanks, Bramblethorn. The second coming of BDSM should have been Mature, and I've edited the OP accordingly.

 

[NotWise]

I'm starting to see why 8letters doesn't reply to comments on his threads. Some of ya'll are like the guys who comment on your stories and think you should have written your story to fit their kink.

Probably, if I were to change anything (aside from the problem that Bramblethorn pointed out) I'd change the title of the thread to "Which categories give the highest ratings?" That was actually the question I was asking.

I'm contemplating a ghost story, possibly for an upcoming contest. I've written two ghost stories, placing one in Erotic Horror and one in Novels and Novellas. The story that went to EH was a very nice little story, but I think it's under-appreciated. Among the comments I've received on and off line was the comment that it would have done better in NonHuman. I never considered putting it in NonHuman.

This certainly improves the odds that, if I write the story, it'll go to NonHuman.

 

[alohadave]

Probably, if I were to change anything (aside from the problem that Bramblethorn pointed out) I'd change the title of the thread to "Which categories give the highest ratings?" That was actually the question I was asking.

You'd still want to not use the top list because by definition they are as highly rated as you are likely to get.

Just for curiosity, I looked at the scores you posted. The median is 4.87, and the std dev is like .03. All the scores are within 2 std dev. They are throughly average to each other.

I'm not a statistician, so maybe it's off base, but using the top lists isn't useful if you want to find the highest ratings of the category.

 

[MillieDynamite]

Just my two cents worth, whose actual value might be closer to 1/2 cent, I think each writer is going to find a category that is friendlier to them than others. I haven't really taken the time to see what that category is for me. I like to spread it around, and my stories are what they are, which is determined by what comes into my rather chaotic mind at the time I'm writing. This also accounts for all the stories I have that are in various stages of production. My mind moves off what I'm working on, no amount of effort can get me back on track, so I must come back when the spirit moves me.

Just glancing at a few of my stories, I see non-human, erotic horror, transgender & cross-dressing have been kind to me, my lesbian stories to a lesser degree. But again, that's my experience, not everyone's. If you're looking for the most accepting readers, I'm not sure one category is better than another, other than avoiding LW.

Your mileage in any one category may vary from mine.

 

[Pureotica]

I wonder if the lower scoring categories might also be the heavier trafficked ones. But even that may be skewed, because the most published stories doesn't necessarily mean the most reads, votes, etc. I wonder if a new writer in incest, for example, gets more reads or gets lost in the crowd, where a Fantasy/SF writer may get more reads because its fans have fewer options. But I write what I write, and try to find the most appropriate category. I do my best, but don't fret the scores because I can't control them anyway. I won't put something in LW if I can avoid it. A lot of my current projects are cheating stories, but the cheating serves a character development purpose. Thankfully I have other places to put them. (If I stop procrastinating in the AH forum and actually finish writing them).

My glacial writing speed makes it tough to gauge my writing's reception because it only gets a few votes, and a single vote has a bigger impact when you only get a few to begin with. But some people I respect give me useful feedback, so that actually means more than the arbitrary score anyway.

 

[SimonDoom]

I'm starting to see why 8letters doesn't reply to comments on his threads. Some of ya'll are like the guys who comment on your stories and think you should have written your story to fit their kink.

Anytime you start a thread like this, where you are offering an interpretation of site data, you're going to get this kind of response. It's not a bad thing. You started an interesting conversation. The fact that it goes in unexpected directions isn't bad at all; it makes it more interesting. Don't take it personally if people challenge your interpretations. Nobody's opinion is privileged over anyone else's opinion.

 

[MillieDynamite]

But Simon, I'm always right cause I'm special and privileged.

Anytime you start a thread like this, where you are offering an interpretation of site data, you're going to get this kind of response. It's not a bad thing. You started an interesting conversation. The fact that it goes in unexpected directions isn't bad at all; it makes it more interesting. Don't take it personally if people challenge your interpretations. Nobody's opinion is privileged over anyone else's opinion.

 

[SimonDoom]

But Simon, I'm always right cause I'm special and privileged.

Millie, it's your keen self-awareness that makes you so special.