Most reviewers are fake. Most reviews are not.

[carlos marvado]

Great statistical work and analysis Brother El Gordo........sort of adds a bit of science to both the selection of escorts and the review system. Some time ago, while studying sites across Central Europe to select an escort for a punthaton of mine, I got into the habit of using height and weight data to calculate their body mass index.....great fun.

I think we should set up a group here for mathematical and statistical anoraks.

[ber]

A great piece of research, certainly makes sense to me,also in the main it vindicates the
review system,provided u know how to evaluate and use it properly.
Very well done,cant wait for ure next piece only problem uve set ureself a high standard

[El Gordo]

I just wanted to clarify one point. Counting first time reviewers is no substitute for reading reviews. For example, you can see above that Karin has an above average number of first time reviewers. Even with 47 reviews there isn't enough data to say that this means anything. If you actually look at the reviews you see, for example, a very positive review from ninebythree written a couple of weeks ago. Personally I wouldn't have any hesitation about seeing Karen even if the number of first time reviewers is slightly above average. On the other hand, there is an escort who has recently returned to Dublin who has a smaller percentage of first time reviewers than Karen. But if you look at those reviews a little you begin to notice things. For example there are two reviewers who have multiple reviews, but all of those reviews are of this one escort. Another has 4 reviews, of which 3 are of this particular escort. There are also unusually many ratted reviews. None of this would convict anyone in a court of law. In fact it could all be innocent, which is one of the reasons I don't want to name the escort in question, but personally I wouldn't pay 200 euro an hour of my own money just to find out.

For those with short attention spans: counting first time reviewers is useful for finding general patterns, like the fact that some escorts in the running for Escort of the Month probably can't resist the temptation to cheat. It's not a magic wand to identify dishonest escorts. You still need to read reviews if you don't want nasty surprises.

[Karin Kiss]

There are currently about 2,163 reviews which accounts for almost 50% of all visible reviews (as some reviews opted out of the review system). If half the reviews are single reviews, an escort having a 15%, a 20% of them woudn't seem to me something to worry about.

True, some escorts do attract first time clients more then others.
Each single day I am working I find at least one person that never visited an escort before.
As El Gordo said it in one of his previous postings, I do look quite innocent (it's just the look BTW ) and nervous first timers are comfortable with me. I used to ask majority of my clients to leave a honest review of our encounters and I often tried to convince them to register for their first review. I am not doing that anymore as clearly this doesn't do me any favours. Facts speak for themselves and you'll probably notice that 2 trusted reviewers on my page had their first reviews on me, I hope this would constitute some sort of evidence that yes, I do attract first timers. Given time, many of those first timers would go on and write more reviews.

The percent calculated on me is in fact smaller. One of my reviewers had in fact 2 reviews, but the other girl he left a review for decided to opt out of the review system. The "karin on karin" review belongs in fact to someone else who left a single review, so they are in fact 2 (both ratted due to the unfortunate circumstances).

I don't want to change the beautiful presentation of El Gordo in something about myself so I will refrain to speak further about myself and my reviews. His second posting (about reading he reviews) makes a wonderful addition to the first. Good statistical procedures are reliable ways of learning from data, ones which are unlikely to lead us into errors, however taking a little time to read the reviews of any escort would bring more reliable information.

[MR JOLLY]

Umm yes a scientific approach to cutting through the bullplop as an aid to selecting the perfect girl.... lets see now. after numerous calculations, computations, numerations, masturbations, statistical analysis, data analysis Boxplots, Confidence Intervals, Curve Fitting, Datasets, Dotplots, Histograms, Hypothesis Testing, Linear Regression, Looking for Patterns, Measures of Location, Measures of Spread, Normal Distribution, Normal Plots, Probability, Sampling, Scatterplots, Stemplots.

and the results are in,




tonight the girl I will be calling IS.....





the one with the biggest boobs.

[Alec Horan]

Karin did an excellent spreadsheet a while back which analysed the reviews perfectly. Unfortunately it was not adopted here. The best review system is the one on adultwork in the uk if you want to see the best system available.

Re: first time reviewers and Karin. Karin is posting here a lot and puts herself across very well. If you are a newbie I would have thought that you are going to be inclined to gravitate towards the ladies that post on here on a regular basis. Ditto Dolly, mysterious Diana, Anna, Elle.

By the way, who is to say that Scankman, zoobanzoo, ber, blackcat etc amongst the multiple reviewers are not pimps. It would seem unusual to me for a pimp not to ensure his ladies were well advertised and reviewed.

p.s. is it in EI's best interests to keep poor reviews of girls if those girls have advertised on here for a considerable length of time ? Thats where EI gets their income (?). And there will always be punters.

[ninebythree]

In fairness, karin analysis was spot on and should have been adopted.

All i can do is take ure suggestions and give them to Patricia.

In fairness people dont realise how slow changes are made around here.......

6 months for client warning section and nearly the same for same for new chatroom.

