No one can predict true love

Did you know OK Cupid has a blog? I had forgotten, but that's probably because they just published the first new post in three years. No wonder they sent a blast email to announce! The post details some "experiments" the popular dating site has done on users, concerning the importance of photos versus profile text, match score, etc. Interesting stuff, actually.

" I’m the first to admit it: we might be popular, we might create a lot of great relationships, we might blah blah blah. But OkCupid doesn’t really know what it’s doing. Neither does any other website."

No kidding. Most of us have known this for a while. Well, except for that one guy who, despite me pointing out some glaring differences between us, insisted that "an 87% match is nothing to ignore."

Yes it is - and that blog post proves it.

"The ultimate question at OkCupid is, does this thing even work? By all our internal measures, the “match percentage” we calculate for users is very good at predicting relationships."

I would agree. OKCupid said I was 90% matched with Trooper, and 98% matched with Big. While neither relationship lasted, each was successful in its own way.

OKCupid uses a series of questions to match users. Some are personality questions, some seem like IQ questions. Of course some are sex questions.

If a user doesn't answer any, I still find we are at least a 10% match. I can only assume that represents the fact that we match in some basic ways (he's a guy who dates women, he is the right age, and lives in the right area).

If users do decide to tackle the questions, it isn't enough to respond. To "improve match accuracy" users also need to indicate which answers they consider acceptable, and assign a level of importance to the question. Of course, answers depend an awful lot on each person's interpretation. So while you might not "match" on a particular question, if you took the time to discuss the issue, you might find you actually agree.

Like everything else on dating sites, I've found match scores need to be taken with the proverbial grain of salt. Sites may be good at predicting the success of initial conversations, or how those first dates will go, but no one can predict chemistry or real attraction.

No one can predict true love.