4.89 at 100 ratings dropped to 4.79 by 104.

[SimonDoom]

Reaching that 100-vote threshold does drastically increase a story's visibility, if it's scoring high enough to make the toplist. I could believe a legitimate drop in the average rating at that point. But what OP described (apparently total 9 stars from 4 votes) sounds a bit too drastic to be from bona fide voting.

I agree but would just caution people about being too certain of conclusions without strong evidence. It's possible for a vote to be motivated by bad intentions yet still be a legitimate reflection of the reader's opinion. We just don't know for sure. In any event, it's something we all face, and I think we're better off accepting that it's part of the landscape here and we have to deal with it. As with so many other aspects of the site, it's not clear to me that there are solutions that would be better than the problem.

 

[MelissaBaby]

I agree but would just caution people about being too certain of conclusions without strong evidence. It's possible for a vote to be motivated by bad intentions yet still be a legitimate reflection of the reader's opinion. We just don't know for sure. In any event, it's something we all face, and I think we're better off accepting that it's part of the landscape here and we have to deal with it. As with so many other aspects of the site, it's not clear to me that there are solutions that would be better than the problem.

I've got this all screenshotted somewhere, but Queen of the Roller Derby hit 100 votes with a score of 4.98. That works out to 98 fives and 2 Fours. For a very brief period, it was #1 on the All Time Top List. 15 minutes later, it had about 5 more votes, and the score was 4.92.

Kind of an amazing coincidence that everyone who read it before it popped up at #1 thought it was worth a high score, and that just at that point, the first One votes come in.

 

[THBGato]

Kind of an amazing coincidence that everyone who read it before it popped up at #1 thought it was worth a high score, and that just at that point, the first One votes come in.

Especially with a long (and brilliant) story like that, there's no way they were legitimate votes. If a reader didn't like it they would just stop before the end.

 

[Wifetheif]

I've been here a long time. I'm happy with a red H and a score over 4.5 BUT I usually hang just under 4.50 at 4.48 0r 4.49 My stories loose and gain red Hs regularly. I'm not really chapped by ratings. Some of my early stories are frankly dreadful. I leave them up to
remind myself to strive and do better. I've had stories read hundreds of thousands of times. NONE of those stories are red H's but they still get viewed and favorited. In the end we are here to get our stories read, to prove that we have some talent and that total strangers will enjoy our stories. I have gotten all of that in spades. The simple fact is the longer a story is up the lower its rating will be with time. Almost NO ONE holds on to a 4.9 forever. Yes red H's cause SOME people to read our stories others go on things like category and writer popularity. A writer with a large number of followers will be read more often than a writer with fewer followers NO MATTER THE RATING OF THE STORY! I've had people tell me I touched their lives through stories I've posted here. Isn't THAT the point? Rating are nice but when you get a response like: "A great read from a very talented writer. Bravos." who gives a crap if it has a red H or not?

 

[Bamagan]

Certain types of bad faith voting are probably undetectable, and certain types of legitimate voting may seem like bad faith from our perspective as authors or readers.

The site has access to all the votes so they can turn various analytical and statistical tools onto that data, which is probably pretty good at picking out the accounts that exclusively or predominantly vote only 1s and 2s. It's not hard to conclude that such an account is designed to troll, and even if it's actually a person who only votes on stories that bother them, they're not likely to be missed if their votes stop getting counted.

It gets a little trickier, perhaps, when deciding what to do about accounts that have the same pattern but their mode is 3, 4, or 5. In the case of 3s and 4s, they're still effective for dragging the score away from the H threshold, which might indicate trollish behavior, but it's also a more believable pattern of behavior for people who enjoy reading erotica in general but are rarely amazed by what's available here. Since 5 is the only 'reward' vote in terms of visible recognition, an account that veers that way probably also gets scrutiny as a possible ballot-stuffer and might also get toggled into vote oblivion. In all of these cases, the site may try to do a second layer of analysis, such as correlation to contest entries or top lists, time spent viewing stories, or ratio of stories viewed to stories voted on. The question of where to draw the line between troll and legitimate but single-minded voter is problematic, though.

