Yes, the Toplists are Broken

[Five_Inch_Heels]

These threads make me wish all lists were removed.

 

[Erozetta]

My E&V stories were being bombed off the Top List long before that.

My response was to RG's claim of it being "as expected" until 2025.

That kind of targeted speculation does nothing but risk pointing fingers at who gains what from such actions..

 

[pink_silk_glove]

My response was to RG's claim of it being "as expected" until 2025.

It also makes one wonder who RG's other account is, since RG has only existed since Oct 2024. Pretty difficult to believe that someone would be watching toplists that closely long term without an account. Yes, plenty of people here read for years and years without an account, but to actually follow toplists so closely over an extended period of time (years) as to make these detailed mathematical claims, it just doesn't make sense.

 

[Voboy]

There are, in fact, ways to fix the lists so that the trolls don't win.

Like a lot of people here who enjoy beating dead horses, you don’t seem to have grasped the fundamental truth of Lit: that its voting systems are not designed with writers in mind, but with readers and site management.

I’m reasonably sure you mean well, but until you accept that truth, you’re unlikely to be happy here. At the end of the day, you need to figure out why you’re here. If the answer is “because I want to be on the top lists,” I don’t think you’re setting yourself up for success.

 

[jsmiam]

…The new thing is the mass flattening of the toplist contenders. 243 of 250 stories on a toplist having the same rating is new. And it's utterly insane. …

On this specific topic, scores are rounded to two decimal places then Sorted by number of votes. In the case of the same exact number of votes, I’m not sure. If I were writing the query I would sort by all decimal places next, but I don’t know what lit did.

So in a busy category, and using the example you cited (I haven’t cross checked, I’ll just use your example), there aren’t necessarily 243 votes of 4.84 (or whatever number you meant), there are 243 votes between 4.83500000001 and 4.8499999999 (there are different rounding algorithms, but the basic point remains regardless of that).

This isn’t to say there aren’t shenanigans. A story of mine spent some time on page 1 of the Exhibitionism & voyeur toplist and there were shenanigans aplenty.

Without blaming the author specifically, maybe it was them or maybe a fanboy, there are stories that stay and stories that get knocked down. and Stories with thousands of votes are harder to knock down than stories with less too. And a threshold where the shenanigans stop.

Could a score obsessed fanboy or author have a script too? Or just their obsession? Either is possible.

Conclusion, I’m not sure I agree on flattening being a new issue.

And a new but old topic: the clutter of series/chapter stories. That is an even bigger problem with the top lists. Series clutter up top lists and push standalone stories out the way with their skewed scoring due to the “bought into the series” followers. series are far worse and more prevalent a problem for the top lists meaning anything.

 

[Five_Inch_Heels]

Like a lot of people here who enjoy beating dead horses, you don’t seem to have grasped the fundamental truth of Lit: that its voting systems are not designed with writers in mind, but with readers and site management.

I’m reasonably sure you mean well, but until you accept that truth, you’re unlikely to be happy here.

The real reality is that they really don't mean anything at all.

They're a visual gimmick at best.

I've read, or tried to read some of the 'top' ones and found them not very good in my opinion. They're fan based, not quality or content based.

 

[BobbyBrandt]

This is not what happens anymore.

I've seen a lot of theories and anecdotal claims that there is some "new" trend affecting scores, but never any proof of actual malicious intent shown. Rumors about automated scripts and conspiracy theories won't impact anything on this site.

The fact that ratings go up and down could be attributed to several factors, and most egos naturally gravitate towards "someone is out to get me" instead of the more natural cycle of loyal follower votes becoming interspersed with other readers who may not like the story as much.

I track my stats religiously and can tell as my followers become aware of a new story, and then vote on it. I have had stories remain over a 4.98 rating for a couple of days before the influx of newer readers discover them and start adding their votes, dropping the story to a more accurate rating of 4.6* or similar where the stories tend to remain for years.

This very trend is active right now with my latest story, so it is still happening.

I will accept that certain writers here seem to have occasionally attracted anonymous members who target their stories with lower ratings and negative feedback, but that is no different than me ticking off a literary critic with some online blog or review site. You are never going to please everyone and I just don't see the reason to fuss over it. Play the long game.

 

[Voboy]

These threads make me wish all lists were removed.

A lot of towns in my area are coming to the same conclusion about Honor Roll or High Honor Roll in their high schools. Publicly ranking achievement, like here on Lit, often just tends to lead to bad blood and whining. Usually it’s parents, and often they’re very mean-spirited. Why should the schools bother subjecting themselves to that?

