Statistical Analysis of LitE Stories

lovecraft68 said:

Its been awhile since I thought on what might max views etc, but wondering if you'd get better results numbers wise if you spread the chapters a week apart rather than put the entire thing out in a week? I'd think you'd get more bang out of each one with some space between.

Click to expand...

I typically don't read chapter stories. I don't want to start a series and then have it never completed, or have to wait forever for its completion. I'm more like to read a chapter story if I see a chapter a day as I have more faith that the series will be completed.

Nice work, 8letters. I compared your data with the snapshot I'd taken of activity in October 2015. My dataset didn't include all the variables you collected, but it's possible to compare the number of stories in each category, the percentage of stand-alone stories, the average ratings, and the percentage of stories that received red Hs. Both datasets include a month's worth of stories (2163 stories from 2018, 2278 stories from 2015), although the ratings were tabulated slightly differently. (In 2018, the rating was taken when a story had been up exactly one week. In 2015, the ratings were taken at the end of the month, so that different stories had been up different lengths of time.) The results for the most popular categories are shown here, with the 2018 data on the left in each column and the 2015 data on the right.

[tr][td]Category___[/td] [td][/td] [td]ries_[/td] [td][/td] [td]Alone[/td] [td][/td] [td]Rating_[/td] [td][/td] [td]Red H[/td][/tr] [tr][td]Incest/Taboo[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]E Couplings[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Sci-Fi Fantasy[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]BDSM[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Loving Wives[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]NC/Reluct[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Gay Male[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Celebrities[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Fetish[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Group Sex[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Mind Control[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Trans & Cross[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Exhib & Voy[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Romance[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Lesbian Sex[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Mature[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]Interracial[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr][tr][td]Novellas____[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td]correlation[/td] [td][/td][td].87[/td] [td][/td][td].88[/td] [td][/td][td].86[/td] [td][/td][td].81[/td][/tr]

The correlations between the two sets of data are quite high (> 0.8) suggesting a certain stability in the site's operation. Authors are writing stories in the different categories at about the same rates now as then, and with about the same percentages of stand-alone vs chapter stories. Readers are rating the different categories at about the same levels now as then. They tend to vote up in Sci-Fi/Fantasy, Romance, and Novella and vote down in Loving Wives. As you point out, the fact that the SF&F and N&N categories having lots of multi-chapter stories with dedicated readerships undoubtedly contributes to this.

So, if nothing else, analyses like these confirm and clarify our hunches about the way things work.

8letters said:

I typically don't read chapter stories. I don't want to start a series and then have it never completed, or have to wait forever for its completion. I'm more like to read a chapter story if I see a chapter a day as I have more faith that the series will be completed.

Click to expand...

You could put something in the disclaimer that it is finished and you'll be posting one a week.

HectorBidon said:

Nice work, 8letters. I compared your data with the snapshot I'd taken of activity in October 2015. My dataset didn't include all the variables you collected, but it's possible to compare the number of stories in each category, the percentage of stand-alone stories, the average ratings, and the percentage of stories that received red Hs. Both datasets include a month's worth of stories (2163 stories from 2018, 2278 stories from 2015), although the ratings were tabulated slightly differently. (In 2018, the rating was taken when a story had been up exactly one week. In 2015, the ratings were taken at the end of the month, so that different stories had been up different lengths of time.) The results for the most popular categories are shown here, with the 2018 data on the left in each column and the 2015 data on the right.

[tr][td]Category___[/td] [td]

_Sto​

[/td] [td]ries_[/td] [td]

%Stnd​

[/td] [td]Alone[/td] [td]

__Avg​

[/td] [td]Rating_[/td] [td]

__%​

[/td] [td]Red H[/td][/tr] [tr][td]Incest/Taboo[/td] [td]

264​

[/td] [td]

249​

[/td] [td]

35​

[/td] [td]

53​

[/td] [td]

4.34​

[/td] [td]

4.25​

[/td] [td]

38​

[/td] [td]

27​

[/td][/tr] [tr][td]E Couplings[/td] [td]

200​

[/td] [td]

246​

[/td] [td]

59​

[/td] [td]

67​

[/td] [td]

4.19​

[/td] [td]

4.20​

[/td] [td]

29​

[/td] [td]

26​

[/td][/tr] [tr][td]Sci-Fi Fantasy[/td] [td]

166​

[/td] [td]

122​

[/td] [td]

16​

[/td] [td]

26​

[/td] [td]

4.56​

[/td] [td]

4.45​

[/td] [td]

64​

[/td] [td]

66​

[/td][/tr] [tr][td]BDSM[/td] [td]

138​

[/td] [td]

144​

[/td] [td]

32​

[/td] [td]

53​

[/td] [td]

