Went to see Horton Hears a Who with the kids, who had a blast.  Afterwards we talked about things that you can’t see, can’t hear, can’t smell and yet exist (despite the kangaroo’s insistence to the contrary).  This was great fun with the conversation ranging from bacteria to magnetic fields.  But we did not touch on something even more profound simply because it is too abstract for the kids (and as I have come to realize, for most adults too): probability.

We live in a world in which randomness plays a huge factor.  I should know, I met my wife in 1991 in a cafe in Paris in a completely chance encounter.  Yet we humans have no way to observe probability directly.  The best we can do is observe outcomes.  This fundamental limitation leads to a lot of mistakes both in planning and in judging outcomes.

A great example of this are financial models for startups companies.  I have seen a lot of these over the years and almost all of them (including some that I have been guilty of putting together) make the same mistake.  They include a ton of detail but do not account at all for randomness.  Sure there is usually a scenario switch to try out different sets of assumptions (pessimistic, base, optimistic), but for a given set of assumptions the model is completely deterministic.

Went to see Horton Hears a Who with the kids, who had a blast.  Afterwards we talked about things that you can’t see, can’t hear, can’t smell and yet exist (despite the kangaroo’s insistence to the contrary).  This was great fun with the conversation ranging from bacteria to magnetic fields.  But we did not touch on something even more profound simply because it is too abstract for the kids (and as I have come to realize, for most adults too): probability.

We live in a world in which randomness plays a huge factor.  I should know, I met my wife in 1991 in a cafe in Paris in a completely chance encounter.  Yet we humans have no way to observe probability directly.  The best we can do is observe outcomes.  This fundamental limitation leads to a lot of mistakes both in planning and in judging outcomes.

A great example of this are financial models for startups companies.  I have seen a lot of these over the years and almost all of them (including some that I have been guilty of putting together) make the same mistake.  They include a ton of detail but do not account at all for randomness.  Sure there is usually a scenario switch to try out different sets of assumptions (pessimistic, base, optimistic), but for a given set of assumptions the model is completely deterministic.