More Review Statistics

[Alec Horan]

Statistics are great but the more you get into them the more you realise what it is you have to consider.

Sterling work by the way. No problem with that. Just I got to thinking about the myriad of issues that effect the figures.
I do love it when median is used, in conjunction with the mean. I have always felt that the median is far less understood and used than it ought to be.

With regards to escorts based in large towns, chances are they will have more reviews. Girls that may only work 2/3 days a week or girls that may hit Ireland for only a couple months every year are going to have slanted figures also.

Even regular reviewers, trusted though they may be need to be taken into account. Just have a look at the average rating of some, scankman usually gives a positive review whereas nbt is less inclined to do so. Therefore a girl may be affected by the luck of the draw as per their reviewer. (perhaps scank is old and ugly and anything is appreciated, whereas nbt is a young buck who can have any woman he wants and finds it a bit of a chore (or he's gay))

The more you learn the more you realise how much you don't know.

As someone stated earlier,after much consideration they decided to chose the girl with the biggest rack !

And there is an awful lot to be said for that. Rationality has its place but Mother Nature has 3.5 billion years of experience.

[Alec Horan]

El Gordo and Karin.

Hats off and kudos to both but please promise us you two won't team up for the new interview competition.... My head hurts...

Shock !Horror! I didnt think it was possible. I remember someone that used to post epics on an almost daily basis.

I kinda miss those posts.........well the first 3 paragraphs not the last 5 !

[Smarty]

Well done El Gordo.

Its no wonder you hav'nt been on the boards much recently having to prepare this fine report.

[Wildthang]

yea yea... well done and all tht blah blah blah stuf. its simple....... fuck the review system cause we all like different things and wht floats one guys boat might not float another.

1: look at the pic (if your mickey twitches= see step 2)
2: Call said escort (if mickey gets harder during convo =see step 3)
3: Go put mickey in said escort ( when mickey spits at escort= see step 4)
4: pay escort and fuck off home till the next time you want to put your mickey in somthing.

[El Gordo]

Excellent work again El Gordo !!! I am very confident you are going to find eventually a fool proof scientific way to calculate the credibility of escort reviews, I am just not very sure that the median number is the solution.

You have more faith in my abilities than I do. I don't think any statistical method will ever be a good substitute for reading reviews.

The median number could indeed be a very useful tool for calculating the reliability of some escorts.
Why am I saying some instead of all? Because in my opinion, this number would help you decide indeed between some relatively new escorts with only a few reviews and for which no other reliable information is available (like recommendation, lady's own postings etc). If you have to chose between escort A, escort B and escort C (with the reviews as below, there is indeed a much better chance for you to pick escort C in this case:

Escort A: 1, 1, 2, 3, 4.
Escort B: 1, 1, 3, 5, 6.
Escort C: 1, 3, 5, 8, 10.

I wouldn't use the median, or any other number, in this situation. First, there isn't enough data. Second, there are only 15 reviews. I would just read them all.

However, the situation complicates a lot when you have to decide between escorts with more reviews. Take for example Mellisa, Kat and Alyssa:

Escort Reviews of Melissa, Irish escort from Escort Ireland
Escort Reviews of Kat, Irish escort from Escort Ireland
Escort Reviews of Alyssa, Irish escort from Escort Ireland

Based on your stats, the reviews reliability for these ladies would be:

1 st place: Melissa
2nd place: Kat
3rd place: Alyssa

Well, looking trough their reviews we can see for a start that Melissa has a total of 30 reviews, from which:

- 10 ratted positive reviews
- 15 positive reviews
- 2 negative reviews
- 1 neutral review
- 1 N/A review

In comparation, Kat would have 21 reviews, from which:

- 1 single ratted positive review
- 19 positive reviews
- 1 negative reviews

Alyssa's situation for her 70 reviews:

- 1 single ratted negative review
- 68 positive reviews
- 1 N/A review

You have certainly picked the right example. Melissa is very unusual. If you read the notes for the ratted reviews then it becomes pretty clear. She, or someone trying to help her, has done exactly what I said an escort would have to do in order to cheat the system: Many fake reviews from several different accounts. If you look through the other escorts at the top of the list you'll see that she's the only one who's done this. I could have made this harder by ignoring additional reviews of the same escort and ratted reviews. It would have required a lot more effort, though, and I'm not getting paid for this.