4.19​

[/td] [td]

4.09​

[/td] [td]

30​

[/td] [td]

21​

[/td][/tr] [tr][td]Loving Wives[/td] [td]

131​

[/td] [td]

150​

[/td] [td]

68​

[/td] [td]

63​

[/td] [td]

3.44​

[/td] [td]

3.64​

[/td] [td]

4​

[/td] [td]

1​

[/td][/tr] [tr][td]NC/Reluct[/td] [td]

126​

[/td] [td]

122​

[/td] [td]

35​

[/td] [td]

50​

[/td] [td]

4.18​

[/td] [td]

4.16​

[/td] [td]

15​

[/td] [td]

20​

[/td][/tr] [tr][td]Gay Male[/td] [td]

108​

[/td] [td]

123​

[/td] [td]

42​

[/td] [td]

48​

[/td] [td]

4.32​

[/td] [td]

4.32​

[/td] [td]

35​

[/td] [td]

37​

[/td][/tr] [tr][td]Celebrities[/td] [td]

103​

[/td] [td]

55​

[/td] [td]

38​

[/td] [td]

53​

[/td] [td]

3.87​

[/td] [td]

4.32​

[/td] [td]

14​

[/td] [td]

44​

[/td][/tr] [tr][td]Fetish[/td] [td]

99​

[/td] [td]

97​

[/td] [td]

43​

[/td] [td]

55​

[/td] [td]

4.09​

[/td] [td]

4.12​

[/td] [td]

23​

[/td] [td]

23​

[/td][/tr] [tr][td]Group Sex[/td] [td]

96​

[/td] [td]

95​

[/td] [td]

46​

[/td] [td]

55​

[/td] [td]

4.23​

[/td] [td]

4.21​

[/td] [td]

39​

[/td] [td]

24​

[/td][/tr] [tr][td]Mind Control[/td] [td]

96​

[/td] [td]

42​

[/td] [td]

30​

[/td] [td]

38​

[/td] [td]

4.31​

[/td] [td]

4.30​

[/td] [td]

40​

[/td] [td]

31​

[/td][/tr] [tr][td]Trans & Cross[/td] [td]

93​

[/td] [td]

58​

[/td] [td]

40​

[/td] [td]

40​

[/td] [td]

4.41​

[/td] [td]

4.29​

[/td] [td]

51​

[/td] [td]

40​

[/td][/tr] [tr][td]Exhib & Voy[/td] [td]

92​

[/td] [td]

83​

[/td] [td]

43​

[/td] [td]

45​

[/td] [td]

4.35​

[/td] [td]

4.28​

[/td] [td]

40​

[/td] [td]

35​

[/td][/tr] [tr][td]Romance[/td] [td]

69​

[/td] [td]

120​

[/td] [td]

42​

[/td] [td]

43​

[/td] [td]

4.46​

[/td] [td]

4.39​

[/td] [td]

57​

[/td] [td]

53​

[/td][/tr] [tr][td]Lesbian Sex[/td] [td]

65​

[/td] [td]

74​

[/td] [td]

49​

[/td] [td]

64​

[/td] [td]

4.39​

[/td] [td]

4.15​

[/td] [td]

44​

[/td] [td]

31​

[/td][/tr] [tr][td]Mature[/td] [td]

54​

[/td] [td]

73​

[/td] [td]

61​

[/td] [td]

62​

[/td] [td]

4.29​

[/td] [td]

4.28​

[/td] [td]

26​

[/td] [td]

30​

[/td][/tr] [tr][td]Interracial[/td] [td]

54​

[/td] [td]

74​

[/td] [td]

57​

[/td] [td]

59​

[/td] [td]

3.84​

[/td] [td]

3.97​

[/td] [td]

16​

[/td] [td]

16​

[/td][/tr][tr][td]Novellas____[/td] [td]

_____53​

[/td] [td]

__56​

[/td] [td]

______6​

[/td] [td]

__23​

[/td] [td]

_____4.53​

[/td] [td]

__4.40​

[/td] [td]

_____74​

[/td] [td]

__55​

[/td][/tr] [tr][td]correlation[/td] [td]

0​

[/td][td].87[/td] [td]

0​

[/td][td].88[/td] [td]

0​

[/td][td].86[/td] [td]

0​

[/td][td].81[/td][/tr]

The correlations between the two sets of data are quite high (> 0.8) suggesting a certain stability in the site's operation. Authors are writing stories in the different categories at about the same rates now as then, and with about the same percentages of stand-alone vs chapter stories. Readers are rating the different categories at about the same levels now as then. They tend to vote up in Sci-Fi/Fantasy, Romance, and Novella and vote down in Loving Wives. As you point out, the fact that the SF&F and N&N categories having lots of multi-chapter stories with dedicated readerships undoubtedly contributes to this.