When i started the review system bbj and i had compiled a list of of 30 odd changes to review system, compiled from suggestions on message board and our own ideas......five of them have been done.

As for the escort of the month, it has been turned in to a farse with four points for badged reveiwer.
I have suggested it before that only one review per reviewer should be counted.

But as every other suggestion ,that has been made with regards to review system falls on deaf ears. At least we know were priorities lie with patricia , and unfortunately for the the genuine punters out there that spend money .........its not with us.


nbt

[Stewie]

I had a look at your reviews I counted 5 one time only reviews

2 reviewers that had reviewed you twice and reviewed no one else I presume that is how el gordo arrived at 9 as he has said he was ignoring reviews marked fake.

However you are correct if I'm right then the calculation of 19% for you is wrong.

You have 47 reviews if we are disregarding fakes then your total is 45
however in my view it is incorrect to count both of the 2 people who reviewed only you twice only their first review should be counted with the second disregarded completely. Hence you end up with 7 out of 43 which is 16% of slightly above the average.


As you say girls opt out and guys drop back to 1 review skews your average. Which proves it lies damn lies and statistics.

As you say given that so many reviews are first time only it would be normal for a girl to have a percentage of them somewhere between 15 to 20.


Also the stats are a little unfair on English Taylor given she only has 26 reviews it might be interesting to count the first 26 girls of those listed to see if he level is as above average as it seems.


Also the verified reviewer has changed things however I think ric has made a mistake making the points for verified reviews so high compared to other reviews given that nbt has not nearly finished with medals yet. I looked at the current standing and they appear to be skewed by Jack sparrows 2 reviews for ttaylor giving her 8 points. Not saying any wrongdoing on eithers part just that it is unfair on other girls whose reviewers may just not have recieved their medal yet.

There are currently about 2,163 reviews which accounts for almost 50% of all visible reviews (as some reviews opted out of the review system). If half the reviews are single reviews, an escort having a 15%, a 20% of them woudn't seem to me something to worry about.

True, some escorts do attract first time clients more then others.
Each single day I am working I find at least one person that never visited an escort before.
As El Gordo said it in one of his previous postings, I do look quite innocent (it's just the look BTW ) and nervous first timers are comfortable with me. I used to ask majority of my clients to leave a honest review of our encounters and I often tried to convince them to register for their first review. I am not doing that anymore as clearly this doesn't do me any favours. Facts speak for themselves and you'll probably notice that 2 trusted reviewers on my page had their first reviews on me, I hope this would constitute some sort of evidence that yes, I do attract first timers. Given time, many of those first timers would go on and write more reviews.

The percent calculated on me is in fact smaller. One of my reviewers had in fact 2 reviews, but the other girl he left a review for decided to opt out of the review system. The "karin on karin" review belongs in fact to someone else who left a single review, so they are in fact 2 (both ratted due to the unfortunate circumstances).

I don't want to change the beautiful presentation of El Gordo in something about myself so I will refrain to speak further about myself and my reviews. His second posting (about reading he reviews) makes a wonderful addition to the first. Good statistical procedures are reliable ways of learning from data, ones which are unlikely to lead us into errors, however taking a little time to read the reviews of any escort would bring more reliable information.

[adogwithabone]

I just wanted to clarify one point. Counting first time reviewers is no substitute for reading reviews. For example, you can see above that Karin has an above average number of first time reviewers. Even with 47 reviews there isn't enough data to say that this means anything. If you actually look at the reviews you see, for example, a very positive review from ninebythree written a couple of weeks ago. Personally I wouldn't have any hesitation about seeing Karen even if the number of first time reviewers is slightly above average. On the other hand, there is an escort who has recently returned to Dublin who has a smaller percentage of first time reviewers than Karen. But if you look at those reviews a little you begin to notice things. For example there are two reviewers who have multiple reviews, but all of those reviews are of this one escort. Another has 4 reviews, of which 3 are of this particular escort. There are also unusually many ratted reviews. None of this would convict anyone in a court of law. In fact it could all be innocent, which is one of the reasons I don't want to name the escort in question, but personally I wouldn't pay 200 euro an hour of my own money just to find out.

For those with short attention spans: counting first time reviewers is useful for finding general patterns, like the fact that some escorts in the running for Escort of the Month probably can't resist the temptation to cheat. It's not a magic wand to identify dishonest escorts. You still need to read reviews if you don't want nasty surprises.

I have said it before, all once-off reviews should automatically delete after say 2 months if there is no further activity by the reviewer. All ratted reviews should automatically delete after 2 weeks unless the reviewer disputes the rat label. Reviewers should only be allowed one review per escort.

[Alyssa Jenkins]

I have said it before, all once-off reviews should automatically delete after say 2 months if there is no further activity by the reviewer. All ratted reviews should automatically delete after 2 weeks unless the reviewer disputes the rat label. Reviewers should only be allowed one review per escort.

dogwithabone I totally agree with your methods here I think it is reasonable or unless like suggested before if you would like to re review a lady then it should replace the older review this makes great sense to me xxxx