But is a person who only votes to show disapproval automatically a troll? If a story contains content a reader finds objectionable, whether that's something fetishy like scat or religiously fraught, are they trolling if they vote a 1? I think most of us would say yes if the story title, blurb, tags, or whatnot make it clear what the reader is in for. Anyone who complains for getting exactly what they should have expected is probably not acting in good faith. I'm not confident that someone voting a 1 because they came across a squick they weren't expecting is voting in bad faith or illegitimately, even if they stop reading at that point and skip to the end to register their distaste. (And, for that matter, how legitimate does that make a vote of 5 on a story simply because it checks a fetish box, or the author is a friend?)

In any case, though, it's difficult (at best) for an algorithm to detect or flag voting patterns like that, much less the reasoning (if any) behind the pattern, so the site might not even try. And if they do, I imagine the false positive rates might mimic the angst of the AI detection situation. And what to do with a voter whose pattern is 50% 1s and 50% 5s? Bot, or someone who only votes like/dislike? At what ratio does that pattern seem inhuman? For that matter, how much of the porn we find here should a human love?

Tl;dr I am a robot, beep boop beep whirr clunk.

 

[RejectReality]

You do absolutely have to remember that making the upper quarter of page 1 of that all-time toplist does also expose you to a whole lot of new readers. Not all of them are going to love it. You're inevitably going to get quite a few "liked it, didn't love it" 4s. Because your score is above 4, every one of those 4s reduces your average score. That's the non-nefarious reason scores can drop when you hit a toplist.

The math here doesn't seem to work out to that kind of light dinging, but you do have to take a moment when you see a drop and determine if that's what's actually happening.

 

[burgwad]

If only ratings were limited to a single kudos/upvote button, a la AO3. Then there would be no way for trolls to affect a story’s standing except to abstain from voting. Site-wide rankings would still emerge, but scores could only climb, not rise and fall. Also, a better, smarter site algorithm that could help to connect readers with stuff they’re apt to like, thereby ideally preventing choosy, opinionated readers from trolling in the first place, would be a huge boon to under-followed talent.

Lit’s rating and ranking architecture is more dog eat dog. I hate it. It promotes genre populism, where only certain kinds of works ever tend to dominate their respective categories. This turns arbitrary features of a work (e.g., brevity, kink selection, body types, etc.) into seemingly necessary sink-or-swim virtues which Anon is only too happy to police, ruthlessly and with abandon.

 

[SimonDoom]

Certain types of bad faith voting are probably undetectable, and certain types of legitimate voting may seem like bad faith from our perspective as authors or readers.

The site has access to all the votes so they can turn various analytical and statistical tools onto that data, which is probably pretty good at picking out the accounts that exclusively or predominantly vote only 1s and 2s. It's not hard to conclude that such an account is designed to troll, and even if it's actually a person who only votes on stories that bother them, they're not likely to be missed if their votes stop getting counted.

It gets a little trickier, perhaps, when deciding what to do about accounts that have the same pattern but their mode is 3, 4, or 5. In the case of 3s and 4s, they're still effective for dragging the score away from the H threshold, which might indicate trollish behavior, but it's also a more believable pattern of behavior for people who enjoy reading erotica in general but are rarely amazed by what's available here. Since 5 is the only 'reward' vote in terms of visible recognition, an account that veers that way probably also gets scrutiny as a possible ballot-stuffer and might also get toggled into vote oblivion. In all of these cases, the site may try to do a second layer of analysis, such as correlation to contest entries or top lists, time spent viewing stories, or ratio of stories viewed to stories voted on. The question of where to draw the line between troll and legitimate but single-minded voter is problematic, though.

But is a person who only votes to show disapproval automatically a troll? If a story contains content a reader finds objectionable, whether that's something fetishy like scat or religiously fraught, are they trolling if they vote a 1? I think most of us would say yes if the story title, blurb, tags, or whatnot make it clear what the reader is in for. Anyone who complains for getting exactly what they should have expected is probably not acting in good faith. I'm not confident that someone voting a 1 because they came across a squick they weren't expecting is voting in bad faith or illegitimately, even if they stop reading at that point and skip to the end to register their distaste. (And, for that matter, how legitimate does that make a vote of 5 on a story simply because it checks a fetish box, or the author is a friend?)

In any case, though, it's difficult (at best) for an algorithm to detect or flag voting patterns like that, much less the reasoning (if any) behind the pattern, so the site might not even try. And if they do, I imagine the false positive rates might mimic the angst of the AI detection situation. And what to do with a voter whose pattern is 50% 1s and 50% 5s? Bot, or someone who only votes like/dislike? At what ratio does that pattern seem inhuman? For that matter, how much of the porn we find here should a human love?

