As we know when considering risk and risk management that risk is comprised of both the consequence of an event and the likelihood (or probability) that that event will actually occur.  And while most people can accurately predict the potential consequence that a risk event may have, like breaking an arm falling off a bicycle, most people have a difficult time predicting the likelihood of that event (and consequence) actually happening.  There are a lot of reasons for this inaccuracy, most of which are unconscious, but some of which are completely intentional.  

Some of the problems associated with accuratley predicting the probability of an event is that our estimation is influenced heavily by our own experiences.  For example, if you have experienced the event and consequence personally (either yourself or someone close to you) you will tend to believe that the probability or likelihood of that event happening is higher.  Conversely, if you have no experience with the event you tend to downplay the likelihood that it could happen.  Additionally, your perception of the level of control that you have over the event, or your personal experience doing the task successfull or unsuccessfully also heavily influences your estimation of the likelihood that something bad could happen.  For example, people who SCUBA dive recereationally have a unrealistic expectation regarding the safety of industrial commercial diving operations.  In general, these people think SCUBA diving is safer than it actually is (i.e. from a risk management perspective).

Finally, in some organizations the level of authority needed to proceed with a task or operation is tied to the risk assessment completed on the task.  The higher the risk of the operation, the higher in the organization one needs to go to get the authorization to proceed.  As a result, some people and some organizations tend to artificially discount and reduce the overall risk (through a conscious reduction in probability) to obtain a lower level authority to proceed with the task.  As sometimes probability is “tougher” to define than the consequence there appears to be leeway, or judgement, in the assignment of the probability.  Additionally, this process often overlooks the ability to research the probability of the event more closely and instead only relies on local, and often, limited knowledge of the probability of that event actually occuring.  The advent of the internet and the access to greater levels knowledge helps to discover what a better assignment of probability could be or should be for the task or operation in question.

Regardless, to get a meaningful estijmation fothe risk of a event, one must spend time to get boththe asessment of the potential consequence and the the probability correct.  Otherwise the overall risk assessment will be skewed.

Probability of a fatality in commerical / industrial diving:  6X10-4 to 1.3X10-3

Probability of a fatal car accident (Canada): 5X10-5 to 7X10-5

Probablity of being struck by lightning:  1X10-6