All Comments on 'How to Analyze Your Scores'

thank you. The formula(s) you provided were quite useful, intuitive and helpful.

Of course, a big question remains.....Why did a voter register a "5" or, worse, a "1"?

What would be most helpful would be Literotica not allowing votes by anonymous. Since most (I guess just about all) names we see on Literotica are pseudonyms already, what can possibly be gained by allowing this extra layer of anonymity?

One more thing: There ought to be a method for Literotica to give extra weight to votes which are accompanied by a few words in the comment box. For sure, comments by readers are more helpful to a writer hoping to improve his/her craft than is a mere raw vote.

At any rate, thanks for your input.

Thank you for taking the time and trouble to prepare and submit this fascinating post. I have often wondered how the scores are actually treated to bring us the numerical values indicated; this goes a long way in explaining just that.

I agree with much of what the previous commentator had to say, particularly regarding anonymous votes; there are other sites where one has, from time to time, to be logged in to vote on stories and always to comment on stories, and I think it would be a good thing to introduce here - we are, after all, to some degree analogous to Alcoholics Anonymous - Perverts Pseudonymous, perhaps? and it would require only a little more effort to pin a name to a comment, if not exactly to a score.

Another idea might be, and I am not suggesting this should be taken up, merely 'flying a kite', that in order to post a story, you should have had to comment on at least one other - 'this is rubbish' type comments wouldn't count. We are all here for our own reasons, but there is not one of us who could not benefit from an honest examination of his or her work.

I would be more than happy to discuss this further by email if anyone is interested, even if it may not be something for the owners/admin to take on board - yet.

mel_pomene

You did a useful thing.

Dissecting the math, as you've done, does provide useful information, much more than a raw score.

I too would support changing to a 10-point scoring system; the current five-point is just too compressive. Is a 4.45 story that much worse than a 4.57? Maybe so, but I dont think your analysis is sufficiently precise to sort out the 0.12 difference. And that's not the fault of your methodology, but rather is inherent in the compressive nature of the scoring system itself. If we're trying to do a sheep-and-goats sorting (and I stress the word "if", because I'll keep writing and posting whatever my scores are or aren't), then differentiation is where it's at...or should be. If 60% (3.0) approval is utter failure, something's wrong with the system.

and dizzying. Good work though.

I have long suspected that they always dump all of the one star votes because they are obviously malicious in nature. I think anything that makes it through the vetting process to get posted is worth more than a zero, which is what a one star vote really is. If I can't give a story at least a 3, I don't vote. Why punish somebody?

As was mentioned elsewhere I too have always favored a 10 point system, but the scores mean less to me than the comments and feedback anyway. I know how to score high when that's my goal. I choose the author name I submit stories under depending on whether I want to score high, get feedback or tick people off.

Also, the relatively new way scores are explained to those choosing to rate stories under the stars has led to lower scoring since this was introduced to replace the old thermometer type scoring system. If you get an H these days, you're doing something right. They are as tough to get in 2012 as they were easy to get in 2008.

Oh my aching brain!

I'm giving you five stars simply for the math. My head hasn't hurt that much since I had to take probability in university. The one missing component for me though is the human element which you failed to address.

If perchance you are writing say in the Loving Wives category and you write a Cheating Wife story or a Wife Sharing story or God forbid a Cuckolds story and score a 2.15 what would your actual score be minus the odd folks who post as Anonymous? I mean does that translate to a 4.25 minus the nuts or is that truly indicative of someone who should never put pen to paper?

Good or bad, if I comment I will notate score for writers benefit/detriment. (5 stars)

But I am an accountant.

As to a ten point system, advocated by some commenters: When you can figure out how to convert the thousands of stories ( with the average of hundreds of votes each ) from a five point system to ten without it turning into a giant mess, I'm sure it will happen immediately :)

Every single one of the votes within the system would have to be converted, and the only thing you could really do is double them. That's going to put the existing stories on a completely different scale from anything new that comes out because none of the existing votes would be odd numbers, and it would eat system resources like crazy for a fair amount of time to get there.

Saying "They should change it" and understanding the enormity of what that actually requires, along with the new problems it would create are worlds apart.

Sites with 10 point scoring systems have exactly the same problem of inflated scores. It is the nature of voting that those who enjoy the story are far more likely to vote than those who don't. Most ( other that those determined to lower the score ) are going to back-click without voting.

As to the essay, I think you left out something rather critical -- at least so far as the title goes. You can evaluate how voters scored your story via this method, but it doesn't tell you much about how that compares to other stories in your own category, each of which have vastly different voting patterns and score averages. A 4.5 in NonHuman isn't remotely the same score as a 4.5 in Loving Wives.

Yay. ^_^ I enjoyed that. Thanks for putting in the effort, CarlusMagnus. Even better than the thread where it came up.

While reading this, it never occurred to me that you should involved the "human" element here in analyzing your score. That would be "interpreting your score" (which isn't the same as analyzing). You as the author (or lecturer..I was having a "professor" moment here, in a good way) explicitly stated that you were only doing the math to determine the range of the differing tiers of votes.

