Decision Superhero
Decision Superhero: Driving Informed Decision-Making with Probability, Explainable Models, and Decision Science, by Eric Torkia and William K. Klimack
Decision Superhero is a modern practical guide to making better decisions using probability. Understand essential psychological and mathematical models for effective decision-making, implement decision science models in Excel, and build an organization that tests assumptions, discerns variables, and predicts outcomes using decision models.
Topics
One: Origin Story for the Modern Decision Superhero
With so many decisions, maybe it should be a science
A science by any other name…
A modern definition of decision science
Decision science and the art of war
Victory disease
Discovering survivorship bias
A solution when intuition leads you astray
Decision science made the monorail faster
From dietary allowances to optimization
The Traveling Salesman Problem
How the ancients calculated the earth’s circumference
Key takeaways
Two: The Decision Superhero (a.k.a. Scientist)
Meet the decision superhero
Be more right than wrong
Decision superheroes have secret identities
Decision science sales challenges
The Zen mind of the decision superhero
Affirmation #1: A decision scientist is objective
Affirmation #2: I NEED less data than I think, and I HAVE more data than I need (to make a decision)
Affirmation #3: Not all decisions work out, not even the best-made ones
Affirmation #4: Remember that decisions impact others
Affirmation #5: Keep the focus on the prize
Affirmation # 6: Know thy domain
Ethics: The truth and nothing but the truth
Who do you serve?
You are both advocate and devil’s advocate
What is being truthful?
The ethics ACID test
Developing your superpowers
Business superpowers
Analytics superpowers
Quantitative methods in decision science
Key takeaways
Three: SuperPower: Developing a Firm that Thinks Differently
Consultants or develop skills internally?
Decisions start with the end in mind
Opportunity framing
Optimal strategy selection
Resource-based strategy and decision science
Resource optimization
Strategic alignment
Risk management
Enhanced decision-making
Dynamic alignment
The dynamic enterprise alignment model
Managing alignment on three levels
Aligning resources and closing gaps
Stakeholder alignment
Business outcomes of the Dynamic Enterprise Approach
Silos versus integration
Fitting decision science into the organization chart
Centralized organizational structures
Decentralized organizational structures
Driving decision science adoption
Key takeaways
Four: When Data and Decision Superheroes Meet
Decision science is to little data what data science is to big data
The models are the same but what is expected is different
When data science can be built out into a decision model
Inductive versus deductive approaches to solving problems
Decision and data superheroes unite!
Why does a decision superhero need a data superhero?
Data wrangling
Exploratory analysis/opportunity identification
Why does a data superhero need a decision superhero?
Targeting data science for more bang for your buck
Political cover
Key takeaways
Five: Decisions are Models and Models are Decisions
Universal model constructors
Models exist in different forms and places
Mental models define the world we see
Symbolic models
Iconic and analog models
Models represent (and are) systems as well
Models, mental models, and biases
Different business modeling approaches
Systems analysis
Financial modeling
Economic modeling
Project risk analysis (schedules and cost estimates)
Discrete event modeling (supply chain, manufacturing, and other physical systems)
Modeling requires the right attitude
Insight is better than specific answers
Don’t get caught up in the weeds of your own mind
Modeling is an application of the scientific method
What kind of model are you trying to build?
Quantitative versus qualitative models
Descriptive models
Predictive models
Prescriptive models
Deterministic versus stochastic (probabilistic)
Analytical versus numerical
Ready-built versus custom (functions versus programs)
Key takeaways
Six: Selecting the Right Modeling Process
Why selecting the right approach matters
Setting a goal for your analytics model
What are the properties of the decisions we make?
What kind of decisions are you planning to make?
To code or not to code? Isn’t there a tool for that?
One-off/low-risk/low-value decisions ($)
One-off/high-risk/high-value decisions ($$$)
High-frequency/low-risk/low-value decisions ($$)
High-frequency/high-risk/high-value decisions ($$$)
Can a model evolve from one type to the next?
Developing models is an iterative process
Triangulating answers and testing ideas before doing the real work
The economics of code versus spreadsheet
Key takeaways
Seven: The Decision Science Roadmap
Do you want faster or better decisions?
Faster decisions
Better decisions
Setting the analytical stage
Exploratory analysis is to understand
Prediction for decisions
The decision science roadmap
Problem framing process
Data collection and assumption research
Model building process
Analysis and optimization
Production models for ongoing analysis and model lifecycle management
Business analysis using 3-tier design principles
Driving the decision science roadmap buy-in with quick wins
Multiple successes increase confidence in the results
Staffing and resourcing for quick wins
Tools and methods for quick wins
Picking winners and not betting it all on one horse
Managing for success
Key takeaways
Eight: Decision Design
The elemental definition of a decision
Preferences and constraints
Setting the value context of a decision
Who’s making the decision?
Basic elements of a decision model
The objective of the decision
Controllable input variables
Uncontrollable (random) input variables
Courses of action
States of nature
Mathematical and/or visual relationships between inputs and outputs
Decision framing like a champ
Better questions and communication
Framing a decision using a payoff matrix―an example
Framing decisions in an organizational context
Good decisions keep the trade-offs in mind
Working-out decisions
Key takeaways
Nine: Superpower: Understanding Probabilities
Uncertainty, risk, and other important distinctions
The binary life of continuous probability
Kolmogorov’s probability axioms
How to estimate or define the probability of something
Objective probability (using data)
Joint and marginal probabilities
Calculating conditional probabilities
Subjective or opinion-based probability
The mathematics of expected value
Visualizing probabilities and decisions as trees
Key takeaways
Ten: Probability and Decision Theory
The Savage axioms
Thinking in terms of usefulness and preference
The mathematics of utility
Is a 1 in 10 chance a big probability?