 

[RejectReality]

The reason people get upset is that hitting the upper fifth or so of page 1 of a toplist = massive new reader influx. Those lists are a major point of reader selection no matter what anyone thinks of them. Running afoul of the gamesmanship and getting knocked down out of that spotlight takes that away. It's like you made it to the front of the pack in a race, and the other racers' fans start throwing banana peels in front of you and shooting bb guns at your knees with every step.

It's a perfectly natural and not unwarranted reaction to be bothered by that.

That being said, no ego-stroking, voodoo math is going to solve that problem, and it is not new. I doubt it's automated either. All it takes is a handful of fans competing to generate these results, which anyone who's been here for any significant length of time has seen happen repeatedly.

 

[MelissaBaby]

The large majority of writers, even many very good writers, have never gotten a story on the top lists. I can't help but wonder how all this handwringing over them looks to those writers. My guess is that it looks like rich people complaining that their chauffeur stopped the limousine too far from the curb.

I've had a lot of stories make it to the top lists. In fact, most of my stories have made at least a brief appearance. And I've have a few there for a long time. I have watched my top rated stories slide down over the last year, even ones that were stable around the 4.90 mark for years. Yeah, something seems to have changed, and I don't like it.

But let's keep some perspective here. It will get fixed or it won't. It doesn't matter to all but a few readers, it doesn't matter to most writers. It doesn't even matter that much to some of us whose ratings have declined.

As I said in another thread, Nigel Tufnel's amp isn't louder because it goes up to 11. Your story is just as good at 4.84 as it was at 4.90.

 

[Five_Inch_Heels]

A lot of towns in my area are coming to the same conclusion about Honor Roll or High Honor Roll in their high schools. Publicly ranking achievement, like here on Lit, often just tends to lead to bad blood and whining. Usually it’s parents, and often they’re very mean-spirited. Why should the schools bother subjecting themselves to that?

Why should the schools bother subjecting the students to that?

You hit a mark a few times and people expect you to always hit it, even under vastly different circumstances.


I see it over on Reddit. I've asked for some of my Mod achievements and accolades to be removed or hidden because of what others come to expect. I'm not always in the mood to try to live up to that.

 

[pink_silk_glove]

As I said in another thread, Nigel Tufnel's amp isn't louder because it goes up to 11. Your story is just as good at 4.84 as it was at 4.90.

Lick My Love Pump.

 

[old_prof]

I've had several stories go through the reach 100 and get slapped down, which is very painful.

I understad LC68's point that all the ratings are inflated. Yes they are. That's the way internet ratings go most places. As much as many of us (myself included) need to get over the pain of getting our better stories 1-bombed, LC68 needs to get over the grade inflation.

I honestly believe that none of my stories would ever make the top list if not for previous stories being 1-bombed. So I try to think that it's at least fun to see a story of mine at the top, even if only ever so briefly. It frustrates the analytical side of me to understand that any rating over about 4.75 is meaningless because of the telling of the top lists. I have said this before, but the sweet spot for stories is to be a "true" 4.83 or 4.84, because they stay below the radar and never get the same 1-bombing and will maintain that rating, while stories higher than that will get beat down to below it. I suspect that will slowly creep back up, only to get beat down again, probably settling at the 4.83-4.85 range eventually. I've not been here long enough to see that settling yet.

 

[crookedletter]

Has anyone written a story yet about the Secret Cabal of Lit Trolls?

I assume they meet in shadowy basements, their locations undisclosed until hours before to avoid leaks.

And they chant incantations and do blood rites and read the entrails to determine which objectively brilliant stories they're going to target, and they cackle maniacally over their cigars when they successfully bring them down a couple hundredths of a point.

 

[Erozetta]

A lot of towns in my area are coming to the same conclusion about Honor Roll or High Honor Roll in their high schools. Publicly ranking achievement, like here on Lit, often just tends to lead to bad blood and whining. Usually it’s parents, and often they’re very mean-spirited. Why should the schools bother subjecting themselves to that?

This is why I pushed against award systems in school that were based only on gpa. I got a couple of teachers to listen and they awarded the students who improved the most over the school year vs raw averages.

This cost me awards and I was fucking thrilled with that because the awards rewarded effort which encouraged classmates. When you know only the top numbers are gonna be awarded, you tend to put in the bare minimum unless you think you can actually get that top spot and then it basically becomes a bloodsport for absolutely no reason.

