Contents
Foreword
Preface
Acknowledgments
Introduction
Foundations
Part 1: The Big Picture
Chapter 1: The Flaw of Averages
Give Me a Number
Statisticians to the Rescue?
Red Lobster
Red River
Red Ink in Orange County
The Red Coats
Why Forecasts Are Always Wrong: A Problem of Dilbertian Proportion
Chapter 2: The Fall of the Algebraic Curtain and Rise of the Flaw of Averages
Becoming a Management Scientist
Abandoning Management Science
Being Reborn as a Management Scientist
On the Road Again
From OR to PR and Back
Chapter 3: Mitigating the Flaw of Averages
Probability Management
Doing for Probability What Edison Did for Electricity
The Consolidated Risk Statement
Chapter 4: The Wright Brothers Versus the Wrong Brothers
A Wind Tunnel for Your Business Plan
Chapter 5: The Most Important Instrument in the Cockpit
Learning About Learning
Flying on Instruments: When Gut Instinct Isn’t Enough
Part 2: Five Basic MINDles for Uncertainty
Chapter 6: MINDles Are to MINDs What HANDles Are to HANDs
Steam Era Statistics: Things to Forget
Chapter 7: Mindle 1: Uncertainty Versus Risk
Risk Is in the Eye of the Beholder
Chapter 8: Mindle 2: An Uncertain Number Is a Shape
New Brunswick Hold ’Em
Uncertain Numbers: The Shape of Things to Come
The Weak Form of the Flaw of Averages
Give Me a Distribution
Black Swans
Some More Stuff on Distributions
Chapter 9: Mindle 3: Combinations of Uncertain Numbers
A Portfolio of Spinners
So What?
Diversification: A Green Word for the CENTRAL LIMIT THEOREM
You Don’t Need to Know
The Power of Diversification
Poster Child for the CENTRAL LIMIT THEOREM
Chapter 10: I Come to Bury SIGMA, Not to Praise it
Outputs of Simulations
Definition of SIGMA
Chapter 11: Mindle 4: Terri Dial and the Drunk in the Road
The Strong Form Versus the Weak Form
It’s Too Hard to Change the Way People Do Business
Tugging at the Distribution
Evaluation of a Gas Well
The Cyclone Hits London
The Human Cyclone Meets the Financial Tsunami
Chapter 12: Who Was Jensen and Why Wasn’t He Equal?
Spreadsheets with Uncertain Inputs
Options, Restrictions, Smiles, and Frowns
Chapter 13: Mindle 5: Interrelated Uncertainties
COVARIANCE and CORRELATION are Red Words
Three Idealized Investments
Portfolios
Three Idealized Portfolios
A Mindle for Interrelated Uncertainties: The Scatter Plot
CORRELATE This!
Part 3: Decisions and Information
Chapter 14: Decision Trees
Decision Trees
Giving Back
Chapter 15: The Value of Information
The Complement of Uncertainty
What’s It Worth to You?
A Military Example
The Value of Obfuscation
Prototyping
Part 4: The Seven Deadly Sins of Averaging
Chapter 16: The Seven Deadly Sins of Averaging
The List
Chapter 17: The Flaw of Extremes
Sandbagging
Layers of Fat
Howard Wainer’s Most Dangerous Equation
Chapter 18: Simpson’s Paradox
Kidney Stones
Baseball
Chapter 19: The Scholtes Revenue Fallacy
Economics 101
Chapter 20: Taking Credit for Chance Occurrences
Advertising Results
Pharmaceutical Effectiveness
An Age-Old Scam
Applications
Part 5: The Flaw of Averages in Finance
Chapter 21: Your Retirement Portfolio
Simulation for the Masses
Smoothing Things Out
Chapter 22: The Birth of Portfolio Theory: The Age of Covariance
Risk Is the New Dimension
Chapter 23: When Harry Met Bill(y)
A Simplified Model of the Relationships Among Securities
All Models Are Wrong
Chapter 24: Mindles for the Financial Planning Client
Bessemer Trust
Interactive Simulation at Bessemer
Financial Engines
Chapter 25: Options: Profiting from Uncertainty
Stock Options
Implied Volatility, the Uncertainty Gauge
Chapter 26: When Fischer and Myron Met Bob: Option Theory
Risk-Neutral Pricing
Publication
Long-Term Capital Management: The Perfect Storm
Skin in the Game
Chapter 27: Prices, Probabilities, and Predictions
Futures and Probabilities
Options and Uncertainty
Prediction Markets
Predicting Terror Attacks
DARPA Gets into and out of the Picture
Predictomania
When Does This Stuff Work?
