Yes, the Toplists are Broken

[ShelbyDawn57]

For a lot of folks it is obviously the rating. But if you suggest that, they will spit acidic fire at you.

Well, at least I didn't suggest they change their style...

 

[RocketGrunt]

I'm glad that you, personally, haven't seen much of an impact to your story from the flattening of the top list. I haven't either. But that doesn't mean it isn't intentional and that it isn't affecting others.

You should absolutely take his statement of not seeing much impact from the flattening with a grain of salt. He has a story that has been sitting at a respectable 4.84 for a decade and has thousands of votes. Very good performance, but not something that would ever be in shouting distance of an all-time top average rating list if it were not for coordinated troll activity. But since everything above his story has been hammered until it is his level or below, his story is now in the top twenty on the Novels and Novellas board, and its position is protected by continued troll activity that prevents stories from rising above his. The toplists are zero sum, for every story brought down, another story is brought up, and he has a story that was brought all the way up to #14 by the simple expedient of hundreds of stories being vandalized until his was in the top twenty by default.

So when he says that he hasn't been affected by the list curation trolls, that is simply not true. He has been affected. He is a beneficiary. Why he elected to not share that fact in this discussion is not something that makes sense to speculate on.

 

[Five_Inch_Heels]

Maybe ....

No story can be on the list for more than 90 days regardless of score?

90 .... or ... ?

 

[StillStunned]

The top lists just are what they are, a very ambiguous list of meaningless numbers that a lot of us take way too seriously.
Sure, I watch my numbers, too, but I don't lose sleep over some rando bombing my stories, I expect it.

Why do you write? What's your goal, your takeaway? Is it the rating, or is it the telling of the story?

New game: try to get your story at no. 69 in the Top List. That's the only spot that counts.

 

[EmilyMiller]

I don't know if that is because there are more sophisticated trolls or something in the sweeps that allows them, intentionally or not.

The trolls have evolved, the site hasn’t.

 

[SimonDoom]

I largely agree with you, although I have trouble living up to this. But the views (and correspondingly largely, the comments, are boosted significantly by being on the first page of the top list. That is a big incentive for many, if not most, writers here.

Scores, and position on toplists, can affect views, but it's also possible to get views and build a reader base without being at the tops of lists. People fret too much about scores, both as an "accurate" measure of story quality (which it isn't), and for their impact on your "success." There are many ways to succeed here. I recommend that authors focus on what they can control -- their own writing, little marketing strategies that can get your stories more eyeballs, that sort of thing. It's a choice, and it makes for a much happier Lit experience.

 

[RocketGrunt]

Maybe ....

No story can be on the list for more than 90 days regardless of score?

90 .... or ... ?

The view they attempted to get working in November has three taps: Popular (Category) Stories: 7 Days, 30 Days, All-Time. That update is a failure, because the tabs don't update properly and the "more stories" link doesn't go anywhere, but it's a good idea and it's where the board is supposed to be going according to the people running it.

It was absolutely intentional to divide the toplist into a non-rotating all-time format and two lists with a recency bias. And I support that. I think the trending stories shouldn't have to compete against 20 year old stories, and I think the all-time list shouldn't be constantly upended by some Chapter 46 that no one has downvoted yet because it has only been up for three days.

Average rating doesn't work unless we can guaranty that the troll votes will all be swept away, and we obviously cannot do that. But having a tab for long term popular and a tab for recently popular is a great idea.

 

[pink_silk_glove]

The toplists are zero sum, for every story brought down, another story is brought up, and he has a story that was brought all the way up to #14 by the simple expedient of hundreds of stories being vandalized until his was in the top twenty by default.

That's not what he claimed at all.

The rating on that story has never varied by more than .03 and has always been in the top 25 since it was published in 2014.

If he has been in the top 25 for over a decade and the score hasn't varied, then he hasn't been bumped up as you claim.

Or have you been diligently following his score since 2014 to claim that his statement is incorrect? I certainly hope for you mental health that you have better things to do than that.

Hey, I would fully expect that many stories will benefit from the downfall of others through trolling, but at this point it smacks of you just making up any contradictory strike on the fly against anyone who opposes your stance in any way.

 

[ShelbyDawn57]

The trolls have evolved, the site hasn’t.