You did exactly what you stated that you'd do. :)

Although a very technical article it is extremely useful for those that continually analyze their scores.

about the problem with going to a ten-point scoring system. It wouldn't be possible to convert prior scores with anything like a fair (or even an accurate) result. And the ultimate argument against going to a ten-point system is still compression: a greater range in possible scores doesn't mandate a greater range of actual scores. Can anybody name a wine Robert Parker rated as a 75? His 100-point system really is a 12-point system. The people who follow Parker would run away from a wine rated at less than a Parker 88. A ten-point system would still be a two-point system: either a 9 or a 10, and anything less is a bomb. So I withdraw my support for a ten-point scoring system: it would do no good and only do harm. Better the devil you know....

A very important consideration when discussing votes is to consider who the votes are for. Are they there to help readers find good stories? To encourage writers to keep writing? or to provide a honest feedback to the writer about what the readers think of the story? Depending on purpose, dismissing any "one star bombs" as trolling might be perfectly appropriate, or way out of line.

As mentioned in previous comments the votes are quite inflated in general on literotica. In my opinion they are inflated to the degree they are almost useless for the first and the third purpose but fulfill the second purpose quite nicely, as almost everyone will have a decent score and therefore encouragement to keep writing.

Anyway thanks for the analysis, and I hope that literotica allows authors at least to se the distribution of the votes directly some time in the future.

if votes are removed why bother voting

And here's to 'em!....... Voting really is too easy for trolls, and for fans too, and I think that skews things....... There are gamers and there are other gamers..... at the same time the really good authors end up leaving, don't they? And the bad ones just keep plugging away. I sympathize with the whole 'logging in to vote' thing, but that's just not LIT....... btw...... I read a few of his stories, and I wouldn't disagree with the scores. I spspect he'll be leaving soon.

And statistically sound! My two cents: You should always take into consideration whether your story is a multi-chapter story or a "single-chapter" story. Why? Look at the Hall of Fame of any category. Very often, the top ranks are filled with the final chapters of (really great) stories. However, if you check out the author's profile, often their equally great, if not better because it has drawn you in, first chapter gets a much lower score. My assumption is: nobody jumps into the final chapter without reading the ones before. And people who give four or less stars stop reading after a couple of chapters. If you're still with the author at chapter 7, you must love his work, and you're more likely to give him a high score. Also, there is the factor of giving the author a vote for his "overall work" in the final chapter, rather than rating the chapter itsself. Plenty of reasons why the final chapter of a piece is more likely to have a high score than a "single chapter" piece.

I've attempted to automate your figures and hit one stumbling block:

> a) if S does not exceed 3.00, n is at most N × (S – 0.995)/2, and

> b) if S is 3.00 or larger, n is at most N × (5.005 – S)/2.

The case of a score of 3 satisfies both conditions. Do you mean for (a) "does not exceed 2.99"?

This posting and its comments have been most informative. I've been writing here for over ten years and still didn't know about the sweeps. Frankly, I think that Literotica does a pretty good job at handling the scoring. And you, CarlusMagnus, have done a great job at shedding a little light on the methodology. Thanks.

I've noticed some of my stories get lots of votes (eg, over 100) and some get much less (the lowest is 29, so far). It's similar with the number of reads. I suspect some categories (such as incest) get more reads and votes than other categories (such as exhibitionism) but I have no real way of knowing. Does Literotica tally that information, and if so, are you willing to share it?

Examples from my own stories of the ratios number of ratings/number of reads is .004, .006., .009. Those are three at chosen random (I have 25 stories so far). These are all exhibitionist stories. Thank you. JBE

Hi RJB -- Thank you for your insightful and helpful comments! I greatly appreciate it. Yours, JBEdwards

Love this, add that you went to the effort to do this.

It would be interesting if someone had access to the full database of scoring, because presumably then you could calculate not just the minimums and maximums, but the probability of the number of X ranks that one received, based upon one's overall score. It would be interesting as well to see if the data vary a lot from one category to another.

Thank you!

where exactly can I see the vote counter? I've only published one story and while I can see my views and the changing average rating at any time, to this point I've never seen a stat that tells me how many people voted on my story! I have a 4.8 average rating, which could be anything from 5 votes technically, (4 five stars and 1 four stars) to 20, 100 or any multiple of 5 that's lower than my view count. (as well as many other numbers in between if someone gave me a 2, 3 etc) I've been inactive for 2 months on literotica because of some personal issues so I missed the opportunity to manually keep count, but now that I came back and I'm working on chapter 2 I was just curious to check how many votes I've got, given that everybody seems to assume on the forums that this is one of the few stats that you have access to...

I assume it's just 5 or 10 votes, but any help with finding the counter to actually check it is much appreciated!

I wish I would have found this post long ago as I’ve gone through many of these machinations in the past couple of years. Cudos to you, this was well done. Ironically I just had a newer story make an unheard of jump of .41 on a reduction of over 50 votes from around 475 to 418. This was obviously a sweep induced outcome. I’m curious what it will do over the next few days. I was ecstatic that it happened, as this was the first story/chapter that I’ve written out of 36 that hadn’t achieved hot status even though I think it was one of the better things I’d written. This bounce moved it from 4.30 to 4.71. 🤞it stays.

Brilliant. Just brilliant. Well analysed.