Ratings

[binderman91]

What is a decent rating on a story
? This is based on the 1-5 scale. I would like to hear your comments

 

[Djmac1031]

I hear people complain about anything under 4.5 because it prevents them from getting a coveted Red H.

But I think a 4 is a perfectly acceptable rating.

4 means the vast majority liked your story.

 

[Trionyx]

It varies somewhat by category. A 4 in Loving Wives is quite good but in Romances it would not be anything remarkable.

 

[ronde]

What is a decent rating on a story
? This is based on the 1-5 scale. I would like to hear your comments

Anything over a 4 is doing pretty well. I get comments all the time from readers who say they never vote higher than a 4 unless the story is somehow exceptional to them.

 

[MelissaBaby]

Anything over a 4 is doing pretty well. I get comments all the time from readers who say they never vote higher than a 4 unless the story is somehow exceptional to them.

I think that is a reasonable approach. If you gave every story you enjoyed a five, what do you do when you find one you think is really special?

 

[Jackie.Hikaru]

I think that is a reasonable approach. If you gave every story you enjoyed a five, what do you do when you find one you think is really special?

I'll only leave a 5 if I liked it. If I think it's special I'll comment. If it's extra special I'll write the author a pm to let them know they got me all hot and bothered.

 

[TheRedChamber]

The truth is that the average in most categories is over 4. The exceptions are Interracial, Erotic Horror, Celebrities and especially Loving Wives (based on data from 8Letter which is a few years old).

After that, while it's fine to aim for 4.5 and the H, the score you're going to get is as much about the proclivities, your writing style and giving people what they want as it is about how good you are as a writer. Oh, knocking your story out of the park will raise you score, but you are not going to please all the people all the time and if you piss off 1 out 7 people you're not going to get to a red H.

It's probably best to write with a view to finding out what readers think of you, then either playing into it or doing your own thing.

 

[AchtungNight]

I think a 3 or higher is acceptable. However I never rate stories here below 4 and I have locked voting on all my stories because I can’t stand to go below three myself on anything. After posting in Loving Wives and losing my longest standing highest rating… I’m done being a slave to ratings. Positive reviews still appreciated, mind you.

 

[Euphony]

I hear people complain about anything under 4.5 because it prevents them from getting a coveted Red H.

But I think a 4 is a perfectly acceptable rating.

4 means the vast majority liked your story.

Some of the most breathtaking pieces I've read here fall between 4.0 and 4.5.

Sometimes readers don't know what to make of things flowing outside the usual currents of the category.

No HEA penalties can weigh down too.

Scores are an imperfect tool. Trust and author focus can unearth real gems hiding below the H.

 

[KeithD]

I think that is a reasonable approach. If you gave every story you enjoyed a five, what do you do when you find one you think is really special?

Leave a really-special comment.

 

[EmilyMiller]

Leave a really-specially comment.

Agree. Comments are real gold. Positive ones obviously. But well-thought-out negative ones as well.

Em

 

[EmilyMiller]

What is a decent rating on a story
? This is based on the 1-5 scale. I would like to hear your comments

I’m as much of a rating whore as the next girl. But that’s just an immediate reaction. What I’ve learned is that ratings are a really poor reflection of anything, unless you have 100 or more votes. What say 30 people think has such inherent variability as to be close to meaningless. And that ignores people who grace you with ones as you have somehow offended them, maybe by just existing.

Don’t sweat ratings.

Em

 

[Euphony]

Voters are inherently lazy. There's a lot of psychological inertia to overcome to get a reader to think with nuance on their current reading experience and past ones.

Smashing 1 bad 5 good is far less mentally taxing for most. So scores drift away from midpoint.

Unless you can somehow get agreement on how, at least most, will use the system, it will always be terribly muddled and should be viewed accordingly.

 

[StillStunned]

What I’ve learned is that ratings are a really poor reflection of anything, unless you have 100 or more votes.

Now I'm sad. "Flesh for Fantasy" has been stuck on 98 ratings for a while.

 

[8letters]

Ratings are strongly influenced by category and page length

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.

 

[Euphony]

Now I'm sad. "Flesh for Fantasy" has been stuck on 98 ratings for a while.

 

[AG31]

Ratings are strongly influenced by category and page length
View attachment 2287828
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.

How do you gather this data? Surely not one by one by hand?

