Seemingly widespread ratings manipulation

[CleverGenericName]

I publish most of my stories in the Romance category, and on a whim, I just went to look at the all-time Top Rankings page, and literally every story (with the exception of the top 12) is currently ranked 4.84. This seems statistically improbable. So I looked at my own story page, and 14 of my 24 stories now share the same rating, which also seems fairly improbable.

I'm just wondering if anyone else has run into this in any of the other categories?

 

[TheLobster]

There might well be some manipulation of the top lists, as many frequent fliers there would readily tell you here on AH.

But the fact that many stories on those lists share the same rating is just math.

 

[NotWise]

I publish most of my stories in the Romance category, and on a whim, I just went to look at the all-time Top Rankings page, and literally every story (with the exception of the top 12) is currently ranked 4.84. This seems statistically improbable. So I looked at my own story page, and 14 of my 24 stories now share the same rating, which also seems fairly improbable.

I'm just wondering if anyone else has run into this in any of the other categories?

That's the result of an intentional attack on the top list that started about a year and a half ago. That also happened in LS and may have happened elsewhere. I don't know if it's still going on.

 

[redgarters]

That's the result of an intentional attack on the top list that started about a year and a half ago. That also happened in LS and may have happened elsewhere. I don't know if it's still going on.

It's still going on and yes, it's all the All Time top lists. Once stories reach 100 votes they get pushed down below 4.85. It's sitewide.

 

[EmilyMiller]

That's the result of an intentional attack on the top list that started about a year and a half ago. That also happened in LS and may have happened elsewhere. I don't know if it's still going on.

It’s far from historical. It happens now when anything breaks 4.85 and 100 votes in Lesbian - and elsewhere.

 

[EmilyMiller]

This seems statistically improbable.

OMG soemone who understands stats. Can we be friends?

 

[Tio_Narratore]

OMG soemone who understands stats. Can we be friends?

Stats are just ways of quantifying what you see. I've designed and taught courses for non-math students to use and understand stats; if you have any questions, feel free to ask.

 

[EmilyMiller]

Stats are just ways of quantifying what you see. I've designed and taught courses for non-math students to use and understand stats; if you have any questions, feel free to ask.

I don’t have any questions. I have a sound knowledge of statistics and a little probability theory.

 

[MelissaBaby]

That's the result of an intentional attack on the top list that started about a year and a half ago. That also happened in LS and may have happened elsewhere. I don't know if it's still going on.

I've got a story in Novels, currently sitting at 4.90 with 92 votes. At some point it will hit 100 votes, probably with the score in the same range.

It will be at 4.84 within a day or two.

I had about a dozen stories with scores above 4.85. Some had been in that high range for years. They all dropped to 4.84.

Nobody is going to convince me that that is a natural statistical outcome, or whatever, especially when other writers are reporting the same thing.

 

[Voboy]

Hmmmm... it's almost like there's not already a thread about this sort of thing...

Welcome, OP!

 

[EmilyMiller]

I've got a story in Novels, currently sitting at 4.90 with 92 votes. At some point it will hit 100 votes, probably with the score in the same range.

It will be at 4.84 within a day or two.

I had about a dozen stories with scores above 4.85. Some had been in that high range for years. They all dropped to 4.84.

Nobody is going to convince me that that is a natural statistical outcome, or whatever, especially when other writers are reporting the same thing.

Yeah - the only people who think it’s natural / random are like this:

 

[EmilyMiller]

Dunning & Kruger strike again!

Anyone who thinks that like 96 of the all time top lesbian stories sharing the same rating is a natural occurrence, one resulting from a stochastic process, really needs to read a basic text book.

And… very tellingly, it didn’t use to be that way.

 

[AwkwardlySet]

There's no need to invoke math. I suspect that few, if any, among the AH-ers have the expertise to tackle this issue in a truly scientific, statistical way.

But regardless of our individual math skills, everyone who has been here for more than a year knows that the scores in the top lists used to be different, in value and spread, for a very, very long time. One doesn't need anything beyond common sense to realize that the current top lists have been engineered.

By whom? I'd say that part is also obvious.

 

[FrancesScott]

I publish most of my stories in the Romance category, and on a whim, I just went to look at the all-time Top Rankings page, and literally every story (with the exception of the top 12) is currently ranked 4.84. This seems statistically improbable. So I looked at my own story page, and 14 of my 24 stories now share the same rating, which also seems fairly improbable.