Part 6: Real Finance
Chapter 28: Holistic Versus Hole-istic
A Serendipitous Sequence of Events
We Win Some and Lose Some
Our Ideas See the Light of Day
The Decision Forest
Chapter 29: Real Portfolios at Shell
The Planets Were Aligned
The Model Survives a Hurricane and Management Changes
Chapter 30: Real Options
Keeping Sailplanes Aloft
Case Study: Restoring a ’64 Porsche
Supplying Thrust in Podunk
Meanwhile Back at Wells Fargo
Chapter 31: Some Gratuitous Inflammatory Remarks on the Accounting Industry
FASB and the Weak Form of the Flaw of Averages
FASB and the Strong Form of the Flaw of Averages
Some Constructive Criticism for the Accounting Industry
Part 7: The Flaw of Averages in Supply Chains
Chapter 32: The DNA of Supply Chains
Just Plain Nuts
When Less Is More
The Bullwhip Effect
The Beer Game
Chapter 33: A Supply Chain of DNA
A Supply Chain at Genentech
Forecasting Done Right
An Interactive Dashboard
Chapter 34: Cawlfield’s Principle
From Remedial Math to Chemical Engineer
A Problem at Olin
The General Principle
Part 8: The Flaw of Averages and Some Hot Button Issues
Chapter 35: The Statistical Research Group of World War II
German Tanks
Protective Armor on Bombers
Half a Battleship
Chapter 36: Probability and the War on Terror
The Magic Bullet
The False Positive Problem
There Is Nothing to Cheer but Fear Itself
Rumsfeld Asks the Right Question
Chapter 37: The Flaw of Averages and Climate Change
Why the Earth’s Average Temperature Is Falling
The Bad News
The Good News
Free Markets and the Tragedy of the Commons
Learning the Controls
Chapter 38: The Flaw of Averages in Health Care
Treating the Average Patient
Medicine Meets Mathematics
Unintended Consequences
I’m Not Dead
Chapter 39: Sex and the Central Limit Theorem
The Results of Diversifying Your Portfolio of X Chromosomes
Which Tail Are You Looking At?
The Myth of the Myth of the Math Gender Gap
On Rethinking Eugenics
Probability Management
Part 9: Toward a Cure for the Flaw of Averages
Chapter 40: The End of Statistics as You Were Taught It
Galton’s Dice
The Bad Boys of Statistics
Chapter 41: Visualization
Visual Statistics
Toward a Theory of Irrational Expectation
Chapter 42: Interactive Simulation: A New Lightbulb
The Return of a Pioneer
Polymorphic
Tip of the Iceberg
Chapter 43: Scenario Libraries: The Power Grid
Modular Risk Models and Age of the Scatter Plot
The Subprime Mortgage Fiasco
Chapter 44: The Fundamental Identity of SLURP Algebra
Chapter 45: Putting It into Practice
The DIST Distribution String
Certification and Auditing
Software Implementation
The Outlook
Chapter 46: The CPO: Managing Probability Management
The CPO Versus the CRO
In the Land of the Averages, the Man with the Wrong Distribution Is King
Probability Management at Merck & Co.
Calibrating Your CPO
To Blow or Not to Blow the Whistle, That Is the Question
Chapter 47: A Posthumous Visit by My Father
Red Word Glossary
About the Author
Index
Copyright © 2009, 2012 by Sam L. Savage. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 646-8600, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permissions.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. For more information about Wiley products, visit our web site at www.wiley.com.