So, if nothing else, analyses like these confirm and clarify our hunches about the way things work.

Click to expand...

Those averages are down from a couple years ago, maybe three at this point. So are votes/views/comments...in general as in some authors will always get the numbers, I'm talking your everyday lit story.

People will charge in and say traffic is as good or better than ever. That could be true, but traffic can be the boards, chat, role plays, the pic threads etc...traffic does not mean reading and reading does not equate voting.

The worst part is that with the average scores down it means the lower vote totals indicate it's the better readers no longer here or voting, and the trolls and bombers are as strong as ever.

I know people will refute any claim that makes it seem like Lit could be lagging, and I'm sure as hell not spending days on end coming up with graphs and stats. I know what I saw before and what I see now.

Where I really see it is on the new story lists and the average scores and votes the stories get compared to awhile back.

Why is it? That I don't know and wouldn't try to guess.

lovecraft68 said:

You could put something in the disclaimer that it is finished and you'll be posting one a week.

Click to expand...

That's what I do with my series. Up front I say it's finished, I say how many chapters there will be, and I give an estimation of when it should all have gotten posted.

electricblue66 said:

You can also submit all chapters at the same time and ask Laurel to schedule the release. I did that for my latest, and each chapter went out every 24 hours - she did let two chapters out at the same time, but the rest went live with a midnight server refresh, someplace US time.

Click to expand...

I know you can do this, but I don't push this "keeping track of it" burden off on Laurel. I can space them myself and when she's giving that extra service to someone she isn't available to be clearing in new submissions.

lovecraft68 said:

Its been awhile since I thought on what might max views etc, but wondering if you'd get better results numbers wise if you spread the chapters a week apart rather than put the entire thing out in a week? I'd think you'd get more bang out of each one with some space between.

Click to expand...

I thought about that, and made the call to "see what happens." My gut feel is the overall take-up would have been less, if it was a weekly release over three months. As it was, several chapters were on the category front page (Sci-Fi and Fantasy) for about three and a half weeks before the last chapter rolled off the front page.

Overall, the chapter three retention rate is around 25% the Chap.1 views, and I can see folk are still reading through to the end at their own pace. I can also see which chapters are being read twice.

There's no way of knowing whether a slower release would gather more reads in the short term, and it's too early to say what the long term numbers will be. However, it does have twice the number of votes/view, compared to everything else of mine, so that's telling me something.

lovecraft68 said:

Where I really see it is on the new story lists and the average scores and votes the stories get compared to awhile back.

Why is it? That I don't know and wouldn't try to guess.

Click to expand...

I don't disagree with you. It seems to me like things were different when I first started posting stories in 2015.

The "random" stories list in each hub gives a way to sample past stories. I've used it before to see if the percentage of stories over 4.5 has changed. I was looking mostly at I/T (where I suspect readers have become more critical) and at a couple hubs where the percentage of stories over 4.5 is very high.

It might be interesting to take a body of random samples, sort them be date and see if there have been changes. You can get every statistic there that you can get off the current hub.

I had a high retention rate with "Lowborn" in Sci-Fi & Fantasy twice a week, which is about what it would be when you submit each chapter as soon as one posts.

As others have said, in the opening note of Ch. 01, say that the story is complete, and will post ( weekly, twice weekly, etc. ) the queue willing and the crick don't rise.

I don't remember exactly what the performance was, but the drop in voters chapter to chapter for the first half of the ten chapters was remarkably low. I'm talking about single digits. It dropped a little more after the first half, but still had high retention throughout, over more than a month to post it all.

Now that it's been out for a while, it's skewed more like most chapter stories, with views and votes weighing down the beginning and the final chapter.

8letters said:

I typically don't read chapter stories. I don't want to start a series and then have it never completed, or have to wait forever for its completion. I'm more like to read a chapter story if I see a chapter a day as I have more faith that the series will be completed.

Click to expand...

HectorBidon said:

Nice work, 8letters. I compared your data with the snapshot I'd taken of activity in October 2015. My dataset didn't include all the variables you collected, but it's possible to compare the number of stories in each category, the percentage of stand-alone stories, the average ratings, and the percentage of stories that received red Hs. Both datasets include a month's worth of stories (2163 stories from 2018, 2278 stories from 2015), although the ratings were tabulated slightly differently. (In 2018, the rating was taken when a story had been up exactly one week. In 2015, the ratings were taken at the end of the month, so that different stories had been up different lengths of time.) The results for the most popular categories are shown here, with the 2018 data on the left in each column and the 2015 data on the right.