I'm just wondering if anyone else has run into this in any of the other categories?

Every time I have had a story hit page one of the LS all time list, it’s immediately been bombed down to 4.84 and often much lower. One of my stories has entered the top ten on multiple occasions only to suffer the same fate. It’s happened to many other people, including @onehitwanda @THBGato @SugarStorm and @redgarters

It’s not a hallucination, it’s all too real.

 

[CleverGenericName]

Sorry to hear that this is a site-wide issue. I was hoping it was confined to Romance. I went back to look at the 12 stories in Romance that somehow managed to avoid getting pushed down to 4.84:

• Five of them had been deleted from the site by the original author, so their ratings are forever set in stone.
• Two of the remaining twelve are all-time classics by Dreamcloud, with tens of thousands of votes, which I assume are harder to manipulate because of the sheer number of votes required to move the rating.
• My all-time highest-rated story is one of the remaining five, and it is still above 4.84, I assume because it has 5000+ votes, but it is also sinking at a steady rate. It's kind of a bummer since it spent a couple of years at 4.91.
  
  And apologies if I missed an earlier thread on this same subject. I did try to search it up, but I must have used the wrong keywords.

 

[EmilyMiller]

I suspect that few, if any, among the AH-ers have the expertise to tackle this issue in a truly scientific, statistical way.

Your suspicion is not very scientific… Some of us have used statistics in both scientific academia and in highly numerate parts of industry.

I simply invoke statistics to counter claims that math explains the 4.84 curation. It’s statistics 101 that it does nothing of the sort.

 

[Voboy]

Your suspicion is not very scientific… Some of us have used statistics in both scientific academia and in highly numerate parts of industry.

I simply invoke statistics to counter claims that math explains the 4.84 curation. It’s statistics 101 that it does nothing of the sort.

Stochastics are sexy.

 

[GuiltyCowboy]

By whom? I'd say that part is also obvious.

Manipulation seems clear enough but, without having much of a clue about any of this, I’m baffled why anyone would care enough to do this?

Wouldn’t it take quite a lot of work to arrange? Without any kind of financial reward?

Or are the people who are being maintained at the top of these lists making money off their writing elsewhere and being featured prominently here helps with that?

Or am I just underestimating the human ego?

 

[NotWise]

And apologies if I missed an earlier thread on this same subject. I did try to search it up, but I must have used the wrong keywords.

No need to apologize. You gave us an opportunity to bitch about it once again. I guess that never gets old.

 

[lc69hunter]

It's still going on and yes, it's all the All Time top lists. Once stories reach 100 votes they get pushed down below 4.85. It's sitewide.

My question is "why".

 

[WittyUserName]

I publish most of my stories in the Romance category, and on a whim, I just went to look at the all-time Top Rankings page, and literally every story (with the exception of the top 12) is currently ranked 4.84. This seems statistically improbable. So I looked at my own story page, and 14 of my 24 stories now share the same rating, which also seems fairly improbable.

I'm just wondering if anyone else has run into this in any of the other categories?

I like your username.

 

[Actingup]

The pattern is perfectly statistically sound, so long as you assume the presence of a group of voters who ruthlessly attack any stories with scores that are ‘too high’. Statistically speaking, they’re arseholes.

 

[Voboy]

The pattern is perfectly statistically sound, so long as you assume the presence of a group of voters who ruthlessly attack any stories with scores that are ‘too high’. Statistically speaking, they’re arseholes.

I prefer the term "stochassholes."

 

[EmilyMiller]

The pattern is perfectly statistically sound, so long as you assume the presence of a group of voters who ruthlessly attack any stories with scores that are ‘too high’. Statistically speaking, they’re arseholes.

I think group is a fiction we tell ourselves. I suspect it’s not very hard to get ChatGTP to write some serviceable code to do what happens. To me it’s too immediate and too widespread and too consistent in its objectives to be an amorphous group. Either a highly organized group, or - much more likely IMO - a handful of individuals.

Why do they do it? Jealousy is one obvious motivation. Beyond that, they don’t make any money doing this, so the right question is cui bono? Any story that doesn’t get hammered back into the pack benefits clearly.

 

[SkyBubble]

Why 4.84 specifically?