Library of Congress Cataloging-in-Publication Data:
Savage, Sam L., 1944–
The flaw of averages : why we underestimate risk in the face of uncertainty / Sam L. Savage.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-471-38197-6 (cloth); ISBN 978-1-118-07375-9 (paper); ISBN 978-1-118-37357-6 (ebk); ISBN 978-1-118-37358-3 (ebk); ISBN 978-1-118-37352-1 (ebk)
1. Risk. 2. Uncertainty. I. Title.
HB615.S313 2009
338.5—dc22
2008052144
For Daryl, Jacob, and Russell, and in memory of Fruipit, our beloved dog
Foreword
Enterprise analysis under uncertainty has long been an academic ideal. In 1954 Sam Savage’s father Leonard “Jimmie” Savage published The Foundations of Statistics, which addressed this subject through an elegant and highly mathematical framework for rational decision making. It had a profound effect on the fields of statistics and economics in the following decades. Sam, whose training is in computer science, has definitely not followed in his father’s footsteps. Although deeply influenced as a child by Jimmie and his academic colleagues, he also spent time as an automobile mechanic, which taught him the value of tactile learning. Sam’s approach to analyzing uncertainty developed relatively late in life, through running simulations on computers. In his book and tightly coupled Web site, Sam covers the knowledge required for grasping the fundamentals of uncertainty and risk, not with mathematical formulas, but with everyday items such as game board spinners and dice. He defines the mindle (pronounced ) as a handle for the mind, and provides five clear mindles for the concepts of risk attitude, uncertain numbers, diversification, plans based on uncertain inputs, and interrelated uncertainties.
He then goes on to show that when we use single numbers to estimate uncertain future outcomes that we are not just usually wrong, but are consistently wrong. He provides numerous examples of what he calls the Flaw of Averages, in which plans based on average assumptions are wrong on average. This is summarized in the Seven Deadly Sins of Averaging, in which it is apparent just how widespread these problems are in today’s society.
In the same informal style he goes on to describe the foundations of modern finance, the first field to formally conquer such problems. He then argues that the concepts of modern finance can be generalized to problems in healthcare, climate change, supply chains, the war on terror, and several other hot-button issues.
The last section of the book is devoted to what he sees as a potential cure for the Flaw of Averages. He recommends that the analysis take the form of a Monte Carlo simulation with which senior executives can interact, so that the user can draw on stochastic libraries that are available throughout the firm, that sublibraries can be combined and substituted, and that the results of any simulation include its pedigree, that is, the audit trail of the assumptions that went into it. While helping large firms better model their portfolios of risky projects, Savage developed a new data type, the Distribution String, which represents a major breakthrough in the communication of risk and uncertainty. It significantly widens the practical applicability of sound theory in these fields. This book describes these advances, which help transform the ideals of enterprise analysis under uncertainty into a practical reality.
Harry Markowitz
Nobel Laureate in Economics
Preface
The Flaw of Averages describes a set of common avoidable mistakes in assessing risk in the face of uncertainty. It helps explain why conventional methods of gauging the future are so wrong so often, and is an accessory to the recent economic catastrophe. Once grasped, these ideas can lead us to more effective forecasting and decision making. Traditionally, these topics have been the domain of probability and statistics. Although I will assume no prior knowledge of these subjects, for those who have had formal training in statistics, it should take only a few chapters to repair the damage.
My perspective no doubt derives largely from my father, Leonard Jimmie Savage. Although well below average on academic scales during his early education, he emerged as a prominent mathematical statistician who collaborated closely with Milton Friedman, among others. One of their students was the founder of modern portfolio theory, Harry Markowitz, who claims that my father “indoctrinated him at point blank range with rational expectation theory.” Thus I am a child of the University of Chicago School of Economics.
