Unexpected change in ratings + votes

Writer61

Englishman abroad
Joined
Feb 17, 2024
Posts
940
I have kept track of the rating and number of votes for my last ten stories. Since each only gets a few votes per week, you can be completely accurate if you work incrementally. I usually do it every week.

Today, I noticed an unexpected change. The number of votes for 9 of the 10 has decreased by 1, and their average ratings have increased. My first thought was that a sweep had eliminated the 1- and 2-star ratings, but a) not all the stories had any of those, and b) mathematically, they are still present. Instead, it is threes and fours that have disappeared. I should note that the 10th story might also have had a vote deleted and, by chance, have received another with the same rating.

It looks methodical, but not like a traditional sweep.

Any ideas?
 
Despite some common wisdom, sweeps do not just eliminate low scores. They eliminate what the algorithm decides are suspicious votes. I will not speculate on what the algorithm looks for (and they ask us not to do so publicly). I'm sure it is not always accurate. But the vast majority of suspicious votes turn out be 1's, so it almost always helps. I did have one story actually drip with this sweeps season (a low vote story that lost a single 5 and nothing else). I think, on the whole, the post-sweeps ratings are more accurate than before the sweeps. Getting a prefect result of the sweeps would need to read into the mind of every voter. Who knows the true rating? Only the shadow knows.
 
The problem I've got with sweeps is they always come too late. If your one bombed on the first day that low score discourages a lot of potential readers. By the time a sweep comes around it's too late.
 
For example: has anyone ever figured out whether or not a reader's votes get removed if they delete their account?
That crossed my mind. I think it is unlikely to work that way because of the extra computation effort, but I could be wrong.

For nine stories to be affected, I would expect it to be a follower, but there has been no change in those in the last week.
 
But the vast majority of suspicious votes turn out be 1's
<adopts Sith persona>Many are also 5s from ‘over enthusiastic’ authors. But yes sweeps don’t care about the rating value just their accompanying characteristics. They are ultra-basic and it seems more and more people are figuring out how to avoid them (it’s pretty simple).

Sweeps are seen as something like a religious totem around here. They are a Band-Aid on a severed jugular. Performative rather than effective, or - heaven forfend - addressing the problem. </adopts Sith persona>

Did I do well, oh my master @AwkwardlySet ?
 
<adopts Sith persona>Many are also 5s from ‘over enthusiastic’ authors. But yes sweeps don’t care about the rating value just their accompanying characteristics. They are ultra-basic and it seems more and more people are figuring out how to avoid them (it’s pretty simple).

Sweeps are seen as something like a religious totem around here. They are a Band-Aid on a severed jugular. Performative rather than effective, or - heaven forfend - addressing the problem. </adopts Sith persona>

Did I do well, oh my master @AwkwardlySet ?
1765728732234.png

And keep in mind that, even if you throw me down a reactor shaft, I'll still return!

Somehow.
 
<adopts Sith persona>Many are also 5s from ‘over enthusiastic’ authors. But yes sweeps don’t care about the rating value just their accompanying characteristics. They are ultra-basic and it seems more and more people are figuring out how to avoid them (it’s pretty simple).

Sweeps are seen as something like a religious totem around here. They are a Band-Aid on a severed jugular. Performative rather than effective, or - heaven forfend - addressing the problem. </adopts Sith persona>

I can only comment on what I have seen on my stories.

I've been pleasantly surprised how well it works overall for cleansing 1's from my contest entries. I have not have as sophisticated enemies yet. In both contest, my entry lost one or two more ones than I was even aware of (plus a handful of 4's and 5's). It has also cleaned up some attacks on non-event stories fo rme as well.

What it seems to totally miss is all the attacks for winning a prize or for making a leaderboard. I have not seen a single one of those bombs swept. Sufficiently strikingly different that either the sophistication of the attacks is quantitatively different OR the site is intentionally ignoring the prize/leaderboard bombs. I would hope for the former, but it would surprise me that the contest bombers (who attack me at least) were fundamentally different from the prize/leaderboard bomber(s)
 
Last edited:
Sweeps are seen as something like a religious totem around here. They are a Band-Aid on a severed jugular. Performative rather than effective, or - heaven forfend - addressing the problem. </adopts Sith persona>

I see no evidence of that at all. Nobody treats sweeps like a religious totem. I think everybody acknowledges that they are an imperfect, and possibly game-able, way to weed out some "illegitimate" votes. Where people disagree with you, and a few others, is on whether this is really that big a problem or not. Some of us, maybe many of us, think downvoting is not something worth worrying about and don't understand all the fretting about it, especially when many of those who fret the most are doing just fine and getting the recognition they deserve for their stories.

