Risk and Uncertainty, what’s the chance that you’re wrong?

Risk and Uncertainty, what’s the chance that you’re wrong?

So here you are driving in peak-hour traffic making your way to a friend’s party.

As of ten minutes ago your biggest concern was finding enough quarters to pay the guy at the toll booth. Unfortunately his flamboyant tie reminded you of wrapping paper and the birthday present you forgot to pick up for George.

You notice a lifeline:

a gas station stocking a range of questionably authentic music.

Provided you come up with a convincing back storey for why you thought George would ‘really get into’ Willie Nelson, you think you’ll be okay.

However, when looking for a place to park, you only find one with a sign ‘Unauthorised vehicles will be towed’.

You quickly consider the alternatives.

You can park in the spot and risk being towed. Or turn up to the party without a present risking a night of awkward excuses…

Although you consider the likelihood of being towed as low, the costs are potentially high.

While the costs of not bringing something for George aren’t as high, the likelihood of you being able to avoid looking bad is low.

 

This is of course the typical context in which the term ‘risk’ is used. However with risk also comes uncertainty, which is in fact a rather separate idea.

Specifically something may be risky when the final outcome of your decision is not certain but on average predictable.

An example of something which can be defined in the region of risk might be whether a coin when flipped lands on heads or tails. Although you cannot be certain which side it will land on in any particular instance, we know that on average it will land on either side about half of the time.

Conversely if the final outcome of your actions is dependent on something which is wholly not on average predictable it may be more accurately described as uncertainty.

For instance guessing what an unlabelled tin can contains is closer to what statisticians might describe as uncertainty, insofar as there is no information available to tell you whether it is in fact food, motor oil or spam.

In the parking scenario we could think of risk as being everything we can predict on average (such as a parking inspector appearing) and uncertainty as everything we cannot (such as George’s reaction to the Willie Nelson CD).

Interestingly enough whether you risk getting a present for George might depend less on what the real risks of your car being towed and more on how well your day went.

In fact new research has illustrated just how subjective our risk assessments tend to be, suggesting we often make choices which are far from rational.

For instance research has found that often people are more willing to bet on an outcome being in their favour before an event than after it.

The implication is that people are more optimistic about their ability to predict an outcome before the event has occurred than after the event. Despite the outcome being unknown in both scenarios.

 

Feeling lucky?

Experiments have also suggested that whether we bet on an outcome, such as flipping a coin, can depend on whether or not we have recently experienced success in the recent past. We may convince ourselves that the chance of successfully choosing an outcome is higher than it actually is as we feel we are on a ‘winning streak’.

Experiments have also shown that when faced with a choice between two levels of risk they will tend to choose the lower of the two, even when the expected pay-off from the riskier option is higher.

An example of this might be giving ten people a choice between $100 or a fifty per cent chance of receiving $1000.

Clearly when considering such an alternative the group would receive more in total by choosing the riskier alternative. However, for a range of reasons people tend to choose the safer, but less rewarding option.

So when it comes to choosing whether or not to park in an area that risks your car being towed it’s best to remind yourself that whatever choice you make might be impacted more by what you had for breakfast than the risks of turning up empty handed (or with a Willie Nelson CD) to the party.

Giles

I'm an economist, data geek and public speaker.

Leave a Reply

Time limit is exhausted. Please reload CAPTCHA.