Decision Trees and Uncertainty

One of the most generic of decision situations arises when we are torn between sticking with something we know and trying something new. This might be staying with a tried and true process as over against innovation, it could be ordering something new at a favorite restaurant, it could be trying some new practice vs. sticking with custom, or it could be whether or not to make a personnel change.

As always, we follow the CCSPP paradigm:

  1. What are the Choices?
  2. What are the Chances?
  3. How are they Sequenced?
  4. What are the Payoffs?
  5. What are the Probabilities?


Here we have (1) stick with the tried and true, OR (2) try something new


Something new might turn out better than the status quo or it could turn out worse. We note that if even a poor outcome with innovation is better than the status quo, then we don't really have a problem, do we?


First we make a choice and then we see how things pan out if we go for innovation.


Let's label them G = good outcome from innovation, B = disappointing outcome from innovation, S = status quo or same ole same ole


In the absence of any other information, let's say we have zero information as to whether the innovation is likely to succeed or disappoint. Thus we set the probabilities at 0.5 and 0.5.