[tr][td]Category___[/td] [td]

_Sto​

[/td] [td]ries_[/td] [td]

%Stnd​

[/td] [td]Alone[/td] [td]

__Avg​

[/td] [td]Rating_[/td] [td]

__%​

[/td] [td]Red H[/td][/tr] [tr][td]Incest/Taboo[/td] [td]

264​

[/td] [td]

249​

[/td] [td]

35​

[/td] [td]

53​

[/td] [td]

4.34​

[/td] [td]

4.25​

[/td] [td]

38​

[/td] [td]

27​

[/td][/tr] [tr][td]E Couplings[/td] [td]

200​

[/td] [td]

246​

[/td] [td]

59​

[/td] [td]

67​

[/td] [td]

4.19​

[/td] [td]

4.20​

[/td] [td]

29​

[/td] [td]

26​

[/td][/tr] [tr][td]Sci-Fi Fantasy[/td] [td]

166​

[/td] [td]

122​

[/td] [td]

16​

[/td] [td]

26​

[/td] [td]

4.56​

[/td] [td]

4.45​

[/td] [td]

64​

[/td] [td]

66​

[/td][/tr] [tr][td]BDSM[/td] [td]

138​

[/td] [td]

144​

[/td] [td]

32​

[/td] [td]

53​

[/td] [td]

4.19​

[/td] [td]

4.09​

[/td] [td]

30​

[/td] [td]

21​

[/td][/tr] [tr][td]Loving Wives[/td] [td]

131​

[/td] [td]

150​

[/td] [td]

68​

[/td] [td]

63​

[/td] [td]

3.44​

[/td] [td]

3.64​

[/td] [td]

4​

[/td] [td]

1​

[/td][/tr] [tr][td]NC/Reluct[/td] [td]

126​

[/td] [td]

122​

[/td] [td]

35​

[/td] [td]

50​

[/td] [td]

4.18​

[/td] [td]

4.16​

[/td] [td]

15​

[/td] [td]

20​

[/td][/tr] [tr][td]Gay Male[/td] [td]

108​

[/td] [td]

123​

[/td] [td]

42​

[/td] [td]

48​

[/td] [td]

4.32​

[/td] [td]

4.32​

[/td] [td]

35​

[/td] [td]

37​

[/td][/tr] [tr][td]Celebrities[/td] [td]

103​

[/td] [td]

55​

[/td] [td]

38​

[/td] [td]

53​

[/td] [td]

3.87​

[/td] [td]

4.32​

[/td] [td]

14​

[/td] [td]

44​

[/td][/tr] [tr][td]Fetish[/td] [td]

99​

[/td] [td]

97​

[/td] [td]

43​

[/td] [td]

55​

[/td] [td]

4.09​

[/td] [td]

4.12​

[/td] [td]

23​

[/td] [td]

23​

[/td][/tr] [tr][td]Group Sex[/td] [td]

96​

[/td] [td]

95​

[/td] [td]

46​

[/td] [td]

55​

[/td] [td]

4.23​

[/td] [td]

4.21​

[/td] [td]

39​

[/td] [td]

24​

[/td][/tr] [tr][td]Mind Control[/td] [td]

96​

[/td] [td]

42​

[/td] [td]

30​

[/td] [td]

38​

[/td] [td]

4.31​

[/td] [td]

4.30​

[/td] [td]

40​

[/td] [td]

31​

[/td][/tr] [tr][td]Trans & Cross[/td] [td]

93​

[/td] [td]

58​

[/td] [td]

40​

[/td] [td]

40​

[/td] [td]

4.41​

[/td] [td]

4.29​

[/td] [td]

51​

[/td] [td]

40​

[/td][/tr] [tr][td]Exhib & Voy[/td] [td]

92​

[/td] [td]

83​

[/td] [td]

43​

[/td] [td]

45​

[/td] [td]

4.35​

[/td] [td]

4.28​

[/td] [td]

40​

[/td] [td]

35​

[/td][/tr] [tr][td]Romance[/td] [td]

69​

[/td] [td]

120​

[/td] [td]

42​

[/td] [td]

43​

[/td] [td]

4.46​

[/td] [td]

4.39​

[/td] [td]

57​

[/td] [td]

53​

[/td][/tr] [tr][td]Lesbian Sex[/td] [td]

65​

[/td] [td]

74​

[/td] [td]

49​

[/td] [td]

64​

[/td] [td]

4.39​

[/td] [td]

4.15​

[/td] [td]

44​

[/td] [td]

31​

[/td][/tr] [tr][td]Mature[/td] [td]

54​

[/td] [td]

73​

[/td] [td]

61​

[/td] [td]

62​

[/td] [td]

4.29​

[/td] [td]

4.28​

[/td] [td]

26​

[/td] [td]

30​

[/td][/tr] [tr][td]Interracial[/td] [td]