Actually, I think the site is partially to blame, that some kind of normalization logic is in play. I don't have anything concrete, but a few years ago the entire Transgender top list got nerfed.
Where there had been a dozen stories over 4.9(my one blue W sat at 4.93 for eighteen months), and a variety of small collections at key rating points on down the list, the top 129 are now all at 4.84 or 4,83 with the entire top 250 being between 4.84 and 4.81. It's hard to believe that's just trolls...

 

[Comshaw]

But it doesn't affect every story equally. If you accept the premise that the only way for 244 stories on a top list to have an identical score of 4.84 is through vote manipulation, then those stories have been treated unequally.

It takes a lot of vote manipulation to drop a hypothetical 4.92 story with 4,000 ratings down to the 4.84 threshold. It takes much less vote manipulation to drop a 4.85 story with 400 ratings to the 4.84 threshold. And once that 4.85 story is dropped to the threshold, it may fall off the top list altogether because there are so many other 4.84 stories with more votes.

I'm glad that you, personally, haven't seen much of an impact to your story from the flattening of the top list. I haven't either. But that doesn't mean it isn't intentional and that it isn't affecting others.

I can see where a person with a story on one of those lists would get all butt-hurt if it got pushed off that list for any reason. That's human nature. Hell, I'd feel the same way. However, another part of human nature is when a person is caught in such a situation, they rarely have the clarity of vision, or lack of bias to assess it fairly.

The bolded part of your post is the problem with most of this thread. There has been lots of speculation, insinuations and accusations that this that or the other is the reason for the problem. The only fact we have is this: something has changed. Everything else stated as the cause of the problem is subjective guesswork and nothing more. Everything. I've seen nothing even close to a concrete fact quote.

For my dollar I'd guess since it appeared to happen all at one time, it has more to do with a change in the website software rather than a huge influx of program savy trolls. It may have been a FUBAR inclusion on the site management's part, or a glitch in a software update, or even intentional by them. I can only guess. I believe (please take note of that word because it denotes what I think it is rather than insisting it is) most trolls aren't intelligent enough to do something that sophisticated. Unless they are twisted evil geniuses. If they are, we can all kiss our numbers and sanity goodbye. Ya' can't fight Gru ya' know.


Comshaw

 

[BobbyBrandt]

You should absolutely take his statement of not seeing much impact from the flattening with a grain of salt. He has a story that has been sitting at a respectable 4.84 for a decade and has thousands of votes. Very good performance, but not something that would ever be in shouting distance of an all-time top average rating list if it were not for coordinated troll activity. But since everything above his story has been hammered until it is his level or below, his story is now in the top twenty on the Novels and Novellas board, and its position is protected by continued troll activity that prevents stories from rising above his. The toplists are zero sum, for every story brought down, another story is brought up, and he has a story that was brought all the way up to #14 by the simple expedient of hundreds of stories being vandalized until his was in the top twenty by default.

So when he says that he hasn't been affected by the list curation trolls, that is simply not true. He has been affected. He is a beneficiary. Why he elected to not share that fact in this discussion is not something that makes sense to speculate on.

The story rating change doesn't affect me. The influence that change had on the position in a top list doesn't affect me because I simply don't care.

I entered this conversation to counter the unfounded (IMO) claims that the top lists are being systemically manipulated. I don't see that and no one has provided proof.

What has happened is that many scores overall have been altered, and that has influenced the position of some stories in the top lists. I have had stories nowhere near a top list see their scores change slightly in the recent past, so why should those bother me more or less than another?

Cause and effect are what you are missing, or choosing to ignore.

 

[Comshaw]

New game: try to get your story at no. 69 in the Top List. That's the only spot that counts.

Oh I don't know about that. I kinda' like 59 and 99 too.

Comshaw

 

[crookedletter]

Scores, and position on toplists, can affect views, but it's also possible to get views and build a reader base without being at the tops of lists. People fret too much about scores, both as an "accurate" measure of story quality (which it isn't), and for their impact on your "success." There are many ways to succeed here. I recommend that authors focus on what they can control -- their own writing, little marketing strategies that can get your stories more eyeballs, that sort of thing. It's a choice, and it makes for a much happier Lit experience.

