Solving the dreaded one-bomb problem

Altissimus

Irreverently Piquant
Joined
Oct 25, 2007
Posts
782
ETA: ignore this thread. I didn't realise it had been done to death, coz nothing came up in "recent threads like this". Didn't mean to rake up old leaves, just wanted to ask the hive
mind for suggestions.


So, what ideas do we have that might be a workable solution to the one-bomb problem?

I know Laurel and Manu get numerous requests to readjust scores on stories - because I'm about to send them another one, for starters!

But I was thinking - what would happen if scoring was moved to a median average instead of a mean average?

Wouldn't that solve the issue?
 
Last edited:
This has been debated into infinity, and lately even. I suggest you check out some earlier threads as I doubt many people will have the energy to talk about this over and over again, especially considering that every discussion soon becomes a dead end. To sum it up for you, there is 0 (zero) consensus about what could be done to improve the fairness of ratings, even if Laurel and Manu would actually consider improving it.

Edit: Btw, what do you mean median? Using a median value instead of mean value would result in every story having an integer score, as we can only vote in integers
 
Last edited:
So, what ideas do we have that might be a workable solution to the one-bomb problem?

I know Laurel and Manu get numerous requests to readjust scores on stories - because I'm about to send them another one, for starters!

But I was thinking - what would happen if scoring was moved to a median average instead of a mean average?

Wouldn't that solve the issue?
They could throw out the top and bottom scores and score on a curve.
 
So, what ideas do we have that might be a workable solution to the one-bomb problem?

I know Laurel and Manu get numerous requests to readjust scores on stories - because I'm about to send them another one, for starters!

But I was thinking - what would happen if scoring was moved to a median average instead of a mean average?

Wouldn't that solve the issue?
Some years back the Web site introduced the sweep system to help with this--sweeping wasn't initially done by the Web site. Since scores invariably go up after a sweep, they are helping with the "one-bomb problem." This is about all we get, and it's more than most Web sites provide.
 
What exactly is the "dreaded one-bomb problem?"

Laurel once pointed out to me that a 1* vote isn't an invalid vote. It could be a reader's opinion, just as valid as any other, and readers' opinions are all the rating system is about.
 
Gee.

Why has it taken this long for someone to post a thread suggesting fixes to the voting system?

I‘m amazed there aren’t more threads like this! Show us all the way, OP!
 
But I was thinking - what would happen if scoring was moved to a median average instead of a mean average?

Wouldn't that solve the issue?

Yes, using the median instead of the mean would negate the impact of outlying scores. But of course there would be no more 4.6 or 4.5 rated stories. Everything would be rated a whole number … 3, 4, 5, whatever.
 
ETA: ignore this thread. I didn't realise it had been done to death, coz nothing came up in "recent threads like this". Didn't mean to rake up old leaves, just wanted to ask the hive
mind for suggestions.


So, what ideas do we have that might be a workable solution to the one-bomb problem?

I know Laurel and Manu get numerous requests to readjust scores on stories - because I'm about to send them another one, for starters!

But I was thinking - what would happen if scoring was moved to a median average instead of a mean average?

Wouldn't that solve the issue?

“The problem is not the problem. The problem is your attitude about the problem.”​

― Captain Jack Sparrow


Merely adjust your own attitude, and there is no more problem. I know I have at least two dedicated followers who 1-bomb my stories. I depend on them, and they never seem to let me down!
 
Yes, using the median instead of the mean would negate the impact of outlying scores. But of course there would be no more 4.6 or 4.5 rated stories. Everything would be rated a whole number … 3, 4, 5, whatever.
When the number of votes is even, the median is the average of the two middle values. So it can be a fractional value.
 
Implications of using the median score:
  • everything currently over 4.5 would become 5.0, which basically kills toplists (not sure this would be such a bad thing but some would disagree!)
  • most of the time, when somebody votes, your score doesn't change at all, but once in a while a single vote will cause a shift of as much as 2.0 to the score.
    • e.g. if a story in LW has 499 1s and 500 5s, for a median of 5.0, the next 1* vote drops that to 3.0 and the one after that drops it to 1.0.
  • more confusion among both readers and authors about how scores work
  • drama when scores suddenly shift
  • threads here suggesting we shift to basing scores on an average
Medians can work well for data on a continuous distribution but they're generally not great when there are only a handful of possible values, as here.
 
What exactly is the "dreaded one-bomb problem?"

Laurel once pointed out to me that a 1* vote isn't an invalid vote. It could be a reader's opinion, just as valid as any other, and readers' opinions are all the rating system is about.
Unless its in a contest and the winner scores 4.96 because apparently all those one votes were invalid.

The issue isn't simply the one bomb, its the sweeps which seem to be as inconsistent as everything else here.
 
Our scores are already an average of all votes. A one doesn't give your story a one, it just pulls down the average.

The issue here is some people can't seem to grasp the fact that this site-even with the spite bombing-on average is a soft voting site. The options are 1-5 the majority of stories on this site are over 4 so....is this really that big of a problem outside of LW? If you take LW out of the equation the average score site wide goes up higher.

Maybe people should just come out and say "What I really want is every story to be a 5 whether it sucks or not, and while I'm getting my five, can I get a gold star too? I mean, I got up and everything this morning.
 