54​

[/td] [td]

74​

[/td] [td]

57​

[/td] [td]

59​

[/td] [td]

3.84​

[/td] [td]

3.97​

[/td] [td]

16​

[/td] [td]

16​

[/td][/tr][tr][td]Novellas____[/td] [td]

_____53​

[/td] [td]

__56​

[/td] [td]

______6​

[/td] [td]

__23​

[/td] [td]

_____4.53​

[/td] [td]

__4.40​

[/td] [td]

_____74​

[/td] [td]

__55​

[/td][/tr] [tr][td]correlation[/td] [td]

0​

[/td][td].87[/td] [td]

0​

[/td][td].88[/td] [td]

0​

[/td][td].86[/td] [td]

0​

[/td][td].81[/td][/tr]

The correlations between the two sets of data are quite high (> 0.8) suggesting a certain stability in the site's operation. Authors are writing stories in the different categories at about the same rates now as then, and with about the same percentages of stand-alone vs chapter stories. Readers are rating the different categories at about the same levels now as then. They tend to vote up in Sci-Fi/Fantasy, Romance, and Novella and vote down in Loving Wives. As you point out, the fact that the SF&F and N&N categories having lots of multi-chapter stories with dedicated readerships undoubtedly contributes to this.

So, if nothing else, analyses like these confirm and clarify our hunches about the way things work.

Click to expand...

Thanks for doing this. And thanks for figuring out how to do tables in the LitE forums. I stole your table formatting for my OP.

There's definitely a lot of noise in the data. The quality of stories is going to vary from week to week. The stand along/chapter split will vary. I'm kind of surprised to see the data match up as well as it did given all that noise.

8letters said:

And thanks for figuring out how to do tables in the LitE forums. I stole your table formatting for my OP.

Click to expand...

How does one do this? I tried setting up a table in a post and it didn't work. I'm curious.

SimonDoom said:

How does one do this? I tried setting up a table in a post and it didn't work. I'm curious.

Click to expand...

Quote the first post and you'll see how it's done. It's a table. If you're familiar with the table concept in HTML, it's similar. You create a bunch of rows, all of which have the same number of cells. You have to have entries in each cell of the table, so "." is used in blank cells. You can specify the justification of each cell. The forum software will automatically size the columns of the table.

8letters said:

There's definitely a lot of noise in the data. The quality of stories is going to vary from week to week. The stand along/chapter split will vary. I'm kind of surprised to see the data match up as well as it did given all that noise.

And thanks for figuring out how to do tables in the LitE forums. I stole your table formatting for my OP.

Click to expand...

Yeah, I was surprised too that the data matched up so well. When you get correlations > 0.8 it means there really is some signal in spite of the noise.

SimonDoom said:

How does one do this? I tried setting up a table in a post and it didn't work. I'm curious.

Click to expand...

The code is actually a subset of html code described here. I don't know of any way to do it automatically. I usually get the data the way I want it in a spreadsheet, then copy and paste-without-formatting into a text editor to get tab delimited text. Then I copy and paste the appropriate html code at the beginning and end of each line. Then I find and replace all the tabs with the html code to finish up the cell on the left and begin the cell on the right. The column and row headings often have different justification than the data cells, and have to be coded separately.

I make horizontal rules by underscoring and using the underscore character (since leading blanks between [u] and [/u] are ignored). 8letters also uses rows of hyphens. I control the column separation by putting in more or fewer underscore characters.

Then I make sure to delete all the "newline" characters so that the text all runs together in a single paragraph. Otherwise the whole table will be shifted down on the page.

Then I go into the Reply to Thread page on Lit, paste the table code into the message box, click Preview Post, and proceed to fix code errors and tweak the headings and rules to get things looking the way I want. A couple dozen iterations usually gets the job done.

lovecraft68 said:

Those averages are down from a couple years ago, maybe three at this point. So are votes/views/comments...in general as in some authors will always get the numbers, I'm talking your everyday lit story.

Where I really see it is on the new story lists and the average scores and votes the stories get compared to awhile back.

Why is it? That I don't know and wouldn't try to guess.

Click to expand...

NotWise said:

I don't disagree with you. It seems to me like things were different when I first started posting stories in 2015.

The "random" stories list in each hub gives a way to sample past stories. I've used it before to see if the percentage of stories over 4.5 has changed. I was looking mostly at I/T (where I suspect readers have become more critical) and at a couple hubs where the percentage of stories over 4.5 is very high.

It might be interesting to take a body of random samples, sort them be date and see if there have been changes. You can get every statistic there that you can get off the current hub.

Click to expand...