Early on it was clear I possessed at least one of my father’s traits. I, too, was a below-average student, displaying neither athletic nor academic aptitude. The defining moment of my high school education came in an after-class conference with my English teacher in my junior year at the University of Chicago Laboratory School. She explained that I was failing the course, but with a monumental effort might achieve a D by the second semester. Then she helpfully explained the underlying problem: The Lab School was for students who would go on to college, which, quite clearly, I would not. Instead she suggested a technical school where I could get practical training as a mechanic or a plumber.
She therefore presented me with my first serious career decision: to work my butt off for a lousy D in English or play my guitar for immediate gratification. I made the obvious choice, and music has brought me joy and solace ever since. Better yet, I was able to have my cake and eat it too, because I ended up with a D anyway. In retrospect I cannot question this individual teacher’s judgment, because all told I received three D’s in four years of high school English, each under a different instructor.
My father was in no position to complain, because when he graduated from high school in Detroit, he too “was classified by his teachers as ‘not college material,’ and consequently was refused admission to the University of Michigan.”1 My grandfather, in desperation, called on personal connections to get him admitted on probation to Wayne State University. Allen Wallis, with whom my father later cofounded the University of Chicago Statistics Department, reported what happened next: “In his year [at Wayne] he established a good enough record to be admitted on probation to the University of Michigan. However, he caused a fire in a chemistry laboratory and was expelled.” 2
Once again I followed in my father’s footsteps, later flunking out of the University of Michigan myself, although based on academics rather than involuntary arson.
As adolescent misfits, then, neither of us was able to conform to the norms expected by our teachers. Thus nonaverageness itself became a family value, perhaps in some way inspiring this book. After being de-Michiganized, however, our careers diverged. My father fought his way back into Michigan, got his PhD in mathematics, and achieved great academic acclaim. I worked as a mechanic and briefly raced a sports car before ultimately getting a degree in computer science, which is, in deference to my English teacher, just plumbing with bits of information instead of water.
Although The Flaw of Averages will discuss concepts from both statistics and economics, I have little formal training in either of these subjects—just the basics picked up at an early age at the dinner table. Therefore, I have written not from the perspective of a statistician or an economist, but from the perspective of a former mechanic and current plumber of information who grew up surrounded by statisticians and economists.
I came up with some of the core ideas and title for this book in 1999, and I started writing. I knew the concept had potential, but somehow the book was not uplifting: The Flaw of Averages asserts that everything is below projection, behind schedule, and beyond budget. Where was the happy ending?
In search of one, I continued to teach, consult, and write articles about various aspects of this problem. Feeling the need to stake out the real estate (in case I ever did finish the book), I wrote an article in October 2000 on the Flaw of Averages for the San Jose Mercury News.3 When it was published, it was, to my surprise, adorned with a drawing by the renowned cartoonist Jeff Danziger depicting a statistician drowning in a river that is on average three feet deep. This is reproduced in Chapter 1 of this book.
Over the years, I have had the good fortune to interact with some exceptional people in academia and industry who were grappling with the Flaw of Averages themselves. As a result of this interaction, an approach that we call Probability Management has recently emerged, offering a potential cure for many variants of this problem. And so at last with a happy ending in hand, I renewed my writing efforts in earnest in 2006. All told, on average, I have written 21 words per day since 1999.
When my stepbrother, John Pearce, first heard of this writing project, he assumed that I was working through some sort of psychodrama involving my late father. Wrong. This work has been fueled by a psychodrama involving my high school English teachers.
Sam L. Savage
Palo Alto, California
April 2009
Notes
1. A. F. M. Smith, The Writings of Leonard Jimmie Savage—A Memorial Collection (Washington, DC: American Statistical Association and The Institute of Mathematical Statistics, 1981), p. 29.