It was the single biggest thing I've ever accomplished and it only made sense to a couple of teachers who are likely retired or dead now. But it made a difference for the time I was in school and I saw a couple of students really thrive with the recognition from that change.

I would love to see a shift from raw GPA awards to actual awards that track and reward increased effort and skill as they learn.

 

[Five_Inch_Heels]

I don't believe any story can/should maintain a rating above 4.8 for any length of time once it gets above X,000 clicks.

I don't believe in top seller lists for printed book eithers. Too easy for mass buyers to inflate those lists.

 

[Five_Inch_Heels]

Has anyone written a story yet about the Secret Cabal of Lit Trolls?

I assume they meet in shadowy basements, their locations undisclosed until hours before to avoid leaks.

And they chant incantations and do blood rites and read the entrails to determine which objectively brilliant stories they're going to target, and they cackle maniacally over their cigars when they successfully bring them down a couple hundredths of a point.

Or raise one.

 

[RocketGrunt]

On this specific topic, scores are rounded to two decimal places then Sorted by number of votes. In the case of the same exact number of votes, I’m not sure. If I were writing the query I would sort by all decimal places next, but I don’t know what lit did.

So in a busy category, and using the example you cited (I haven’t cross checked, I’ll just use your example), there aren’t necessarily 243 votes of 4.84 (or whatever number you meant), there are 243 votes between 4.83500000001 and 4.8499999999 (there are different rounding algorithms, but the basic point remains regardless of that).

Conclusion, I’m not sure I agree on flattening being a new issue.

The flattening definitely is a new issue, and it's one that happens every day. It happens in the middle of the night in America, or possibly as some sort of breakfast activity in Australia. The toplists don't update continuously, there's a several hour period between when it checks for vote totals, meaning that the list is remade only a few times a day. And entropy increases on each of those revisions until the one where it all condenses down again. And it condenses down, not up.

So this time, the middle of the night update gave us 243 stories with one single rating, and 7 with a higher rating. The following morning update showed us that 26 stories had a higher rating, with 19 stories having clicked up by .01 average. Tonight it's going to condense down again as those stories are whack-a-moled down and a few others will rise up to take their place - and get hammered down with the next nighttime trollscript pass.

And a new but old topic: the clutter of series/chapter stories. That is an even bigger problem with the top lists. Series clutter up top lists and push standalone stories out the way with their skewed scoring due to the “bought into the series” followers. series are far worse and more prevalent a problem for the top lists meaning anything.

This is absolutely true, and why despite being a serial writer myself, I advocate for a total upvotes criteria. Absent troll curation, it is natural for series to have many episodes with high average ratings without having hugely high numbers of ratings. Having weekly or monthly popular tabs will give plenty of exposure to those, because being chapter 25 or 55 or whatever will get a rapid influx of ratings as the fans get the thing they were waiting for. They don't need to live perpetually on the All-Time list.

Except the Non-Human board, obviously. That toplist is going to be the personal property of one author's series pretty much no matter what you do, because there's a single series with over a hundred chapters that is wildly more popular than anything else on the board by any possible metric.

 

[Five_Inch_Heels]

I would love to see a shift from raw GPA awards to actual awards that track and reward increased effort and skill as they learn.

And where does that put the poor saps who literally can't learn or improve no matter how hard they try? They work their asses off to try and do their best. They see it isn't getting them anywhere and they just stop. They no longer even try for anything over a C.

'I just want to get this over with, get out of school and move on to something else'.

Some just do better with their hands than they ever could with words.

 

[Erozetta]

And where does that put the poor saps who literally can't learn or improve no matter how hard they try? They work their asses off to try and do their best. They see it isn't getting them anywhere and they just stop. They no longer even try for anything over a C.

'I just want to get this over with, get out of school and move on to something else'.

Some just do better with their hands than they ever could with words.

That's kind of the point. The ones who put in effort got the awards despite their final grades. There were a few C students who got awards over the A students, which absolutely pissed off a bunch of A students, but being rewarded for something that comes fairly naturally isn't an award. It's like being in a debate with yourself and being proud you won. It's pointless.

That D student who got up to a C because they started asking questions and *trying* to learn the thing they weren't getting? That's worth awarding. That's noticeable effort and should be commended.


There's always going to be people who aren't awarded, that's just the nature of awards, but when you have a chance to set the criteria and you see an opportunity to lift effort over final results? That effort is going to compound over time and create better results in the long run.