So I used the Search Stories page to find all the IT stories sorted by date, then went through and counted how many were published each year. I did the same for I/T stories with red Hs. Here are the results:

[tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td].[/td] [td].[/td] [td].[/td] [td][/td] [td][/td] [td][/td][/tr]

Basically the number of stories per year increased only very modestly from 2003 to 2013. It's increased somewhat more since then, and may still be accelerating. The percentage of Red Hs has remained fairly constant, varying between about 30% and 40%. There doesn't seem to be any pronounced trend to the Red H percentages. If anything, the last four years have had higher than average Red H percentages.

I didn't look at the view, vote, favorite, or comment counts. I don't think it's possible to get the vote counts of arbitrary older stories or historical view, favorite, or comment counts (e.g., the counts of an arbitrary story when it had been up for 1 year). It would be possible to get the current view, favorite, and comment counts, but it would take a lot of work for 43k stories. Plus, it's not obvious how to compare this data between newer and older stories.

So while this data doesn't say anything about readership, it seems to me that it does say that story production and ratings in this category haven't really changed that much over the lifetime of the site.

HectorBidon said:

So while this data doesn't say anything about readership, it seems to me that it does say that story production and ratings in this category haven't really changed that much over the lifetime of the site.

Click to expand...

Well, there is that little factor of two difference between the % over 4.5 in 2000-2001 and more recent times.

2015 (when I started) gave us the highest percentage of red H's, but it really hasn't slipped that much since. See? I shouldn't trust my gut.

AwkwardMD said:

Since this topic has come up from a number of angles, I’d like to throw a curious case study into the mix. On 4/14/18, I posted a story called The Favor, and the following day (4/15/18) I posted another story called Derelict 0006.

Click to expand...

This is indeed a very interesting case study that shows that two stories that are alike in many ways (same category, same author, same month of publication) can end up with quite different viewing and voting histories. I tried to extend this study by looking at all the stories published in Lesbian Sex during that month. This amounted to 83 stories in all (an average of 2.8 stories per day), of which 46 (55%) have red Hs. These statistics are right in line with the yearly averages for the category (10/20/2017-10/19/2018, obtained from Search), which were 2.7 stories/day and 56% red H's.

The view counts for the stories ranged considerably, from 98k down to only 1.9k, as shown in Fig. 1. The counts were distributed fairly evenly across the range so that there wasn't really any value that could be considered typical. The mean number of views was 13.8k, but any given story might have considerably more or considerably fewer. AMD actually had several stories appear during the month, which are marked by the red and purple bars. "Favor" was toward the high end of the distribution and "Derelict" toward the low end, but they didn't stick out in any particular way.

The fact that the view counts are so widely distributed shows that the readers didn't just view these storied indiscriminately. They were more attracted to some than to others.

One factor that could affect views is author prestige. The five most viewed stories in the list stood out from the others in having considerably more views. The top story got twice as many views as the second. It is "Could you be mine?" by careythomas. She is a relatively new author (joined January 2017) with 605 followers and 10 red H's out of 10 stories. Moreover, 7 of her stories are in the Lesbian Sex Hall of Fame(!), including the story on our list and its sequel, which is in fact the top story in the Hall of Fame(!!). So it's pretty clear that the exceptionally high view count for her story is due to the her superstar status. The authors of our five most viewed stories are here:

[tr][td]Views(k)[/td] [td]Author_______[/td] [td]__Joined[/td] [td]_Followers[/td] [td]_Stories[/td] [td]__Red H[/td] [td]H Fame[/td][/tr] [tr] [td][/td] [td]careythomas[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td]silkstockingslover[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td]Limentina[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td]QuartzErotica[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr] [tr][td][/td] [td]vanillawithcinnamon[/td] [td][/td] [td][/td] [td][/td] [td][/td] [td][/td][/tr]

So careythomas is a Superstar, silkstockingslover is a Legend, Limentina is a Hall of Famer. That probably accounts for the high viewing of their stories. However, silkstockingslover had one other story on our list, and Limentina had two other stories, and they didn't get nearly as many views. Furthermore, QuartzErotica and vanillawithcinnamon are relatively unknown authors, yet their stories got considerably more views than other stories on our list by authors with considerably more followers. So, author fame isn't the whole story.

Another factor that can affect views is whether the story is a stand-alone story or a chapter story. The stories on our list that were chapters >1 are shown in the plot by the yellow and purple bars. Many of the least-viewed stories were chapter stories, but quite a few of the chapter stories had view counts in the mid-to-high range too. So chapter vs stand-alone doesn't tell the whole story either.

AMD brings up the day of the week on which the story comes out as another possible factor. This would probably have the greatest effect during the first week when the stories are going on and coming off the New lists. However, there is no obvious indication that the stories on our list that came out on any particular day tended to have more or fewer views than those that came out on any other day.

