How Much Analysis of Outcomes is too Much Analysis?

How Much Analysis of Outcomes is too Much Analysis?

Diligence promptly executes what intelligence slowly excogitates. Hurry is the failing of fools; they know not the crucial point and set to work without preparation. On the other hand, the wise more often fail from procrastination; foresight begets deliberation, and remiss action often nullifies prompt judgment. Celerity is the mother of good fortune. He has done much who leaves nothing over till to-morrow. Festina lente is a royal motto. (Balthasar Gracian - The Art of Wordly Wisdom)

I'm trying to make a habit of reading 30 aphorisms each day from the book of wisdom, The Art of Wordly Wisdom by Balthasar Gracian. I did something similar last year when I read The Daily Stoic by Ryan Holiday, I haven't read it daily, what I've done instead was to read a bunch of 'dailies' when I needed Stoic advice and wisdom the most.

From time to time it happens to come across advice that in one source seems contrary to advice given in other source. In this aphorism Balthasar Gracian tells about the importance of diligence over hastiness. On the other hand, too much rationalization and analysis of outcomes can lead to procrastination.

What-If Analysis What-If Analysis. Source:Flickr

The wise usually have more life experience, a better view and understanding of the world, can see the problem from more angles and in more depth. They can also see more solutions. Like chess players, they have a better understanding of the board and can think several moves ahead. Chess is a simple game with simple rules, a limited number of chess pieces yet after the first opening moves it gets very complicated and complex. There are 72.048 possible positions after 2 moves, 9+ million possible positions after 3 moves and 288+ billion possible positions after 4 moves and the number of possibilities keeps growing from there. We can agree that the world is even more complex than the game of chess and the number of possibilities of events happening is even greater.

Given this complexity, there is no wonder that having a great number of possibilities and choices can lead to paralysis by analysis. In philosophy we have the illustration of this paralysis by analysis in the paradox of Buridan's Donkey : a donkey that is equally thirsty and hungry is placed midway between a stack oh hay and a pail of water, unable to chose between the two options the donkey eventually dies.

How much preparation and diligence is too much? Is it really necessary to predict how the chess board will look like after five moves or is it sufficient to see only two-three moves ahead? In real life and in e-commerce, the answer is clear, the fewer options you show your customers the better the conversion rate will be, two-three options will lower their choice anxiety and perform better than 5 options. But this is only a small subset of the total set of possible choices that we have to do during our lifetime.

I did a search for "how to reduce the number of choices", the search engine didn't do that autocomplete thing that it does for popular search terms, which leads me to believe that not a lot of people are interested in this. The results that are returned are unsatisfactory, there is no theory, or platform or tool or anything that is useful in regards to reducing the number of choices. Choices belong to a wide array of domains and they are virtually infinite so there is no wonder that there is some universal way (that I know of) of reducing them.

However, a choice filtering system that I know of is the Bayesian Filter. The Bayesian filter is an algorithm that calculates and estimates the probability of a variable. That variable can be anything from a location, for example the next most probable location of a robot or a missile, a file that is probable to be a virus or not, an email that is probable to be spam or not. Many email clients are using this filter to filter out spam. The filter gives an estimated probability ranging from 0% to 100% (0 to 1), for example an estimation of 10% spam will mark the email as not spam, an estimation of 90% spam will mark the email as spam. It's a good system in classifying variables in two ways, and as we have seen, the fewer choices the better and when these choices can be reduced to two is even better because they can be reduced even further using a Bayesian filter: 0 or 1? spam or not? good or bad? legit or not? chocolate or vanilla?

Bayesian filters are pretty good at doing certain things but their use is limited. Humans are complex machines and their decisions are influenced by a lot of factors, most of the time they are not even aware of them, and humans don't usually think in terms of probabilities of events.

How much thinking and analysis of outcomes is too much? How much fast decision making and action taking is too fast? The balance between diligence and celerity is worth studying in more depth and this is exactly what I intend to do in future posts.

Update: Part 2 can be seen here