Risk aversion on television

January 12th, 2006 | by Andrew Ó Baoill |

The Freakonomics blog has had a number of posts about Deal or No Deal, a television show that had passed me by until I came back to Ireland for Christmas. I’ve now seen parts of several episodes of the British version, which I’m guessing is fairly close to the U.S. version.

It’s a fascinating premise, since it’s a mix of luck and calculation of risks and returns. For those unfamiliar with the show, a contestant is faced with 22 boxes. On the inside of the lid of each box is written a different cash amount (in the UK case, from 10p to £250,000). One of the boxes is ‘their’ box, the default prize they will take home. But first they have to open each of the other boxes one by one, thereby narrowing down the range of possible amounts in ‘their’ box. After they open three boxes, or so, they get an offer from a ‘banker’ to buy their box (with the unknown amount) from them for a set amount of cash. They can ‘deal’ – take the fixed sum – or choose ‘no deal’ and continue opening boxes, until the banker offers them another offer.

Stephen Dubner has noted that economists are interested in the show because of the data it provides on the decision-making of individuals (when sitting in a television studio in front of a live audience). Watching the British version I was struck also by the manner in which the contestant I watched relied on ‘lucky’ numbers to choose which boxes to open – no surprise really, but interesting nonetheless. There was also more concentration – in the patter of the presenter, and on the part of the contestant – on the single largest prize than on the spread of the prizes on the board, which struck me as inappropriate (though I may be wrong – I haven’t done much number-crunching on this). The odds being quoted to them for success also seemed wrong to me, seeming to treat all closed boxes as of equal value, rather than noting that the player was playing with one particular box – though, again, this was an off-the-cuff analysis.

Finally, the ‘banker’ (who appears only through phone-calls to the presenter) was being treated by the contestant as an actual person, or at least personality – it seemed she was at least in part treating it as a psychological contest (talking about how the banker wasn’t respecting her, etc.). The Freakonomics posts haven’t noted anything definitive about the formula used for the bids, with various commenters arguing over whether it’s a fixed formula or responds to the perceived personality of the contestant (or other factors such as their existing income) but it seemed to me that early offers were well under the expected return (average of possible results) while later bids became more generous – thus encouraging the contestant to play on near the beginning but bow out rather than play the last game (bearing in mind that the perception of new bids will be coloured by previous bids).

Sorry, comments for this entry are closed at this time.