Laurel: A question and a suggestion or two

[Mr_Neb]

Is is possible to keep track of the votes that each story gets in order? This may be helpful in determining if a story is in fact getting slammed by 1s unnecessarily. If a story is posted and get 20-30 votes with all 4s and 5s and then there is a streak of 1s, it is pretty clear that there is something odd going on or else a group of people with a particual dislike for a certain story all just happen to read and vote at the same time.

If that can't be done, another option would be to set it up so that an email is automatically sent with the vote value to the author each time a vote is cast. The email would be anonymous. Authors can elect to turn this option on or off. This seems like it would be relatively easy to do.

My suggestion pertains to the cash awards for stories. If it is clear that the voting system can't honestly assess a story's true popularity, then the prizes are somewhat meaningless. Wouldn't it be better to discontinue them?

Thanks for listening.

Respectfully,

Mr. Neb

 

[Laurel]

Thank you for the feedback. We have the ability to review the voting stats of any story, including IP addresses, frequency of votes, etc.

If you think in terms of a single story, then the possibility of fraud may seem to interfere with the accuracy of the scoring. However, when you look at the big picture involving thousands of votes and thousands of stories, you suddenly see that the system is relatively solid.

For example, if one story reaches #1 and a certain percentage of people 1-vote it because they want their story to rise to the top, then that seems to mess with the system. But if you assume, instead, that EVERY story that reaches the top spot will receive a certain percentage of 1-votes, then you begin to realize that the 1-votes, in reality, have no affect on any individual story. The same is true of authors and/or author's friends who may 5-vote their own stories. In the end, the tiny amount of "fraudulent" voting that may or may not exist for any given story is far outweighed by the sheer volume of honest voting.

Even with that in mind, there is no rule against someone voting a 1 on a story simply because they personally didn't like it, for whatever reason. If you see 100 great reviews for a movie, and you go see it based on those reviews, and you personally dislike the movie, you may be angry because you feel you were misled into seeing the movie based on the positive reviews. In the same way, if a story reaches #1 on the Top List, and a reader is less than impressed with it, then that reader may "punish" the story for having so much praise, which he or she does not believe is justified. This may result in a 1-vote, but it is as valid as any other vote, is it not?

Conversely, if a reader sees that a story he or she really likes has a low rating, they may very well become activist and try to raise that story's ranking by giving it a 5, even though perhaps they honestly think it a 4. Again, that is a valid vote by someone who read the story, as much as any other vote.

There are also other factors that can influence the score of a certain story. We have seen, on numerous occasions, where a story will get linked from another website, a forum, or the Usenet. If there is a site somewhere on the web dedicated to blonde green-eyed girls with 3 legs, and an author on Lit happens to write a story about that subject - and the other site links directly to that story - then there is a good chance that that story will rocket up in the voting. We have had, for example, The Hun link to specific stories. The Hun gets over 1 million visitors to his site per day. That type of link will obviously affect the number of votes and, quite possibly, the quality of votes for that specific story. Once again, this seems to skew the system unless you look at the big picture, where a certain number of stories are being linked to each day, every day, all year long. In the end, these things tend to balance each other out.

At the current time, there is no perfect system. The voting system we use works for IMDB, Amazon, AmIHotOrNot, and millions of other sites on the web. On a micro level, it appears flawed, but on a macro level it tends to be very accurate in terms of analysing people's favorites.

One thing to keep in mind is that no voting system EVER - even if 100% accurate - can determine the best story. These are only the opinions of specific readers about their favorites. My favorites will not be the same as your favorites, and the world wouldn't be as interesting a place if they were.

We definitely appreciate your suggestions and feedback. Please keep them coming!

 

[Mr_Neb]

I have to dissagree on a few points. The number of votes that a story gets, at least in my case, is not large enough to make insignificant a series of 1s intentionally cast to lower a rating. My stories usually fall in the 40-90 range.

I am not referring to thousands of votes cast against thousands of stories over a long period of time. I am referring only to the votes cast for stories that have initially been received very well and then fallen off dramatically. These reflect a much smaller percentage of your story database. Expressed another way, you have a small problem because it seems to only affect your most popular new stories.

I have no objection to a story receiving a 1-vote because the reader dislikes a story, mine or anyone elses. But to use your movie analogy, is a movie more likely to receive a higher precentage of negative reviews later in its lifespan than initially? And is a movie more likely to receive this disporpotionate number of negative reviews only if the movie starts out being popular?

I believe that there is a much higher probability that 1s are intentionally cast to bring down a story that 5s are cast to inflate them. As far as I know, there has never been a post from a surprised author that had a story initially ranking at a 3.5 and then all of a sudden soar to a 4.7

I am at a loss to understand why there is a predictable interest in artificially lowering top rated stories. I see nothing to be gained by it but it clearly exists. Allow me to demonstrate.

My latest story, "What I Want" was to my pleasant surprise at 4.75 after 17 votes. At 20, it was 4.65 and at now at 25, it is at 4.32. I would very much like to know what the votes were and in what order. The same has happened with several other stories. If the voting were truely a reasonable reflection of the popularity of a story as determined by even a slightyl imperfect system, the distrubution of votes would be relatively uniform. The probability of having a reader like a story or dislike it is the same at any one point in time.