2. Ibid., p. 14.
3. Sam L. Savage, “The Flaw of Averages,” Soapbox column, San Jose Mercury News, October 8, 2000.
Acknowledgments
I must start by chronologically acknowledging those who were directly involved in the evolution of Probability Management. I am indebted to Ben Ball of MIT, first for infecting me with his interest in portfolios of petroleum exploration projects in the late 1980s, and second for the collaboration that laid the foundations for much that lay ahead. In 1992 Mark Broadie of Columbia University gave me a key (a simple spreadsheet model) that unlocked a world of stochastic modeling. In 2003 I had the pleasure of working with Andy Parker of Bessemer Trust on a retirement planning model that pioneered some important ideas in interactive simulation. In 2004, I began an exciting three-way collaboration with Stefan Scholtes of Cambridge University and Daniel Zweidler, then at Shell. This truly put Probability Management on the map with a large interactive simulation application at Shell and a coauthored article in ORMS Today. During this time, Dan Fylstra of Frontline Systems made a breakthrough in interactive simulation, turning my dream of interactive simulation in spreadsheets into reality.
The following group also played critical roles in the development of this book. My father, Leonard Jimmie Savage, and his colleagues Milton Friedman and Allen Wallis served as towering intellectual role models from my earliest memories. Next, I must thank Linus Schrage of the University of Chicago for his collaboration on What’s Best!, without which I would not have been reborn as a management scientist. By supporting my seminar series on management science in spreadsheets, Jack Gould, then dean of the University of Chicago Graduate School of Business, helped launch the odyssey during which I discovered the Flaw of Averages. Stanford’s Department of Management Science and Engineering, with which I have been affiliated since 1990, has been the ideal environment in which to experiment with and teach the ideas underlying the book. I owe special thanks to Peter Bernstein, whose own book, Capital Ideas, assisted me in my own work and who personally helped get this book off the ground. In 1999, Mina Samuels, who was then an editor for John Wiley & Sons, was inspirational in helping me conceive the book and, when I tracked her down in 2007, was even more supportive as a midwife. In the meantime, Bill Falloon, who inherited my nine-year project at John Wiley, deserves the Most Patient Editor of the Century Award: Thanks. Bill Perry of Stanford University has served as both an inspiration and a foundation of support. Marc Van Allen, of the law firm Jenner and Block, realized that the Flaw of Averages underlies the nation’s accounting standards and collaborated in researching and publicizing the issue. Several chapters were inspired by discussions with Howard Wainer, and by a prepublication draft of his book, Picturing the Uncertain World: How to Understand, Communicate and Control Uncertainty Through Graphical Display, which I highly recommend. Finally, I owe special thanks to David Empey and Ronald Roth for their programming support over the years and in particular for the implementation of the application at Shell and subsequent development of the DIST (Distribution String) data type.
When it takes you nine years to write a book, there is plenty of time to pick up useful ideas from others. So many people provided assistance, contributions, or comments over the years that they won’t fit into a paragraph. Therefore I have used the following table. The laws of probability ensure that I have missed a few people who belong here, for which I apologize in advance.
Dick Abraham
Bob Ameo
Ted Anderson
Matthias Bichsel
Adam Borison
Jerry Brashear
Stewart Buckingham
Mike Campbell
David Cawlfield
Kevin Chang
Terri Dial
Mike Dubis
Ken Dueker
David Eddy
Brad Efron
Martin Farncombe
Roland Frenk
Chris Geczy
Bob Glick
Peter Glynn
Joe Grundfest
Deborah Gordon
Kevin Hankins
Ward Hanson
Warren Hausman
Wynship Hillier
Gloria Hom
Ron Howard
John Howell
Doug Hubbard
Darren Johnson
Martin Keane
Gary Klein
Michael Kubica
Paul Kucik
Andrew Levitch
Bob Loew
David Luenberger
Jeff Magill
Harry Markowitz
John Marquis
Michael May
Rick Medress
Robert Merton
Mike Naylor
Abby Ocean
Greg Parnell
John Pearce
Mark Permann
Bill Perry
Tyson Pyles
Matthew Raphaelson
Andrew Reynolds
John Rivlin
Aaron Rosenberg
The late Rick Rosenthal
Mark Rubino
Sanjay Saigal
John Sall
Jim Scanlan
Karl Schmedders
Myron Scholes
Michael Schrage
Randy Schultz
Adam Seiver
William Sharpe
Rob Shearer
John Sterman
Stephen Stigler
Jeff Strnad
Steve Tani
Janet Tavakoli
John Taylor
Carol Weaver
Bill Wecker
Roman Weil
Justin Wolfers
A separate category of appreciation goes to those who contributed to the specifications of the DIST data type, in particular Dave Empey, Dan Fylstra, Harry Markowitz, Ivo Nenov, John Rivlin, Ron Roth, John Sall, Stefan Scholtes, Eric Wainwright, and Whitney Winston.