Anecdotally, the story I have that is kind of riding up and down the top list for its category -- more visible in 12 months than in all time, though it's somewhere buried in that one too -- seemed to get a larger surge of views/engagement when I published a new story in the same category than it did when it, for example, moved into page 1 of the 12 month view.

 

[old_prof]

Actually, I think the site is partially to blame, that some kind of normalization logic is in play. I don't have anything concrete, but a few years ago the entire Transgender top list got nerfed.
Where there had been a dozen stories over 4.9(my one blue W sat at 4.93 for eighteen months), and a variety of small collections at key rating points on down the list, the top 129 are now all at 4.84 or 4,83 with the entire top 250 being between 4.84 and 4.81. It's hard to believe that's just trolls...

I do believe that the most likely scenario is that someone became determined enough to create a tool to bomb at will. I don't know whether it's all one troll, or that troll shared the weapon, or the weapon was invented multiple times. I have no doubt that dozens of people on this forum could in fact do so. And that there are people who who deploy it.

I can see an argument for the site intentionally wanting to do this (shuffling the top lists regularly gets more stories added visibility, which improves the quantity of "top-shelf" stories). But I don't see Laurel as being someone who would buy off on that action. Technically, it would be trivial for them to do. But I'm also dubious that Manu ha the time to bother or to be brutally honest, that he wouldn't noticeably screw it up, which I have not heard any one reporting.

 

[Five_Inch_Heels]

The view they attempted to get working in November has three taps: Popular (Category) Stories: 7 Days, 30 Days, All-Time. That update is a failure, because the tabs don't update properly and the "more stories" link doesn't go anywhere, but it's a good idea and it's where the board is supposed to be going according to the people running it.

It was absolutely intentional to divide the toplist into a non-rotating all-time format and two lists with a recency bias. And I support that. I think the trending stories shouldn't have to compete against 20 year old stories, and I think the all-time list shouldn't be constantly upended by some Chapter 46 that no one has downvoted yet because it has only been up for three days.

Average rating doesn't work unless we can guaranty that the troll votes will all be swept away, and we obviously cannot do that. But having a tab for long term popular and a tab for recently popular is a great idea.

Just eliminate 'All Time'. Or change it to the last 5 years maybe.

If not for those kinds of lists, would anybody even find stories from 2002?

At the very least, disable voting for any story over 5 years old.

 

[old_prof]

Just eliminate 'All Time'. Or change it to the last 5 years maybe.

If not for those kinds of lists, would anybody even find stories from 2002?

At the very least, disable voting for any story over 5 years old.

Many readers like the ability to vote, it makes them feel more connected to the site. (Although fewer readers do than most of us would like.)
So I do not see the site locking off voting for people.

I could see building five year or decadal lists to show more stories. Maybe the older lists would be less likely to be bombed? I can see the site liking that, but not enough to actually make the change.

I could see creating an unranked HOF list (just alphabetic or chronological) that was every story with 500 or more votes, say, that had a 4.75 or 4.8 rating. I think that would be a less attractive target for bombing. And I would have tried to read through the whole list for categories I tended to read when I was purely a reader here.

 

[EmilyMiller]

I do believe that the most likely scenario is that someone became determined enough to create a tool to bomb at will. I don't know whether it's all one troll, or that troll

It’s overwhelmingly likely to be tool based and even automated. One of the giveaways is that the sweeps seem to not pick up on it. Which would necessitate a human spending an inordinate amount of time to do what is necessary to be sweep-proof.

It’s also telling that it has got much worse.

 

[EmilyMiller]

What the site could do is:

1. Only count the middle 90% of votes cast, discarding the bottom and top 5%
2. Restrict voting to named accounts
3. Make who has voted public the same way that who has favorited is
4. Do some basic statistical analysis of voting patterns / times / relationship to view patterns - and use this to highlight suspect activity (clusters of low votes, votes cast with no uptick in views, etc.)
5. Create profiles of member voting to highlight rogue activity (e.g. someone who only ever votes 1 - 3, except on work by author X, where it is always 5)
6. Respond to author complaints in timely manner

None of these would eliminate the problem, but they would contain it.

But it’s all moonshine as demonstrably the site is quite comfortable with the status quo.

 

[EmilyMiller]

You could also have trust profiles for people voting, high trust and your full vote counts. Low trust and only 10% of your vote counts.