An argument can be made to support the idea that there is a higher likelihood that a story will get a higher rating if it is initially popular if the system provided a level paying field. If a reader decides to read a story from the Top Lists, I would suspect that they would be more likely to go along with the existing support for the story rather than go against it. I think it is a bit of a stretch to assume that the opposite is equally true in that a reader, a total stranger to the author, would cast a higher vote as a personal, single vote crusade.

Another argument for the higher likelihood of stories being given higher scores rather than lower ones is that I would suspect that if a reader does not prefer a story, they are more than likely to not vote at all.

I've touched on it already bit there is still the issue of timing. There are many ways to get to the stories on Lit but I would suspect that most are accessed one of two ways. One is through the New Stories list where the story's ratings are not shown. The other is through the Top List that does show the ratings. There have been many instances where new stories have increasing scores in the beginning and then have dramatic fall offs after they have moved up on the Top List.

The opposite is not true however. I have had stories that have received mediocre scores right from the outset. They never moved significantly up on the Top List. These stories never fell in their scores. I would be very interested in knowing the distribution of scores for those stories as compared to the initially popular ones whose scored fell dramatically. I would suspect that there were actually fewer 1s cast for the mediocre stories than were cast for the more popular ones.

In my case, I can assure you that there is no immense fan base that is waiting to boist my scores. I tend to stay on the periphery in terms of most of the forum topics. I have neither intense fans or vindictive enemies so I don't take any of it personally. I am just offering my experiences here in hopes of sheading some light on this puzzling situation.

Clearly there is something going on that has undermined the wonderful intentions of this board and until that is fully understood, it might be a good idea to remove at least one temptation for abuse, namely the cash awards.

Mr. Neb

 

[Laurel]

Hi Mr_Neb,

Thanks for taking the time to write out your concerns in detail. We absolutely appreciate hearing what you have to say, and I will attempt to address the issues you brought up.

I understand your concern that your stories could be intentionally "slammed" with 1-votes. I took a quick check of your stories. Since the 1-votes and 5-votes are the extremes, I only paid attention to those in my quick review.

You have 25 stories posted on the site, with a total of 942 votes total as of when I checked. Of those, 46.6% (439 votes) were 5-votes. Only 3.5% (33 votes) were 1-votes. The most 1-votes garnered by any one of your stories was 6. (That same story had 33 5-votes.) The majority of your stories (15 of them) had only a single 1-vote, and 5 of your stories had no 1-votes at all, though every story had at least one 5-vote.

Just by looking at those numbers, and the related data, I did not see any indication of obvious fraud. Quite the contrary, it looked to me like your stories were - and are - very popular with the vast majority of Literotica readers. I've read all of your stories myself, and I must say that the fact that they are popular does not come as a surprise to me personally.

As far as the movie analogy, my point was that a viewer's expectations can influence their experience. If a movie has won 17 Oscars, then your expectations of that movie may be higher than your expectations for a low-budget independent movie. You need go no further than the User Reviews section of IMDB to see individual people pummeling popular movies, and very specifically calling them "over-rated" and "all hype". If a highly touted movie, or in our case a story, doesn't live up to a some people's expectations, I don't think it's unreasonable to assume that a small percentage of those people may be reactionary, or activist, and try to lower the expectations of that story by giving it a lesser review than they would have had it not been so highly recommended.

As I've said previously, the improvements to the voting system that are about to be implemented will give us a little better idea of whether votes are coming from registered or unregistered members. We are also always working on improving the cheat protection mechanism. At some point in the future - hardware, software, and data willing - I think your idea of allowing authors to see when votes were cast is not a bad one. Manu and I were talking about it, and we were thinking that maybe at some point each story could have a graph that showed the voting stats over the life of the story. I have no idea how hard this would be to do. I'm sure it wouldn't be any time really soon, but it's definitely an interesting idea for the future. Thanks again for taking the time to help improve the site!

 

[Mr_Neb]

Thanks for doing all that digging. Before I go any further, I want to make sure that there is no misunderstanding. I don't mean to be negative or unappreciative of you or the board. I do appreciate all the hard work that goes into providing this resource for authors and readers alike.

Yes, I do seem to enjoy some level of popularity (blush) and my comments aren't any attempt to improve on that. It's just that readership, at least as can be estimated by the number of votes cast, is affected by a story's scores. Anyone's stories that have modest scores in the low to mid 4s stay on the Top List longer than those that score much higher. The higher scoring stories fall much lower on the list and get fewer total votes and therefore I presume fewer readers. Few readers means fewer feedback emails and those are what I enjoy the most.

If possible, can you determine the order of the votes? I would be very interested in when the 1s came in on "What I Want".

Thanks again for your efforts and your patience with all of us.

Mr. Neb

 

[Maximus]

A mere 3% of your votes are ones and you're still complaining? Laurel, you have a saint's patience. If I ran this site I would tell these yahoos to take a walk off a pier.

