Community Comment: Part 3
The comments I provided in reaction to a community discussion thread:
https://www.linkedin.com/feed/update/urn:li:activity:6729270967148810240?commentUrn=urn%3Ali%3Acomment%3A%28activity%3A6729270967148810240%2C6733315058924867584%29
Global Head of Innovative Data Science at Bayer: Yesterday, I published my vlog session, where I also discussed the contraposition between possibility and probability. Today I want to clarify this concept. In ordinary language, a thing is possible when it can happen. It doesn't matter how many times a phenomenon might be observed. In other words, this property can be attributed either using logical or empirical considerations, but it always remains binary. If A is possible, NOT(A) is impossible and vice-versa. In our lives, surprisingly, we seldom rely on this concept. I know that "it's possible" that "it can rain." However, such a statement is unsatisfactory and merely qualitative. The "non-rigorous" concept of probability is the attempt to transform the binary, Aristotelean statement into a fuzzier one. When I say "the probability of rain is high," I'm still in a qualitative domain, but I'm introducing a "degree" that qualifies the level of possibility. A formal probability becomes a measure proportional to the likelihood of observing an event (this is not a precise definition, but it suffices in this context). A quantitative data-driven approach is indeed a strategical and aware shift towards a domain where the uncertainty has the same dignity of binary statements.
Gfesser: Thanks for this post. When this topic arises, I typically simplify my message to clients by saying that everything is "possible", but not everything is "probable", so saying that something is possible doesn't really provide value. For example, is it possible that a given team can complete a series of tasks within a given time period? Yes, it's possible, but perhaps not probable, and as you indicate here, only probability can be assigned a range of values. One of the reasons time boxing can be so effective.. If I were to explain this concept with a graphic, I would simply remove the "impossible" label, and adjust the "sure" label so that it covers the entire span of probability. Individuals far removed from the tasks being discussed tend to think in the realm of their own definitions of possibility rather than probability, so when this term gets brought up by someone it makes sense to explore what they are really trying to say.