 

[THBGato]

Why do you write? What's your goal, your takeaway? Is it the rating, or is it the telling of the story?

I write because I want to. But I publish here because I want people to read my stories. That's surely true of everyone.

I largely agree with you, although I have trouble living up to this. But the views (and correspondingly largely, the comments, are boosted significantly by being on the first page of the top list. That is a big incentive for many, if not most, writers here.

Exactly. It's about attracting readers, isn't it?

I'll admit, it's also about legacy. I'm never going to win a competition because I don't enter them and the monthly prizes have stopped. So getting on the top list seemed - when I started writing here and the top lists were pretty stable - to be a way to ensure readers would still come across my stories five, ten, fifteen years from now. After all, that's how old some of the stories I was sharing the top twenty with were. That was attractive. We publish to be read, right?

Okay, that's gone now, c'est la vie. But that's why some writers cared about them: for me it wasn't about the score or the status (okay, I didn't hate those!) but about the visibility. That's why I started putting my stories up elsewhere - having gone to the effort to write them, I want others to be able to read them.

 

[EmilyMiller]

I write because I want to. But I publish here because I want people to read my stories. That's surely true of everyone.

Exactly. It's about attracting readers, isn't it?

I'll admit, it's also about legacy. I'm never going to win a competition because I don't enter them and the monthly prizes have stopped. So getting on the top list seemed - when I started writing here and the top lists were pretty stable - to be a way to ensure readers would still come across my stories five, ten, fifteen years from now. After all, that's how old some of the stories I was sharing the top twenty with were. That was attractive. We publish to be read, right?

Okay, that's gone now, c'est la vie. But that's why some writers cared about them: for me it wasn't about the score or the status (okay, I didn't hate those!) but about the visibility. That's why I started putting my stories up elsewhere - having gone to the effort to write them, I want others to be able to read them.

This

 

[BobbyBrandt]

I do believe that the most likely scenario is that someone became determined enough to create a tool to bomb at will.

Anyone who tracks all their scores and doesn't focus only on the top lists would see that the scores overall are what changed, regardless of their position on any top list.

I have had stories with ratings below 4.00 see the same type of change (up and down) as any that I have on a top list. What reason would any "trolls" have to target stories rated lower than 4.00?

 

[BobbyBrandt]

I write because I want to. But I publish here because I want people to read my stories. That's surely true of everyone.

The axiom goes, "Writers want to get published, authors want to get paid."

Reader engagement is the only payment we receive here, but it comes in multiple ways. If you focus too much on only one aspect of reader engagement, you'll miss much of it.

 

[RocketGrunt]

What the site could do is:

1. Only count the middle 90% of votes cast, discarding the bottom and top 5%

Imma gonna stop you there, because that is totally incoherent. There is no top 5%. There CAN'T be a top 5%. The Mode is above the average. More than 80% of votes are the highest possible value. As far as stories in contention for toplists are concerned, 5* is the normal vote and all other ratings are variable strength downvotes. There's no top 5% to discard, the top 5% is indistinguishable from the top 88%.

Which of course, is one of the reasons we shouldn't be using average ratings as our point of comparison. We have hundreds and hundreds of stories that have 12% or less votes that are different from the highest possible vote. The only reason that one gets a 4.86 and one gets a 4.85 is that one of them got a few less 4* and a few more 2* among the people who voted down. The average is simply a measure of downvotes, a measure of how much the people who didn't like it decided to express the fact that they didn't like it.

 

[EmilyMiller]

Imma gonna stop you there, because that is totally incoherent. There is no top 5%. There CAN'T be a top 5%. The Mode is above the average. More than 80% of votes are the highest possible value. As far as stories in contention for toplists are concerned, 5* is the normal vote and all other ratings are variable strength downvotes. There's no top 5% to discard, the top 5% is indistinguishable from the top 88%.

If there are 100 votes and 60 of them are 5️, 30 are 4️, 6 are 3️, 2 are 2️, and 2 are 1️. Drop 5 x 5️, 1 x 3️, 2 x 2️, and 2 x 1️. I’ve got a much longer essay about it on my blog, it makes adjustments for stories with low vote counts.

You may not like it, but calling it incoherent is the sort of thing that people get rightfully pissed about.