The possible effect of titles and taglines has also been brought up. This is what the reader sees when he or she is browsing, the "book cover" and "blurb." so to speak, that entice him or her to click. As Simon put it, "Literotica readers have many options, and many have little time to exercise them. Small differences in the attractiveness of title, taglines, and tags can make big differences in views and reads." It's hard to quantify this effect objectively, but here are the titles and taglines of the 4th, 5th, 82nd and 83rd stories on our list, not necessarily in that order:


I'll just say that, to my mind, two of these do a much better job of selling their story than the other two do. I agree with Simon and EB that title attractiveness has an important effect on views.

There are undoubtedly other factors that affect view count as well. AMD suggests that tags may be important. Whether or not a story ends up on "Similar Stories" lists surely has an effect. The bottom line is that view count is a complicated, multifactorial quantity. As much as we want to unlock the secret of Achieving Big Views, it's not that simple.

The other big issue, of course, is what a view even means. This has been discussed a lot in the forum lately, and the consensus opinion is that only a fraction of views translate into someone actually reading the story. There are various estimates about how large the read to view ratio might be.

Whatever the actual value is, if it's consistent across stories it means that, in general, the more views a story gets the more reads it gets too. The data I looked at goes along with this. The number of times the stories on our list were favorited and the number of comments they received were both very highly correlated with the view count (r = 0.88 and 0.85, respectively). It's not possible to obtain the vote counts from the Search page, but I would suspect that they are correlated with the view counts too. So whatever a view actually means, getting more of them means that your story is getting more attention.

The other big thing that the data shows is that the story ratings were not closely correlated with the view counts (r = 0.26). This is shown in Fig. 2. Most of the stories with high view counts (including "Favor") received pretty good ratings. However, some of the more highly viewed stories weren't rated as highly. They attracted too many eyeballs, perhaps, and failed to deliver what the readers were looking for. On the other hand, some of the less viewed stories (including "Derelict") also received very high ratings. These were presumably stories that found their target audience. The chapter 1 stories pretty much fell into the pack with the stand alone stories. The chapter >1 stories tended to be less viewed but highly rated, suggesting that they're being followed by a loyal audience. However a couple chapter >2 stories received lower ratings. Just like there can be clunker stories, there can be clunker series.

Fiddling with all these statistics is enjoyable (for me at least), but in the end what they show is that my little blue squares aren't that much different from anybody else's little blue squares amidst the vast scatterplot of Literotica. In the words of another Legend (>2087 followers, >1140 stories):

KeithD said:

When I'm done writing one, I just move on to writing the next one. The relative reception of stories on Literotica is a mostly meaningless and manipulatable crap shoot.

Click to expand...

Still, though, what I'd tell an aspiring writer is this. Lit provides a huge audience. If you post a story it will get read. If you write good stories and title them appropriately, there's a good chance that at least a few appreciative readers will find them.

Kudos to 8Letters and Hector Biden. This is the best and most informative thread I've seen yet on story category stats.

so it turns out some really smart people have looked at a similar case with regards to music downloads:

Experimental Study of Inequality and Unpredictability in an Artificial Cultural Market
Matthew J. Salganik, Peter Sheridan Dodds, Duncan J. Watts

Abstract
Hit songs, books, and movies are many times more successful than average, suggesting that “the best” alternatives are qualitatively different from “the rest”; yet experts routinely fail to predict which products will succeed. We investigated this paradox experimentally, by creating an artificial “music market” in which 14,341 participants downloaded previously unknown songs either with or without knowledge of previous participants' choices. Increasing the strength of social influence increased both inequality and unpredictability of success. Success was also only partly determined by quality: The best songs rarely did poorly, and the worst rarely did well, but any other result was possible.

astuffedshirt_perv said:

so it turns out some really smart people have looked at a similar case with regards to music downloads:

Experimental Study of Inequality and Unpredictability in an Artificial Cultural Market
Matthew J. Salganik, Peter Sheridan Dodds, Duncan J. Watts

Click to expand...

This is a really interesting study. What they did was set up a website where people could rate songs. They put up 48 songs from unknown bands and then let people rate them. Some of the people could see the ratings of the people who had voted before them, some couldn't. This setup isn't exactly the same as Lit, but it's similar in several ways, and I think it offers the following insights:

1. The fact that readers on Lit can see how other readers have already voted affects the results. The paper refers to this as "social influence." If we wanted a more objective measure of a story quality, it would be better to have readers rate stories independently of everyone else. Social influence on Lit tends to amplify the difference in the "success" of stories that are not really that different in quality. Stories A and B may be of equal quality, but A gets a lead in early voting, and social influence tends to make this lead even bigger. However, it could just as well have gone the other way around. B could have gotten the early lead and ended up being more successful. The paper makes the point that experts in the music industry can't always predict which songs will be big hits because this depends as much on the way that social influence plays out as on any objective measure of the song itself.