You can't make someone learn something they don't want to learn. But you can encourage the people who want to improve by rewarding effort.

 

[ThatNewGuy]

The third good reason is that unlike average ratings, which show increased resilience to trolls the older the story is (and therefore disadvantage newer stories), the raw number of 5*'s can be made fair(er) across stories of all ages. The all-time toplist can can still count every single vote of that 20-year-old story, but after scaling the result against this long span of time the story can still be fairly positioned among those published only a few months ago.

I think using raw number of five votes as the basis for the all-time top lists sounds like a promising solution to the recent flattening that has occurred. My two cents:

1. The all-time top list problem is a niche concern that, in my opinion, doesn't warrant that amount of digital ink that has been spilled recently.
2. That doesn't mean that the problem shouldn't be addressed if there is a way to do it.
3. Using raw number of five votes sounds like it would greatly improve things by removing the power of downvoting scripts.
4. My one concern after seeing it proposed in a separate thread a little while ago was that it would unfairly weight the top lists in favor of very old stories that have had the most time to accumulate the highest raw total of five-star votes. @TheLobster addressed this concern in his bolded text above.
5. A caveat: math and statistics are not my area of expertise, so I concede there may be disadvantages to the proposed solution that I'm simply not able to see.

I say all this as someone with stories that have actually benefitted from the recent top list flattening. The scores of my top list stories have been pushed down because of the flattening campaign, just like everyone else.

However, a few of my stories happen to have high scores and a large number of votes, which insulates them to some degree. Those stories benefit when scores are tied across the board because total number of votes is used as the tie breaker on the all-time top lists.

The upshot is that two of my stories have climbed even higher within their respective top lists. One of them even reached the top 10 all time across all categories. The proposed solution would certainly knock it out of that position. But as far as I can see, the proposed solution would also be a fairer and more troll-proof way to list all-time stories.

We're tilting at windmills here because I don't think Lit is going to devote the time or resources to solve an issue that affects a vanishingly small percentage of its population. The site's main concern is giving readers a general sense of what stories are considered popular/successful so that readers can more easily find "good" stories. Flattened top lists don't affect reader satisfaction one iota, so it's not something that I think will be fixed. But that doesn't mean we can't have fun talking about how to fix it!

 

[THBGato]

The systematic trolling that targets every story has been going on since Nov 2024. I have the data to back this up. I reported it back in Dec 2024, but...

It hurt me to start with (I had 4 stories in the top twenty at the time), now I'm more phlegmatic.

However, to those who say it doesn't matter, well you're wrong. There are two ways for you story to be linked to from a category hub for longer than a few days:
1. Win a contest
2. Be at the top of the top list
So, in terms of increasing your stories' visibility, being on the top list matters.

I know as a reader that I found several stories I wouldn't have otherwise by browsing the lesbian top list. Maybe you don't, but other readers do. So if your story gets voted down by the bot/troll, then that hurts your visibility. This is especially true when every story has basically the same score, as the top list ranks stories by number of votes - so older stories will always win out, gain more viewers, more votes, etc. So the trolling harms new stories more.

 

[ronde]

Someone should take the time to graph out the performance of a story of theirs over a one, two or three year period. I'm pretty sure that it would show that over time, the rating of a story finds its level spot and will deviate little from that point.

Loyal followers have had their one vote, trolls have lost interest, hurt feelings get forgotten, sweeps come and go, and life moves on.

I have been doing that for every story in my catalogue and the results are exactly as you state. I see a typically lower rating on the first day that progresses to what the story will level out at over the next week. After that, all I've seen is weekly rating changes between -0.01 to +0.01. Sometimes the same story will gain ratings one money and lose the next. Most of my older stories haven't changed rating in years. I do occasionally see the impact of the "sweeps", but the result is usually no bigger than +0.02.

 

[EmilyMiller]

Yes, the Toplists are Broken​

Yes they are, and ratings in general are a shit show, wide open to manipulation, and with next to nothing done about it by the site.

Also, even if not vandalized by the nefarious, there is little correlation between merit and rating.

But nothing is ever going to be done, so your options are:

1. Put up with it
2. Don’t publish here

I went with option 2, but YMMV.

 

[RocketGrunt]

The systematic trolling that targets every story has been going on since Nov 2024. I have the data to back this up. I reported it back in Dec 2024, but...

I would believe that. The contents of the toplists were still recognizable to me until the end of January 2025, but it makes sense that the systematic bullying started a couple months earlier.