Tl;dr I am a robot, beep boop beep whirr clunk.

My view is that as long as a reader actually read the whole story and as long as they truly believed their score reflected how they felt, their score is legitimate. They aren't obligated to look away because the subject isn't to their liking. I personally do not downvote stories just because I dislike the subject matter, but many readers DO vote that way, and there's no objective way to say they are wrong. Their vote conveys useful information to other like-minded readers, of whom I think there are many.

 

[Bramblethorn]

Especially with a long (and brilliant) story like that, there's no way they were legitimate votes. If a reader didn't like it they would just stop before the end.

And if Melissa's "15 minutes" was literal, no way they actually read the story in that time.

I don't think readers are automatically obliged to read the whole thing before they vote; if page 1 is sufficiently bad it might be reasonable for the reader to skip ahead and downvote it without slogging through the whole thing. But there's nothing I can recall in Queen of the Roller Derby that'd be remotely likely to attract that kind of reaction.

 

[intim8]

It happens. I had a top-ten all time in Romance, and even with over 800 votes, it suddenly dropped 0.02, enough to push it down below a lot of other stories. Probably malicious voting, but with a really high rating, it doesn't take a lot of less-than-fives to take it down a notch. Could be that more people found because it was up there on the list list, and some genuinely didn't like it.

Meh. People like it. The comments are great, and I'm very happy with it, and its votes.

 

[MelissaBaby]

And if Melissa's "15 minutes" was literal, no way they actually read the story in that time.

I don't think readers are automatically obliged to read the whole thing before they vote; if page 1 is sufficiently bad it might be reasonable for the reader to skip ahead and downvote it without slogging through the whole thing. But there's nothing I can recall in Queen of the Roller Derby that'd be remotely likely to attract that kind of reaction.

It's literal. I've had it happen a few times, but not so dramatically.

I did have one reader that was upset because I placed a novel about lesbians in Novels and not Lesbian. They may have bombed it.

 

[ThanksForAllTheFish]

After a year and a half, my story Victor or Vanquished of the Vees made got it's 100th review, and made it onto the all-time Exhibitionist & Voyeur toplist. It's not for me to say whether this was deserved. Within a day it had four new reviews and dropped to 4.79, just low enough to drop it off the toplist. Is this common? Are there people so desperate to push their own work that they down-vote the work of others? Sigh. I don't suppose there's much to do about it. Site administrators are no doubt way overworked as is.

Exact want thing happened to me with multiple stories one was 4.91 and had been anywhere between first to third for four years, now it's under 4.80 for the first time ever

 

[Plathfan]

Exact want thing happened to me with multiple stories one was 4.91 and had been anywhere between first to third for four years, now it's under 4.80 for the first time ever

It’s unlikely that a single factor is targeting all categories and all top stories across the board.

 

[RejectReality]

It wouldn't be the first time someone(s) bombed every toplist continuously for an extended period of time. It took a while to correct the last couple times that happened.

 

[Bramblethorn]

It's literal. I've had it happen a few times, but not so dramatically.

I did have one reader that was upset because I placed a novel about lesbians in Novels and not Lesbian. They may have bombed it.

Ah, those people. I heard from one of them for putting a lesbian romance in Romance. (Which, TBH, was one of the reasons I put it there in the first place though not the main reason.) I could see that resulting in a one-star vote once in a while, but five individual homophobes in a row sounds less likely than one homophobe voting five times.

 

[ThanksForAllTheFish]

It’s unlikely that a single factor is targeting all categories and all top stories across the board.

All 7 stories dropping out the top list in one day after being there for two years? Crazy coincidence if not imo

 

[MelissaBaby]

Ah, those people. I heard from one of them for putting a lesbian romance in Romance. (Which, TBH, was one of the reasons I put it there in the first place though not the main reason.) I could see that resulting in a one-star vote once in a while, but five individual homophobes in a row sounds less likely than one homophobe voting five times.

I suspect it was more likely "how dare you knock my favorites down the list" fan behavior.

 

[pink_silk_glove]

It's literal. I've had it happen a few times, but not so dramatically.

One benefit I have of never ever being close to a toplist and never will.