Lies, Damned Lies And Statistics

Introduction

One of the great things about having a store of data such as this is that we can analyze it in all sorts of interesting ways. What we need to remember, however, is that in converting real world concepts to numbers, interpreting those numbers, and then converting them back to real world concepts, false assumptions are very easy to make.

I will not begin to claim that the assumptions made by the site are in any way perfect. In most cases there are multiple interpretations and any argument put forward for one interpretation has a counter argument I could equally suggest. The aim of this page is to at least explain the assumptions that have been made so that, even if they are not perfect, the reasoning behind them can be understood.

Suggested Scenes

For Single Participants

Each rating is then squared. So, Hate is 1x1 (1) while Essential is 6x6 (36). Finally the score is doubled if it has been tried. This gives a value from 1 (Hate it, never tried it) to 72 (Done it and it is Essential).

Now all of the activities are entered in to a list with one place per point. The activities are chosen from this list, meaning that something that is Essential and has been tried is 72 times as likely to come up as something that is Hated and never tried.

For Couples

Now the scores for both partners are added together. Any activities that do not have a combined score of 7 or more (Essential(6) + Hate(1); Love(5)+Dislike(2); Like(4)+Don't Mind(3)) are discarded.

Next we take the original scores for both the Dom and the sub, separately and subtract three - so Don't Mind now counts as zero. This score is multiplied by 2 if they have tried the act. So, if they Hated the act and have tried it, it counts as (1-3 = -2 x 2 = -4) and if they find an act Essential but haven't tried it, it counts as (6-3 = 3). The individual scores for each person are now added together. Any combined scores of zero or less are discarded. This means that if one person Hates an act and has tried it and the other one Loves it but has not tried it, the weight of experience will swing it.

Finally, this combined score is squared and used to populate a table, just as in the single participant case. The range of possible entries goes from 1 (everything lower is discarded) to 144 (Essential (6) - 3 = 3 * 2 (tried) = 6 for both of them, which gives 6 + 6 = 12, squared = 144).

Compared Checklists

Again, a score of 1-6 for Hate-Essential is used. Curious is given an unbiased 3. Hard Limits are -2 and Soft Limits are -1. The missing zero entry is deliberate as it adds a degree of extra emphasis to Limits.

Now the difference between the two scores is found. That gives a range of zero (for both exactly the same) through to 8 (for one is Essential (6) and the other Hate (-2). Starting from 100%, 25% is removed for each point's difference. So, two identical scores get 100%, 1 apart (i.e. Love (5) and Like (4)) get 75%, 2 apart get 50%, 3 apart get 25% and 4 or more apart get 0%.

Finally, the average, overall, is found and given, rounded off to the nearest whole number.

The average only really gains any degree of accuracy as more entries are given. After all, comparing two or three aspects of two people's sexuality does not give much of a clue, while comparing two hundred aspects can give a very good impression.

As a quick analysis, an average of 75% compatibility or more should be considered a very good match. It implies that, across the whole spectrum, the average ratings matched to within a point of each other - that's a pretty impressive match.

An average even as low as 50% is still pretty positive. After all, that is the equivalent difference of one person finding something Essential and the other merely Liking it. While not perfect for each other, that still implies they can make things work fairly easily.

Alternative Methods

We looked at whether we should use a boolean (100% or 0%) scoring system. After all, if one person hates something and the other has a hard limit, surely they are still a perfect match - after all, neither one is going to want to do it. While that is largely true, we felt that this gave no degree of flexibility. It did not address cases where, while something may be hated, perhaps doing it occasionally, for someone else, brings a degree of pleasure.

We also looked at separate rankings for each possible combination. While this has a lot of potential, it is also very open to our subjective opinions. After all, just because we believed that the difference from say an Essential to a Like is greater than a Love to a Don't Mind, does not mean that is always the case.

So, as we mentioned at the beginning, while it is not by-any-means perfect, we went with the simple system because it is at least consistent. If it means a lot to you, I am sure you can perform your own analysis.