Book Review – “Fooled by Randomness” by Nassim Nicholas Taleb

I picked up Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets, by Nassim Nicholas Taleb, on the recommendation of a mathematician (Vipul Naik, my son). I was expecting a heavy treatise on economics and statistics. It was however a very engrossing book written in a lucid and conversational style, with historical events and everyday situations used freely to provide insights.

Here is the book summary/ key insights (that I picked up from the book):

1)      Human beings are wired in a way that they are unable to intuitively handle randomness and chance.

2)      We are adept at explaining everything through a cause-effect; because we just can’t handle uncertainty. And a statistical correlation does not necessarily mean one causes the other.

3)      Ignoring rare events (outliers) in building prediction models is fooling ourselves – rare events are a part of the process and environment, and their impact is rarely understood or considered by people.

4)      We try to explain extraordinary successes as the result of brilliant strategy or business model or formula or leadership skills or intelligence; while it is often just dumb luck. This is more so of domains like stock trading, marketing, and running a business. We try to learn from and emulate the “winners”, without much success ourselves (by trying to implement the so-called strategies of successful people). Basically, according to the book, many of the winners are just lucky fools :-).

5)      Nice symmetrical probability distributions cannot be expected of any human endeavor (symmetrical distributions may be used to understand controlled situations like gambling – toss of a coin, or rolling of a dice). When we simplify probability distributions and approximate them to neat curves, the results that we get are unreliable.

6)      Though Monte Carlo simulations are looked down upon (“that is cheating, it is not statistics!”) by purists, it is still be the best way to model complex, real situations and understand the potential randomness of the outcomes, and can be used for informed decision making.

7)      Past performance cannot be blindly used to predict future performance. Hence, we should not overestimate the accuracy of our beliefs just because we have been successful in the past, we should reexamine our beliefs based on logic, and always have a backup plan.

One of issues with the book is that it lacks structure and tends to jump from topic to topic. The tone is also snobbish and contemptuous at places, and it may make some people (who are secretly think that their success may be attributable to luck :-)) annoyed or even angry.

The author Nassim Nicholas Taleb is Distinguished Professor of Risk Engineering at New York University’s Polytechnic Institute. He has been a mathematical trader, essayist, philosopher, and researcher. He specializes in understanding uncertainty, luck, probability, knowledge, and decision making. Taleb has been described as a dissident thinker, maverick, irreverent, iconoclastic, and unconventional.

Another book by Taleb in a similar vein is The Black Swan – this is an earlier book, and again very interesting to read. Taleb has also authored AntiFragile, The Bed of Procrustes and Dynamic Hedging.

I recommend this book very highly for anyone involved in high maturity implementation of the CMMI®/ People CMM® models.

For people who are looking for quick-fix templates and control chart macros, this book is not for you (as if high maturity practices can be implemented using quick-fix solutions :-).

Here are some details of the book, in case you want to get your hands on it:

By the way, you DON’T need a Kindle device to read a Kindle ebook.

Fooled By Randomness Book Cover

 

Title: Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets

Author: Nassim Nicholas Taleb

Publishing Date: First Version Around 2001

Publisher: Random House/ Penguin

Available at: Amazon.om, Amazon.in.

Available as eBook in Amazon Kindle.

 

By the way, you DON’T need a Kindle device to read a Kindle ebook.

Please feel free to share your views, experiences or queries, using the “comments” feature available at the top of this article/ post.

Also, please add other insights that you may have got from the book, using the “comments” feature available at the top of this article/ post.

Notes:

Nothing Official About It! – The views presented above are in no manner reflective of the official views of any organization, community, group, institute, or association.


Book Review – “The Checklist Manifesto: How to Get Things Right” by Atul Gawande

I had earlier read two engrossing books by Atul Gawande – Complications and Better. So when I saw The Checklist Manifesto while browsing in the neighborhood library, I decided to pick it up looking forward to an interesting read (some reviewers had strongly recommended it). The title also indicated that it may a good book to review for this weblog.

Well, I was partially right – it was suitable to post a review on this blog. So, here goes…

The main theme of the book/ book summary is as follows:

  • Over the years, many activities have become extremely complex.
  • Even experts struggle to master and remember all the tasks they have to perform.
  • Use of checklists can minimize human errors of oversight. In many cases this it can improve the performance significantly.
  • Use of checklists can also help the experts focus on the difficult, tricky parts of a situation, rather than worry about the mundane activities.
  • There is need to create better checklists, organize them for easy use and ensure that they are used.