In conclusion: the median number cannot be used to check if escort reviews are more reliable then others. Who would be more credible, an escort with a third of her reviews ratted & 50 % positive unrated reviews (Mellisa) or an escort with 90 % positive unrated reviews (Kat) or another escort with 97 % of positive unrated reviews (Alyssa) ???

Just to be clear, reliable and good are not the same thing. Suppose an escort has three negative reviews, from scankman, ninebythree and zoobanzoo. Her median is 86! I would think her reviews are reliable, meaning I would believe them. They just aren't good.

In your example, I would pick Kat or Alyssa over Mellisa. I know you didn't ask, but I would also pick you over many escorts who appear higher up in the list.

P.S. In reality, as the ladies I picked as an example here have each a sufficient number of reviews to make them very credible as escorts, a punter will see any of the ladies I picked as an example based on how attractive he would find them, their price and their availability. If I am allowed to quote Mr Jolly's joke that he made on your previous thread about reviews: "after several computations, permutation etc ... the one I'm going to see tonite is ... the one with bigger tits ".

True. And once I find some reviews I trust I don't worry so much about whether the others are real.

P.P.S. I really find your work, your time & efforts simply excellent and I do apologise if my reply to your thread may seem like I contest them. I firmly believe that scientific measures could be useful in this case, I just think that something as simple as the median number is not a measure powerful enough to be applied across all reviews and escorts.

I'm aware that the median is a pretty crude measure. I hope no clients are working their way down the list, calling the escorts in order. I think the median is still useful if applied with caution. For escorts I've actually met (not you so far ) I find that they are roughly where I would expect them to be. Your number is low, but I think I understand the reasons. Some simple fixes, like discarding additional and ratted reviews, would help. Beyond that, I don't think it's possible to come up with a single number which will tell someone which escort to call.

[luther]

And great analysis of it Karin.........

[El Gordo]

yea yea... well done and all tht blah blah blah stuf. its simple....... fuck the review system cause we all like different things and wht floats one guys boat might not float another.

1: look at the pic (if your mickey twitches= see step 2)
2: Call said escort (if mickey gets harder during convo =see step 3)
3: Go put mickey in said escort ( when mickey spits at escort= see step 4)
4: pay escort and fuck off home till the next time you want to put your mickey in somthing.

I usually want to know if the escort will look like the pic. My time isn't free. Also, I usually pay before step 3. The escorts I've visited seem pretty insistent about that.

[El Gordo]

Well done El Gordo.

Its no wonder you hav'nt been on the boards much recently having to prepare this fine report.

I was able to automate most of it. A lot of it happened while I was asleep.

[El Gordo]

Great work but I think that the results are distorted by multiple reviews from high-number reviewers. Fori nsteance, an escort with only three reviews but all from Skankman would have a number of 108!!!

For escorts with lots of reviews this isn't a problem. I gave Eve's numbers in the original post: 1 2 2 2 2 3 3 3 4 5 5 5 5 5 6 6 7 8 9 10 10 11 11 11 12 12 13 13 13 17 18 18 21 21 29 29 29 36 36 36 36 74. Now let's say Scankman visits her four times. An excellent idea, if I may say so! I wish I could afford to do that. If he writes a review each time then her numbers change to 1 2 2 2 2 3 3 3 4 5 5 5 5 5 6 6 7 8 9 10 10 11 11 11 12 12 13 13 13 17 18 18 21 21 29 29 29 36 36 36 36 74 108 108 108 108. I've marked the median in both series. As you can see, it's still 11. That's the advantage of the mean: It doesn't change much.

[Wildthang]

I usually want to know if the escort will look like the pic. My time isn't free. Also, I usually pay before step 3. The escorts I've visited seem pretty insistent about that.

U must come across as untrustworthy then. Also you need to remember its only money at the end of the day.