Instead of numbers, readers could choose from among an array of colors, choosing their favorite color as a way of indicating the story is their favorite. The "score" would be a color swath that averaged all the different chosen colors.

I'm confident that this would solve the problem and I can't see any downside. And it would be interesting to see if the colors assigned were different for different categories.

Another method: swipe left/swipe right, like the dating sites.
 
Another method: swipe left/swipe right, like the dating sites
An up/down vote like this might be better. Up equals one, down equals zero, then take an average. Someone wants to downvote, fine. But they don't get to do a downvote with averaging strength equivalent to 5 upvotes. If all stories were scored in the full range of 1 to 5 stars and an average story was a 3, the 5 star system would be fine, but it seems like here it's a base of 5* votes with 4s thrown in to create averages like 4.56 or whatever, and then a single 1* vote can skew it down to 4.32 or something. To be clear, scores are whatever, who cares. But with most voters using 4 and 5 stars, the 1 bombs carry undue weight, equivalent to 4 or 5 normal votes of 4 or 5 stars. In this situation a simple 1 or 0 gives every voter equal voice in determining the average between two choices.
 
Last edited:
...I'm confident that this would solve the problem ...
As I indicated earlier, the only "problem" is the self-inflicted one of unrealistic expectations and ego.

People should try to understand that not everyone LOVES their story.

I might write a good, descriptive, realistic and true story about my wife gagging on my cock as she tries to learn deepthroating. And there are authors/readers here who would HATE it. I wouldn't obsess over their 1-bombs, because I'm still enjoying the feeling in the aftermath of my wife's attempts. In other words, the 1-bombs have zero impact on my enjoyment.

And the 1-star ratings carry no more weight than an unearned 5-star (which many authors give themselves.) Those authors who write a shitty story and think others who don't give them a 5 should not rate it at all are just delusional.
 
Instead of numbers, readers could choose from among an array of colors, choosing their favorite color as a way of indicating the story is their favorite. The "score" would be a color swath that averaged all the different chosen colors.

I'm confident that this would solve the problem and I can't see any downside. And it would be interesting to see if the colors assigned were different for different categories.

Another method: swipe left/swipe right, like the dating sites.

Or how about this. . . The readers vote by selecting from a set of five symbols. Say, $ % & * ], for instance. Then the total score is the supersposition of all the votes. After a few votes, the score for every story would be a dark blob.
 
And the 1-star ratings carry no more weight than an unearned 5-star (which many authors give themselves.) Those authors who write a shitty story and think others who don't give them a 5 should not rate it at all are just delusional.

But the primary issue is that, because the system highlights 4.5 as the "read me" point for many story shoppers, a 1 star does carry significantly more weight than a 5 star rating. It takes eight 5-stars to move one 1-star to the 4.5 "read me" point here. That is the real issue in play here. The stars applied do not have equal weight in this system. Those willing to use the 1-star have considerable control over how a story rates and whether it's read here.

That said, it's frustrating to see all the systems change suggestions given here. The Web site isn't listening to users. The system isn't going to change. Those discussing it are spitting into the wind--continuously.
 
Last edited:
But the primary issue is that, because the system highlights 4.5 as the "read me" point for many story shoppers, a 1 star does carry significantly more weight than a 5 star rating. It takes eight 5-stars to move one 1-star to the 4.5 "read me" point here. That is the real issue in play here. The stars applied do not have equal weight in this system. Those willing to use the 1-star have considerable control over how a story rates and whether it's read here.

That said, it's frustrating to see all the systems change suggestions given here. The Web site isn't listening to users. The system isn't going to change. Those discussing it are spitting into the wind--continuously.

I'm reminded of the scene in The Big Lebowski where the Dude and his friends stand on a cliff over the ocean and try to toss their dead friend's ashes into the ocean . . . and the wind blows their ashes back into their faces.
 
I'm reminded of the scene in The Big Lebowski where the Dude and his friends stand on a cliff over the ocean and try to toss their dead friend's ashes into the ocean . . . and the wind blows their ashes back into their faces.
Should have done what Keith Richards did with his dad's ashes - put them in a firework and lit the fuse.
 
Or how about this. . . The readers vote by selecting from a set of five symbols. Say, $ % & * ], for instance. Then the total score is the supersposition of all the votes. After a few votes, the score for every story would be a dark blob.

And when you've collected all the symbols, you know you've written a $*&%ing good story.
 
Or how about this. . . The readers vote by selecting from a set of five symbols. Say, $ % & * ], for instance. Then the total score is the supersposition of all the votes. After a few votes, the score for every story would be a dark blob.

The result might be more attractive than the puke-brown that likely would result from my color-merging idea.
 
But the primary issue is that, because the system highlights 4.5 as the "read me" point for many story shoppers, a 1 star does carry significantly more weight than a 5 star rating. It takes eight 5-stars to move one 1-star to the 4.5 "read me" point here. That is the real issue in play here. The stars applied do not have equal weight in this system. Those willing to use the 1-star have considerable control over how a story rates and whether it's read here.

That said, it's frustrating to see all the systems change suggestions given here. The Web site isn't listening to users. The system isn't going to change. Those discussing it are spitting into the wind--continuously.
Then it seems the simpler solution is to get rid of that Red-H at 4.5, rather than waving it at the authors as if it really means something. Allow the readers to make their own decisions based only on the average.
 
Back
Top