Risk seeking
Risk averting
Risk neutral
The principle of certain monetary equivalence
Financial approach to CME (f-CME)
Economic approach to CME (e-CME)
GreenGear Super-Widgets
Decision-making paradigms
Decision-making under certainty
Decision-making under complete ignorance
Decision-making under risk or partial uncertainty
Decision-making under conflict
Decision-making under complete uncertainty
Key takeaways
Eleven: SuperPower: Calculating The Value of Perfect Information
Calculating the impact of better information using simple probabilities
Defining the naïve state (no information scenario)
Updating analysis with probability estimates for weather
Calculating the value of perfect information
Using Bayes theorem for decision-making under risk
Calculating the conditionals
Analyzing payoff scenarios
Analyzing the value of better prediction using Bayes
Key takeaways
Twelve: SuperPower: Making Decisions Under Uncertainty
Defining the initial problem frame
Organizing the decision problem
Identifying alternatives to analyze
Analyzing the decision options
Payoff analysis
Analyzing regrets
Accounting for uncertainties using probability
Calculating the utility of alternatives
Comparing alternatives
Making the decision
Key Takeaways
Thirteen: Remembering the Important Stuff
Eric Torkia and Bill Klimack have masterfully distilled decades of experience into a guide that is both accessible and profoundly insightful. This book is not just a manual for analysts but a blueprint for anyone looking to navigate the complex landscape of modern decision-making.
In a world where data is king, “Decision Superhero” empowers its readers to become the architects of their own destiny, leveraging the principles of decision science to drive impactful outcomes. Whether you’re a seasoned professional or just starting your journey, this book will equip you with the tools and mindset needed to make better, faster, and more informed decisions.
With its blend of theory, practical application, and real-world examples, “Decision Superhero” stands out as an essential read for anyone looking to elevate their decision-making prowess. Eric and Bill have created a resource that is sure to become a staple on the bookshelf of every data-driven leader. This is the book that will turn good analysts into great decision-makers.
Jordan Goldmeier, Author of Becoming a Data Head and Data Smart
Decision Superhero is a compelling guide to mastering decision science, crafted with clarity and depth. This book invites readers to explore the art and science of making better choices, blending real-world examples, historical anecdotes, and practical insights to create a rich learning experience. From operations research during WWII to contemporary decision analysis, the authors present a cohesive narrative that demonstrates the evolution of this critical field. The early chapters offer fascinating stories of decision-making challenges, such as Abraham Wald’s work on survivorship bias, grounding readers in the timeless relevance of decision science.
Douglas W. Hubbard, Author of How to Measure Anything: Finding the Value of Intangibles in Business and The Failure of Risk Management: Why It’s Broken and How to Fix It
Unafraid to blend insights from famous statisticians and Arnold Schwarzenegger in the same breath, Torkia and Klimack provide spirited advocacy for the notion that our mental models should often be replaced or augmented by more explicit, conscious modeling and careful analysis; that this can lead to surprising results and drive organizational value; and that it can often be accomplished with fewer prerequisites than we might be inclined to assume.
Peter Cotton, Author of Microprediction: Buildiung an Open AI Network and, Co-Founder and Chief Scientific Officer at Crunch Labs
Decision Superhero by Eric Torkia and Bill Klimack is a standout guide in the field of decision science, penned by a pioneer whose expertise illuminates every chapter. They masterfully bridge the gap between complex theoretical concepts and practical business applications, making decision science accessible to data professionals of varying technical skill levels. Through historical anecdotes and real-world examples, they illustrate the impactful role of decision science across various industries and its pairing with data science to optimize business strategies.
George Mount, Author of Advancing into Analytics and Modern Data Analytics in Excel
Torkia and Klimack have successfully tied together high-level theories to real life anecdotes, making the concepts not only practical but actionable. The content being light-hearted and effervescent, making a topic that is often somewhat daunting for some, extremely accessible.
Sandra Abi-Rashed, VP Business Growth & Development at Digilant and Founder of Mentoro
About Eric and Bill
Eric Torkia holds a Master’s degree in Management Information Systems and brings a wealth of experience in supporting strategic decision-making across various industries. As the executive partner at Technology Partnerz Ltd., he leads a team dedicated to delivering advanced decision science solutions focused on forecasting, simulation, and optimization. Eric advises senior leadership on business and analytics strategies, has trained hundreds of business analysts in predictive analytics, and spearheads risk analysis initiatives in finance and operations. He is also the lead developer of MCHammer.jl, a Julia package that streamlines Monte Carlo simulations. Deeply involved in the fields of decision science and business analytics, Eric balances his professional pursuits with the tranquility provided by his cats, affectionately known as his “corner assistants.”<br><Br>
William K. Klimack has extensive experience applying operations research in energy and pharmaceutical companies. A retired U.S. Army infantry Colonel, he held faculty positions at West Point and the University of Oklahoma. He has a BS in chemical engineering from Lehigh University, a MS in applied mathematics from Johns Hopkins University, a MMAS from the US Army Command and Staff College, and a PhD in operations research from the US Air Force Institute of Technology. He is a Certified System Engineering Professional, a Professional Engineer, a Certified Analytics Professional, and a Fellow of the Society of Decision Professionals, the National Speleological Society, and the Explorers Club.
Bestsellers
Faculty may request complimentary digital desk copies
Please complete all fields.