Saturday, September 25, 2010

Revisiting Reality TV Game Theory

Last week the New York Times sponsored Freakonomics blog was gracious enough to quote part of my Bachelor Pad Game Theory post. Not only did I appreciate the extra couple hundred hits, I got some great comments from the loyal Freakonomics readers. One of them mentioned a similar situation in a British reality TV show called Golden Balls. Here's how it works:

I mentioned in my original post that I thought the glances exchanged by the Bachelor Pad contestants were important for their ability to trust each other. In the video above, and in other examples online, it seems a lot of exchange between contestants doesn't improve the likelihood of sharing. In fact in a study done on the British game show, contestants were not more likely to cooperate if their opponent was expected to cooperate. Overall, 45% of contestants in the study cooperated and apparently higher stakes slightly decreased the chances of cooperation, unless the contestant's occupation requires a trustworthy reputation.


  1. Whoa! The freakin' NYT blog? That's awesome! Congratulations, friend!

  2. With my limited knowledge of human psyche....I think that if the contestants are are partners or in first relationship within the family, then there is higher probability of chosing 'split' rather than steal'.

    Do we have any research paper on it ??

  3. I couldn't find any research on whether family makes a difference, but my bet is that the result would be similar to that of "contestant's occupation requires a trustworthy reputation".

  4. Oh and thanks Justin, it was pretty exciting.

  5. Congrats Harrison.

    Remember to take time for the little people Harrison.


You are the reason why I do not write privately. I would love to hear your thoughts, whether you agree or not.