 

[Mr_Neb]

Maximus,

What is it exactly that you think I am complaing about? If it is the low scores of my stories, then you have missed the point completely.

I am not concerned with 1-votes cast if the voter did not like the story. I am perturbed (and a lot less than others on this board) if ANY story gets a 1 just because of it's position on the Top List.

I hope that clears things up.

Yahoo Neb

 

[Guest]

Lies, damned lies, and ...

To start off, I left university as a mathematical Logician, so I know a little about statistics. I al over 60 years old, so I have seen a lot of people's behaviour.

quote from Mr_Neb
---------------------------------------------
My latest story, "What I Want" was to my pleasant surprise at 4.75 after 17 votes. At 20, it was 4.65 and at now at 25, it is at 4.32. I would very much like to know what the votes were and in what order. The same has happened with several other stories. If the voting were truely a reasonable reflection of the popularity of a story as determined by even a slightyl imperfect system, the distrubution of votes would be relatively uniform. The probability of having a reader like a story or dislike it is the same at any one point in time.
---------------------------------------------

On a purely statistical basis over a large readership of many stories this would be true with one enormous proviso.

If, and only if, the readers did not have any criteria for preselecting stories they expect to like.

I, and I expect some others, look at the new list and if I see one by an author I have previously enjoyed, I read it. Most authors who write one story I have enjoyed, continue to please me. I suspect that this effect is enough to cause a skewing of the popularity of the story with early readers.

In lay terms, your fan club reads you first and votes you higher than the random reader.

The same sort of effect will happen with number one stories. People will vote them to number one by reading them from an author name, and from a category, and randomly.

Then some people will read them because they are number one and for no other reason. Is it not likely that those readers will, on average, like them less because they are not ones they have otherwise selected?

Statistically speaking, to win the monthly prize, ensure your new story is posted early enough in the month for your real fans to read it (and post enough votes to qualify) and late enough for the rest not to find it yet.

Finally, I don't care because I write elsewhere professionally and therefore don't qualify for amateur prizes.

PS to Laurel - I hope that doesn't count as "vote hacking information"

 

[Guest]

When I started that last reply I *was* logged in.

This has happened before. I seem to have to log in several times a session!

Anyway, Unregistered of a few moments ago was I.

No apology this time because I am sure it wasn't my fault.

 

[Guest]

I have just seen the times read count and, as someone else points out, give your story the title "A AAA Alcoholic ..." and put it in "Incest" to get lots of random reads.

Beware, they may not mark you as high as your fan club.

PS Laurel ----ditto----

 

[DVS]

Laurel, I just want to say thank you for addressing these questions for us. I for one just want to understand all of this. I feel I must agree with Mr_Neb that there is at least one person going around and voting 1s on high rated stories.

You mentioned the stats of his stories, and the ratio of 5s against 1s voted. I have not seen where his stories are rated at present, but I would be curious to know when the 5s were cast and when the 1s were cast. Let me explain a little for you.

I have a story broken into chapters posted here. I do not have a lot of friends who will come and vote highly for me. Actually, I wish I did. I do agree that might be something that happens.

The voting on my chapters has been slow and increasing up to the point they will get an H rating. Now, this H rating won't have a chance to continue much more than the threshold because there will be someone come along and vote a 1 to effectively kill that H.

Then, the voting slowly increases again until the H is achieved once again. Then, another attack to remove the H. This has happened several times. It seems to me that if the 1s were random, I might never reach the H status. But, because I do, it seems to be more a vindictive thing, waiting until I do.

And, in addition to the 1 votes, the chapters that don't have this H are not changed. Only the chapters with an H are 'hit'. This is all one story and has the same content throughout. So, why would someone only vote against a few chapters, instead of all? If the voter doesn't like my story, why only vote against a few chapters, and why just the H rated ones?

OK, I can see 1 person doing this, but this has happened more than once. Sure, I know people will vote for or against one chapter and not the whole story, but not so consistently. I can't see this as being random or coincidence.

It has also been said in other threads that the 1 vote has more weight than a 5. I would like to know if this is true. It just seems so easy for someone to come and whack a rating, and so difficult to regain it. I recently lost 3 Hs to one vote each. This put these chapters so low, it will be quite a while before I will see another H.
But, I can't wait to see if the H is killed again, if and when it does show.

It is also interesting that the BDSM category doesn't have any stories much higher than the threshold of the 4.50 rating. That could be because the stories receive a 1 when they get that far. A 1 will just remove the H and not much more. So, there are a lot of stories hovering just below that point.

I may seem overly concerned about my H status, but that isn't the point. I don't care that much about what happens, if the voting is fair and honest. And, my voting history is all I have to help me explain what I feel is happening.
If there is a vendetta of some sort happening, it should be addressed, for the better of all voters and writers.

I think the voting system is fine, if it is used correctly by voters. I am just trying to understand if that is truly the case.

Again, I think this is a very healthy talk about this voting thing. I sincerely appreciate you responding to our rants.

To better understand my point, please look at the below thread. I tried to explain it there.

www.literotica.com/forum/showthread.php?s=&threadid=88599