Statistical Approach to Relationships
I’ve now got a list of roughly 20 women I’ve dated or wanted to date; enough to start looking for patterns. I’ve listed the following attributes for each woman: Name, Occupation/Major, Relative Age to me (older, younger, same), If I pursued her, if she pursued me, my assessment of the Quality [potential] relationship (good, fair, poor), and the Duration of the relationship in months.
What’s crazy is that I find myself applying Data Mining techniques (with my brain instead of a computer) to the data set.
So far, the results I’ve found that are true 100% of the time are
- when I DO NOT PURSUE a woman the relationship is NOT GOOD (meaning it is either poor, or fair) – 5 cases
- when I DO PURSUE a woman AND she PURSUES ME, the relationship is GOOD – 2 cases
- when her RELATIVE AGE is the SAME, the QUALITY is NOT POOR – 10 cases
Conclusion: For the best relationship on this next go-around I should seek a woman who is also seeking me and is the same age as me.
Possible Revisions: Generalize Occupation/Major to “Liberal Arts” and “Math/Science” to allow better grouping. Also, include information pertaining to location, maybe both relative and absolute location.
EDIT: I’ve updated the sheet (but not my findings) and put it on Google Docs. The names have not been published to protect the innocent/guilty.