Of course this doesn't mean that there's no relation between a story's score and its quality. A 4.6 story is probably "better" than a 3.6 story in the same category. But it's perhaps not necessarily all that much better than a 4.4. Authors like me can rest assured that there are alternative universes in which we have more little red H's than we do in this one.

2. Of course it's not so easy to compile data from other universes. In the music study, they divided their social influence subjects into 8 groups who only had access to rating information from people in their same group. In this way they basically ran the same experiment 8 times in parallel and were able to see social influence playing out differently in the different groups.

Astuffedshirt_perv looked at this in a different way by looking at several stories that, for one reason or another, had been published on Lit more than once (Voting scores and double posts). She found that the same story can receive a rating that differs by as much as +/- 0.5 points when it appears a second time.

Some authors have taken stories down and then reposted them later. Perhaps they've compiled data about how they fared the second time compared to the first?

3. The music study used a five star rating system like in Lit, but they measured a song's success in terms of what they called its "market share." Once someone had rated a song, they could select to download it for later personal use. (This was actually the "compensation" for participating in the study.) Each song's success was measured in terms of the number of people who downloaded it. This seems like a pretty legitimate measure since people are only going to download songs they really like. Lit doesn't really have anything comparable except perhaps "favoriting." I haven't paid too much attention to favoriting in the past, but now I'm wondering whether it might be a good way (alongside "rating") to measure a story's success. Votes and comments can be negative as well as positive, but hearts are presumably only positive, and so perhaps a clearer indication of popularity.

The music study didn't mention what we would call "view count." The difficulty of interpreting view counts has been discussed at length in the forum lately.

4. Categories, relentless publication of new stories, one bombing, author reputations, and long-term effects are some of the Lit issues that don't feature in the music study.

HectorBidon said:

...

2...

Astuffedshirt_perv looked at this in a different way by looking at several stories that, for one reason or another, had been published on Lit more than once (Voting scores and double posts). She found that the same story can receive a rating that differs by as much as +/- 0.5 points when it appears a second time.

Some authors have taken stories down and then reposted them later. Perhaps they've compiled data about how they fared the second time compared to the first?

...

4. Categories, relentless publication of new stories, one bombing, author reputations, and long-term effects are some of the Lit issues that don't feature in the music study.

Click to expand...

I posted two almost identical stories a few months apart. It was my own error. I had upgraded my computer, transferred files, and not realised that I had completed and posted a version of the story I had as shown as 'pending'.

I have left both stories on Literotica as an interesting experiment. The two stories are:

Escaping Cybele

Posted 10 April 2014 Rating 4.40; Votes 58; Views 14657; PCs 4

Limit of Authority

Posted 14 July 2014 Rating 4.51; Votes 318; Views 19123; PCs 4

oggbashan said:

I posted two almost identical stories a few months apart. It was my own error. I had upgraded my computer, transferred files, and not realised that I had completed and posted a version of the story I had as shown as 'pending'.

I have left both stories on Literotica as an interesting experiment. The two stories are:

Escaping Cybele

Posted 10 April 2014 Rating 4.40; Votes 58; Views 14657; PCs 4

Limit of Authority

Posted 14 July 2014 Rating 4.51; Votes 318; Views 19123; PCs 4

Click to expand...

You published them in two different categories, which means they were read by different groups of people with different preferences, and that might explain the rating difference.

The screwiest thing about those results is the extremely different vote:view ratios. That makes no sense to me.

The social influence factor at this site is compounded by the existence of favorites and follower lists, and toplists. Stories that show up on such lists get seen more, and read more, and in turn favorited more, and the process, I assume, compounds.

I published very similar standalone I/T stories one month apart in 2017. The second was a sequel to the first, but you can't tell that from the title or tagline. The second story has a higher score by .03 to .04, but it has less than one-third the views and favorites as the first story. What's even stranger is that the second story now receives about 15% the daily views that the first story receives, so the gap between them is growing despite the fact the second story has a higher rating. I suspect the number of favorites is a significant factor in explaining the difference.

SimonDoom said:

You published them in two different categories, which means they were read by different groups of people with different preferences, and that might explain the rating difference.

The screwiest thing about those results is the extremely different vote:view ratios. That makes no sense to me.

...

Click to expand...

The first one I put in Fetish. I know, as far as my stories are concerned, that Fetish has a low vote ratio. I assume that is because a Fetish story either fits your particular fetish scenario, or not.

The second one I put in Romance. That has a wider readership and I think the readers of Romance are less critical than Fetish.

But I agree. Those results make no sense.