People are free to disagree. But one doesn't establish that downvoting is objectively a major problem just by insisting, over and over, that one is upset by it. Some of us aren't upset by it. I suspect the vast majority of authors who never contribute to this forum don't give it any thought at all.

Also, my experience with sweeps is different from yours. My experience is that over time sweeps seem to restore story scores to about where they should be. Since they are intermittent, they don't prevent whatever adverse effect downvoting has in the short term on a story's performance, but they seem reasonably effective in the long term, according to my appraisal of my stories' scores and the scores of other people's stories. I have yet to see the case of a single author, other--possibly-- than KeithD, whose long-term story scores appeared to me to be adversely and unfairly affected by deliberate downvoting compared to what I would have expected them to be. And I'm not even sure in his case.
 
I see no evidence of that at all. Nobody treats sweeps like a religious totem. I think everybody acknowledges that they are an imperfect, and possibly game-able, way to weed out some "illegitimate" votes. Where people disagree with you, and a few others, is on whether this is really that big a problem or not. Some of us, maybe many of us, think downvoting is not something worth worrying about and don't understand all the fretting about it, especially when many of those who fret the most are doing just fine and getting the recognition they deserve for their stories.

People are free to disagree. But one doesn't establish that downvoting is objectively a major problem just by insisting, over and over, that one is upset by it. Some of us aren't upset by it. I suspect the vast majority of authors who never contribute to this forum don't give it any thought at all.

Also, my experience with sweeps is different from yours. My experience is that over time sweeps seem to restore story scores to about where they should be. Since they are intermittent, they don't prevent whatever adverse effect downvoting has in the short term on a story's performance, but they seem reasonably effective in the long term, according to my appraisal of my stories' scores and the scores of other people's stories. I have yet to see the case of a single author, other--possibly-- than KeithD, whose long-term story scores appeared to me to be adversely and unfairly affected by deliberate downvoting compared to what I would have expected them to be. And I'm not even sure in his case.
I general I’d say you’re correct, and while it being personal is questionable, it’s difficult not to think a story starting at 4.9+ with the first twenty or so votes(obviously followers), then dropping to 4.63 with a small handful of new votes, and working back up to 4.8 with 450+ votes, jumping to 4.88 after a sweep and only losing 12 votes isn’t malicious.
 
I general I’d say you’re correct, and while it being personal is questionable, it’s difficult not to think a story starting at 4.9+ with the first twenty or so votes(obviously followers), then dropping to 4.63 with a small handful of new votes, and working back up to 4.8 with 450+ votes, jumping to 4.88 after a sweep and only losing 12 votes isn’t malicious.

It may be. But if one's story ends up with a long-term score of 4.88, whatever the system is, I don't think one can plausibly complain that the system is in any objective sense "broken."

The system, as far as I can tell, has been in place for the site's 27 year existence. We all know what it is, and we're all subject to it. They could have chosen a different system, but they chose this one. The emphasis is on maximizing the ability of readers to access, read, vote on, favorite, and comment on stories. That's a plausible, defensible goal, and it may well be one of the reasons this site has more traffic, more readers, and more stories than other erotic story sites do. It comes at a cost in the form of some element of voting gamesmanship, but the site obviously has decided that this is an acceptable cost, and, objectively speaking, they're not wrong.
 
It may be. But if one's story ends up with a long-term score of 4.88, whatever the system is, I don't think one can plausibly complain that the system is in any objective sense "broken."

The system, as far as I can tell, has been in place for the site's 27 year existence. We all know what it is, and we're all subject to it. They could have chosen a different system, but they chose this one. The emphasis is on maximizing the ability of readers to access, read, vote on, favorite, and comment on stories. That's a plausible, defensible goal, and it may well be one of the reasons this site has more traffic, more readers, and more stories than other erotic story sites do. It comes at a cost in the form of some element of voting gamesmanship, but the site obviously has decided that this is an acceptable cost, and, objectively speaking, they're not wrong.
I’m certainly not one to complain about ratings. I inferred from your previous post a minimization of malicious voting, and was simply trying to illustrate that it does seem to exist. As you say, the system is what it is. Without in-depth knowledge of how it is implemented, all I can suggest is that, since they have a sweep algorithm, perhaps they front load it, making it proactive. There may be technical reasons for not doing this but only they know. They vette comments, so…
 
It may be. But if one's story ends up with a long-term score of 4.88, whatever the system is, I don't think one can plausibly complain that the system is in any objective sense "broken."