The author uses examples from multiple industries and situations. The best ones are from hospitals and medical emergencies (Dr Gawande is a surgeon :-)). There are other examples from the airline industry (where pilots use checklists for normal as well as abnormal situations), construction industry, retail, and restaurants.

There is a whole chapter dedicated to research where the impact of the use of checklists in hospitals was studied. The research showed that there was a significant reduction in deaths (47% reduction) and major complications (36% reduction) for surgical patients. One interesting finding was that though only 80% of the hospital staff found the checklists useful, 93% of them said they would want a checklist to be used if they were themselves getting operated!

The book sometimes extends the concept of “checklist” beyond its normal usage. Here are a few examples of things that are treated under the concept of checklist in the book (though I believe they are different concepts, with their own place in “how to get things right”):

  • Preparing detailed project plans, dependencies, action items, schedules and list of deliverables (example of a building construction project)
  • Use of collaboration meetings (of experts) to handle non-routine situations (e.g., a building floor developing unforeseen problems)
  • Empowerment for doing something extraordinary (how Wal-Mart employees went beyond their formal authority to help people affected by Hurricane Katrina)
  • Use of focus, expertise and wits (how a pilot saved lives by crash landing on Hudson river in 2009 – by focusing on flying the plane, not on using a checklist!)

In trying to bring everything under the umbrella of “checklists”, the author dilutes the concept and utility of checklists as well as other equally important concepts of detailed planning, collaboration, empowerment, dedication, competence and focus. Maybe the title “The Process Manifesto” would have been more apt.

For people who are already convinced about the use of checklists, procedures, plans, collaboration meetings, etc., this book can provide you with interesting examples to relate to process skeptics in your organization. It can also provide process trainers with interesting case studies to relate to the class. You may also consider gifting this book to colleagues who resist the use of formal processes – the book is an easy read and is able to hold the reader’s attention reasonably well.

If you are looking for readymade checklists that will help you reach some level in CMMI®/ People CMM®, then this book is not for you :-).

Those who have read Gawande’s earlier books – Complications and Better may find The Checklist Manifesto a bit disappointing – it is not as engrossing as the earlier two. This is possibly because the earlier books focused primarily on hospitals, medicine and healthcare based scenarios, where Gawande has accumulated loads of experience. And in Checklist, he provides examples from other industries (aircraft manufacture, real estate, retail stores, restaurants, and so on) where he may not have had the same level of familiarity and insight.

Here are some details of the book, in case you want to get your hands on it:

By the way, you DON’T need a Kindle device to read a Kindle ebook.

Book Cover ImageBook Title: The Checklist Manifesto: How to Get Things Right

Author: Atul Gawande

ISBN-10: 1846683130

ISBN-13: 978-1846683138

Publishing Date: Jan 2010

Publisher: Profile Books

Available at: Amazon.com, Amazon.in.

Available as eBook in Amazon Kindle.

By the way, you DON’T need a Kindle device to read a Kindle ebook.


Why Can’t Metrics be Used for Performance Appraisals?

While discussing collection and usage of metrics, one often hears an emphatic “We should not use metrics for individual performance management!”. The statement is made as if it is an unquestionable tenant of the religion called process management.

“And pray, why not?” Why should the performance management process be deprived of metrics? A process oriented organization would definitely not like to boast that their performance management system is completely subjective.

Here are some reasons why metrics should be used for individual performance management.

*    An individual performance management (including the appraisal part) needs to be SMART – the “M” stands for measurable.

*    Most individual performance parameters are the similar to and derived from the project, product and process objectives, they typically relate to cycle time, quality (defects), meeting commitments (schedule) and productivity (cost, effort and usage of resources).

*    A strong metrics system, that provides accurate, precise and valid data can support the project, process and individual performance management requirements.

*    Using the same sources of data, we can create a more aligned organization – the individual objectives are aligned to the project, product and process objectives. In this manner, individuals know that meeting their individual goals helps in meeting the other goals (and vice versa); conflict of interest is minimized.

The situations where we may not want to use process/ project metrics for managing individual performance are:

*    The metrics collection system is not stable, and there questions on the credibility of the data. In such a case, the use of the data for managing the project/ process is also diluted.

*    Usage of the data for individual performance management may make the individuals sabotage the process and the accuracy of the metrics. In which case, we need to strengthen the process and make it sabotage proof.

In the old SW-CMM® days, most metrics collection systems were unstable, and hence many experts of that time were pretty insistent on the metrics not being used for performance appraisals – some organizations even have policy level statements for the same!

We have now moved on from the SW-CMM® days for process management, so we need to move on in other aspects too.

Your comments?


Operational Excellence. Unplugged.