Fun with pivot tables and match.com

I recently considered signing up at an internet dating site, just to see what was available. Naturally I was quite skeptical (since I am not divorced with kids and over 40). Match.com claims to have the largest number of members of any dating site, so I checked it out first. Fortunately, match.com allows anonymous non-members to do very extensive searching of their catalogue without signing up.

I started off with a basic search: women aged 20-27 within 25 miles of my zip code (this includes most of the Seattle metropolitan area). I drilled down from there based on various factors such as education, income, religion, etc... After I had collected a good amount of data, I dumped it into Excel and analyzed it. Using pivot tables I could see how various factors affected the number of potential matches. For example, here is a plot of the 'Income' and 'Education' factors:

There aren't too many surprises here: more education correlates to a higher average income, but not that much higher. Other plots such as Age vs Income or Religion vs Education didn't reveal too many surprises either.

The real surprises came in the magnitude of the numbers of matches. For my search of single women aged 20-27 in Seattle, there were only 530 results. 530! Clearly there are many more than 530 single girls in Seattle, so match.com's selection seems to really suck.

Out of those 530 results, if I restricted the search to include only matches whose Education and Income buckets matched or exceeded my own, there were only 7 results (though there were 64 if I also included one income bucket below mine).

I know for a fact that there are way more than 7 girls in all of Seattle with comparable income and education levels, so one thing is clear: they aren't using match.com.

I don't think I will either.