The system, as far as I can tell, has been in place for the site's 27 year existence. We all know what it is, and we're all subject to it. They could have chosen a different system, but they chose this one. The emphasis is on maximizing the ability of readers to access, read, vote on, favorite, and comment on stories. That's a plausible, defensible goal, and it may well be one of the reasons this site has more traffic, more readers, and more stories than other erotic story sites do. It comes at a cost in the form of some element of voting gamesmanship, but the site obviously has decided that this is an acceptable cost, and, objectively speaking, they're not wrong.
I always find these threads give me a chuckle. If you were to go back to this same forum in 2001, you'll find the same threads with the same comments about "1 bombs" and a whole host of reasons why voters vote like they do. There were also discussions about "5 bombs" where supposedly the clique of people who favor one author all conspire to keep that author's story top rated. One of the reasons for the "5 bomb" theorized was that it was college age readers taking advantage of the number if IP addresses available on a college campus that let each member vote multiple times.

It was threads like those that resulted in the "sweeps", but other than giving some people personal satisfaction that they were able to change something, they don't do much to the scores. The biggest change to any of my ratings I've seen is -0.02 but I've also seen a change of +0.02. Of course the impact is affected by the total number of votes a story has, but losing a couple "1" votes out of a hundred won't have much impact. For example, a story rated as 4.5 with 100 votes , 2 of which are "1" votes only increases to a rating of 4.57 if those "1" votes are removed. If the two votes removed are "5" votes, the rating changes to 4.49. The more total votes, the less the effect.

Yes, those changes are enough to change the placeent in a story contest or the award of a red "H", and I already know my numbers aren't absolute because Literotica rounds off the scores to two decimal places, but we'd all be better off working to improve our writing than spending time debating a rating system that has been around since the site's inception and in my experience is better than the rating systems on other sites.
 
I always find these threads give me a chuckle. If you were to go back to this same forum in 2001, you'll find the same threads with the same comments about "1 bombs" and a whole host of reasons why voters vote like they do. There were also discussions about "5 bombs" where supposedly the clique of people who favor one author all conspire to keep that author's story top rated. One of the reasons for the "5 bomb" theorized was that it was college age readers taking advantage of the number if IP addresses available on a college campus that let each member vote multiple times.

It was threads like those that resulted in the "sweeps", but other than giving some people personal satisfaction that they were able to change something, they don't do much to the scores. The biggest change to any of my ratings I've seen is -0.02 but I've also seen a change of +0.02. Of course the impact is affected by the total number of votes a story has, but losing a couple "1" votes out of a hundred won't have much impact. For example, a story rated as 4.5 with 100 votes , 2 of which are "1" votes only increases to a rating of 4.57 if those "1" votes are removed. If the two votes removed are "5" votes, the rating changes to 4.49. The more total votes, the less the effect.

Yes, those changes are enough to change the placeent in a story contest or the award of a red "H", and I already know my numbers aren't absolute because Literotica rounds off the scores to two decimal places, but we'd all be better off working to improve our writing than spending time debating a rating system that has been around since the site's inception and in my experience is better than the rating systems on other sites.

I've seen VERY significant changes to the scores of my stories as a result of sweeps within the first month of publication. On one occasion I placed in a contest as a result. In other cases I've seen other contest stories vault over mine as a result.

Whatever. I happily admit I'm less technologically and mathematically capable than many others who contribute here, and my brain hurts if I try to wander into the weeds of HOW it works too much.

Your last point, I think, is inarguable. We KNOW they aren't going to change the system, so you can accept it, or not. But there's not much point served in grousing about it too much. Unless one likes to grouse.
 
I've seen VERY significant changes to the scores of my stories as a result of sweeps within the first month of publication. On one occasion I placed in a contest as a result. In other cases I've seen other contest stories vault over mine as a result.

Whatever. I happily admit I'm less technologically and mathematically capable than many others who contribute here, and my brain hurts if I try to wander into the weeds of HOW it works too much.

Your last point, I think, is inarguable. We KNOW they aren't going to change the system, so you can accept it, or not. But there's not much point served in grousing about it too much. Unless one likes to grouse.
Nah, not in AH...🙊
 
I can only comment on what I have seen on my stories.

I've been pleasantly surprised how well it works overall for cleansing 1's from my contest entries.
I hope that it is always this way for you. But it’s not for everyone and anyone could be targeted by a Lit-savvy hater at any time for any reason. Not being a victim of something doesn’t mean thing doesn’t exist. As I say I hope your good fortune holds.
 
Back
Top