Special thanks to Aishwarya Vasudevan for her help with the graphics, Debbie Asakawa for her suggestions on the entire manuscript, and Jeff Danziger for his drawings.
In the end, I could not possibly have written this without the guiding light of my wife Daryl, who helped extensively with the editing and who continues to make life so much fun.
S. L. S.
INTRODUCTION
Connecting the Seat of the Intellect to the Seat of the Pants
The only certainty is that nothing is certain.
—Pliny the Elder, Roman scholar, 23–79 CE
As the financial meltdown of 2008 has demonstrated, Pliny is still pretty much on target two millennia later. Despite all its promise, the Information Age is fraught with a dizzying array of technological, economic, and political uncertainties. But on the flip side, the Information Age also offers electronic extensions of our intuition that can provide a new experiential feel for risk and uncertainty. This book shows how.
Let’s start off with a simple everyday example in which most people’s intellects fail. Imagine that you and your spouse have an invitation to a ritzy reception with a bunch of VIPs. You must leave home by 6 p.m. or risk being late. Although you work in different parts of town, each of your average commute times is 30 minutes. So if you both depart work at 5:30, then you should have at least a 50/50 chance of leaving home together for the reception by 6 o’clock.
This thinking sounds right. But your instinct warns that you will probably be late. Which is correct: your brain or your gut?
Your gut is correct, but not being particularly good with words, it may have difficulty winning the argument intellectually. So here, in terms that even a brain can understand, is why you’ll probably be late.
Suppose there really is a 50/50 chance that each of you will make it home by 6:00. Then the trip is like a coin toss in which heads is equivalent to arriving by 6:00 and tails to arriving after 6:00. Four things can happen:
- Heads/tails: You are home by 6:00 but your spouse isn’t.
- Tails/heads: Your spouse is home by 6:00 but you aren’t.
- Tails/tails: Neither of you is home by 6:00.
- Heads/heads: Both of you are home by 6:00.
The only way you can leave by 6:00 is if you flip two heads, for which there is only one chance in four.
Now imagine that your brother, who also works 30 minutes away, is going to join you. The chance of your all leaving on time now drops to one in eight. Or suppose you, your spouse, and five friends and relations all plan to pile into your minivan for the trip to the reception. Assuming that everyone leaves work at 5:30 and has a different 30-minute route to your house, then the chance of leaving on time is the same as flipping 7 heads in a row; that’s 0.5 raised to the 7th power, or 1 in 128.
No wonder people are always late!
If you want to teach yourself to get a better grasp on uncertainty and risk, you have to recognize two very different types of learning: intellectual and experiential. To set the stage, let’s start with something that everyone has understood since childhood. It may be expressed as follows:
Actually, these are the differential equations of the motion of a bicycle. You have solved them for most of your life, not through the seat of your intellect, but experientially, through the seat of your pants.
The theory of probability and statistics can likewise be presented in terms of mind-numbing equations, and that’s the way it’s usually taught in school. This is probably why Nobel Prize–winning research in behavioral economics has shown that even people trained in the field consistently make mistakes when faced with day-to-day uncertainties.1, 2
Steve Jobs, cofounder of Apple, said that “personal computers are bicycles for the mind.” Increasingly, through a process called simulation, they are being applied to problems involving uncertainty and risk, allowing us to bypass the equations of the traditional statistics course and to gain an experiential understanding of the subject. In the past few years, I have had the good fortune of collaborating with colleagues in academia and industry in advancing the development of such techniques and applying them in practice. I call our approach Probability Management, and it has been applied to problems as wide-ranging as assessing risks in retirement portfolios, investing in petroleum exploration projects, and designing incentive programs for bankers. It has been an exhilarating and sometimes exhausting ride, and it is far from over.