 

[Bamagan]

Given that pretty much the only certain thing about the ratings is that bigger=better, high scores are probably more about audience targeting than anything else. So any score that allows you to taunt another author because yours is higher is therefore a 'decent' one.
Boring stuff no one wants to read follows:
Even 100 votes isn't necessarily very meaningful, although it kind of depends on how big the 'population' of readers is for any given story (which you can only estimate from views). Also, the people who vote aren't necessarily representative of all readers, so that's another layer of uncertainty about the 'true' rating. If you get 400 votes then you might be reasonably confident that your score is accurately reflecting your audiences opinions, even allowing for the likelihood of some sample bias in the voter pool.
On the other hand, most of the authors posting here report that 1%-3% votes-to-views is typical for them. Scores would plummet if we took that level of engagement to infer that a significant percentage of the non-voters would have assigned a rating of 1 (backed out quickly because they could see it wasn't for them), 2 (read for a while but disengaged before getting to the bottom/last page), or 3 (finished but had no major reaction). I'm not saying we should, and I don't think the data that can be gleaned from existing voter patterns can be used to infer anything significant about how to apportion the opinions of non-voters. But if they more-or-less averaged out to a rating of 2.XX, the scores on the site would look a lot closer to 3 (the theoretical average or mean score) than they do currently.

 

[8letters]

How do you gather this data? Surely not one by one by hand?

Something I wrote goes out and gathers data every day. It's not great stuff, but it works.

 

[AG31]

Something I wrote goes out and gathers data every day. It's not great stuff, but it works.

I assume, then, that you are some sort of programmer. Can I ask how you find out about the underlying structure of literotica.com? Is it published somewhere? I'm a programmer from back in the day when Dartmouth was inventing Basic. I never did get into the internet. But I remain curious.

 

[MillieDynamite]

I write with the hope it will evoke an emotional reaction. Love or hate are both acceptable reactions. So, to me, essentially, the rating isn't all that important. Yes, I like having a red H, but I don't need it to feel good about a story. I'm not the best writer. I'm not the worst. Where I fall between the two extremes isn't something I can calculate. I'm adequate for the type of writing I do, and I hope my work improves with each story. All quibbling about scores doesn't change the kind of writer, the quality of the work, or type of story we tell. Putting too much stock in votes only deflates or inflates your opinion of yourself.

Well, that's my opinion, and I value my opinion over other folks views. For better or worse, I am what I am.

 

[8letters]

I assume, then, that you are some sort of programmer. Can I ask how you find out about the underlying structure of literotica.com? Is it published somewhere? I'm a programmer from back in the day when Dartmouth was inventing Basic. I never did get into the internet. But I remain curious.

There is no underlying structure to Literotica. What the rating is of a story varies depending on how you look it up. I've gotten different ratings for a story on the search page based on how I searched. I take all the numbers I can find, and then I try to pick the most reasonable value.

 

[onehitwanda]

I assume, then, that you are some sort of programmer. Can I ask how you find out about the underlying structure of literotica.com? Is it published somewhere? I'm a programmer from back in the day when Dartmouth was inventing Basic. I never did get into the internet. But I remain curious.

It's very easy to work out the structure of the site from the html. Every story [in the new layout] has a small section of html with the following id: tabpanel-info

nested within this will be the following elements:

<div class="aT_H" title="Rating"><i class="icon icon-star aT_ck"></i><span class="aT_cl">4.38</span></div>
<div class="aT_H" title="Views"><i class="icon icon-diagram aT_ck"></i><span class="aT_cl">1.4k</span></div>
<div class="aT_H" title="Favorites"><i class="icon icon-heart aT_ck aT_rn"></i><span class="aT_cl">7</span></div>
<div class="aT_H" title="Comments"><i class="icon icon-comment aT_ck aT_ro"></i><a title="Link to comments" class="aT_cl aT_mG" href="/s/${story-identifier}/comments">15</a></div>

there are a multiplicity of html parsing libraries that can extract these values.

Simplistically, what you'd do is:

1. index the new page
2. extract links to all stories listed there
3. visit each story page, extract the data above and write it to a database.

it's a couple of hours work for a competent software engineer. The hardest part is not running foul of any heuristics detection that sees spammy requests coming from a single ip or ip address range.

oh. shit. I put my geek hat on for a moment and look what happens.

 

[joy_of_cooking]

If I understand correctly, there's another hard part, which is that Literotica shows lots of different snapshots of the same value, because different parts of the site get updated on their own schedules.

 

[8letters]

If I understand correctly, there's another hard part, which is that Literotica shows lots of different snapshots of the same value, because different parts of the site get updated on their own schedules.

I think the inconsistencies in numbers have to do with other things like sweeps. For example, let's talk about A Mirror in the Dark Ch. 02:

What's the rating? 4.75, right?

Or maybe not:


I've been pulling the 30-Days Top List data, and here's what it is for that story:

So, on 10/30, it had a rating of 4.75 on 20 votes. It looks like some votes were swept, and the rating jumped up to 4.89 while the number of votes dropped by 1. That sweeping seems to have frozen the stats on the story to that of 10/30. If I do a search for the story:

It shows more views and more favorites than what the story shows.