The book has three sections. The first, Foundations, provides a basis for visualizing risk and uncertainty using simple everyday props such as gameboard spinners and dice. It describes the sorts of consistent errors that occur when uncertain numbers are replaced by single “average” values: the Flaw of Averages. The second part, Applications, describes classic cases of the Flaw of Averages in the real world. The third part describes a potential path toward a cure for the Flaw of Averages: Probability Management.
Foundations
The foundations are intended to help you intuitively grasp and visualize the consequences of uncertainty and risk. If you were learning to ride a bicycle, for example, the foundations phase would end as soon as you no longer required training wheels.
Hey, wait a minute. Didn’t I just argue that you can’t learn to ride a bike from a book? Yes, I did. Paradoxically, I will attempt to do what I claim is impossible. Here’s how. At various stages along the way, you will see this bicycle in the margin.
At that point you will have the option to put the book down and visit FlawOfAverages.com, where you will be able to go for an actual ride. There are plenty of animations, simulations, and other experiential demonstrations to improve your intuition concerning these issues.
Applications
I begin the second section of the book with applications of the concepts of Part 1 to the field of finance, where the Flaw of Averages was first conquered in managing the risk of and return on investments. Although these models are being recalibrated and refined in the light of the economic turmoil of 2008–2009, they provide an excellent foundation for managing uncertainty and risk. Furthermore, they have the potential to be generalized to many other areas of industry and government that are still blind to the Flaw of Averages. I will discuss examples in supply chain management, project portfolios, national defense, health care, climate change, and even sex.
Probability Management
The book concludes with a discussion of the field of Probability Management, an approach toward a general cure for the Flaw of Averages, which is based on recent breakthroughs in technology, coupled to new data structures and management protocols. The approach is being adopted by some large companies today, and your organization can do it too.
In his book Blink,3 Malcolm Gladwell describes the power of snap judgments as thinking “without thinking.” He writes that “just as we can teach ourselves to think logically and deliberately, we can also teach ourselves to make better snap judgments,” a process I refer to as connecting the seat of the intellect to the seat of the pants. The goal of this book is to help you make better judgments involving uncertainty and risk, both when you have the leisure to deliberate and, more importantly, when you don’t.
A Note from Your Author
I now interrupt this book to bring you an important announcement. Some of the material is a bit mathematical and may challenge certain readers. Accordingly, to accommodate a wide variety of technical backgrounds, I will occasionally offer opportunities to jump ahead without missing the main thrust of the argument. What you choose to do at these forks in the road will depend on your aptitude, previous knowledge of the subject, and how badly you want to get to the juicy chapters on investments, the war on terror, and sex.
Notes
1. “Daniel Kahnemann: The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2002,” Nobelprize.org, http://nobelprize.org/nobel_prizes/economics/laureates/2002/kahneman-lecture.html
2. Daniel Kahneman, Amos Tversky, and Paul Slovic (Eds.), Judgment Under Uncertainty: Heuristics and Biases (New York: Cambridge University Press, 1982).
3. Malcom Gladwell, Blink—The Power of Thinking Without Thinking (Boston: Little Brown and Company, 2005).
FOUNDATIONS
The foundations are intended to help you intuitively grasp and visualize the consequences of uncertainty and risk. If you were learning to ride a bicycle, for example, the foundations phase would end as soon as you no longer required training wheels.
PART 1
THE BIG PICTURE
In Part 1, I will provide an overview of the Flaw of Averages, how it rose to prominence, and how technology and new business practices have the potential to provide a cure. I will finish with some general thoughts on the use and benefit of analytical management models.