Analytics Stories: Using Data to Make Good Things Happen

Cover
Title Page
Copyright Page
About the Author
About the Technical Editor
Acknowledgments
Contents at a Glance
Contents
Introduction
Part I What Happened?
Chapter 1 Preliminaries
Basic Concepts in Data Analysis
What Is a Random Variable?
Excel Calculations
Chapter 2 Was the 1969 Draft Lottery Fair?
The Data
The Analysis
Excel Calculations
Chapter 3 Who Won the 2000 Election: Bush or Gore?
Projecting the Undervotes
What Happened with the Overvotes?
The Butterfly Did It!
Excel Calculations
Chapter 4 Was Liverpool Over Barcelona the Greatest Upset in  Sports History?
How Should We Rank Upsets?
Leicester Wins the 2015–2016 Premier League
#16 Seed UMBC Beats #1 Seed Virginia
The Jets Win Super Bowl III
Other Big Upsets
Chapter 5 How Did Bernie Madoff Keep His Fund Going?
The Mathematics of Ponzi Schemes
Madoff’s Purported Strategy
The Sharpe Ratio Proves Madoff Was a Fraud
Benford’s Law and Madoff’s Fraud
Excel Calculations
Chapter 6 Is the Lot of the American Worker Improving?
Is U.S. Family Income Skewed?
Median Income and Politics
Causes of Increasing U.S. Income Inequality
Money Isn’t Everything: The Human Development Index
Create Your Own Ranking of Well-Being
Are Other Countries Catching Up to the U.S.?
Excel Calculations
Chapter 7 Measuring Income Inequality with the Gini, Palm,  and Atkinson Indices
The Gini Index
The Palma Index
The Atkinson Index
Excel Calculations
Chapter 8 Modeling Relationships Between Two Variables
Examples of Relationships Between Two Variables
Finding the Best-Fitting (Least Squares) Line
Computing the Beta of a Stock
What Is a Good R2?
Correlation and R2
We Are Not Living in a Linear World
Excel Calculations
Chapter 9 Intergenerational Mobility
Absolute Intergenerational Mobility
Intergenerational Elasticity
Rank-Rank Mobility
Comparing IGE and Rank-Rank Mobility
Measuring Mobility with Quintiles
The Great Gatsby Curve
Excel Calculations
Chapter 10 Is Anderson Elementary School a Bad School?
How Can We Adjust for Family Income?
Estimating the Least Squares Line
Can We Compare Standardized Test Performance for Students in Different States?
Excel Calculations
Chapter 11 Value-Added Assessments of Teacher Effectiveness
Simple Gain Score Assessment
Covariate Adjustment Assessment
Layered Assessment Model
Cross-Classified Constant Growth Assessment
Problems with VAA
How Much Is a Good Teacher Worth?
Excel Calculations
Chapter 12 Berkeley, Buses, Cars, and Planes
Simpson’s Paradox and College Admissions
The Waiting Time Paradox
When Is the Average of 40 and 80 Not 60?
Why Pre COVID Were There Never Empty Seats on My Flight?
Excel Calculations
Chapter 13 Is Carmelo Anthony a Hall of Famer?
What Metric Defines Basketball Ability?
Wins Above Replacement Player (WARP)
Manu, Melo, Dirk, and Dwayne
How Do 25,000 Points Lead to So Few Wins?
Chapter 14 Was Derek Jeter a Great Fielder?
Fielding Statistics: The First Hundred Years
Range Factor
The Fielding Bible: A Great Leap Forward
The Next Frontier
Chapter 15 “Drive for Show and Putt for Dough?”
Strokes Gained
The Myth Exposed
Chapter 16 What’s Wrong with the NFL QB Rating?
NFL Quarterback Rating
ESPN’s Total Quarterback Rating
Excel Calculations
Chapter 17 Some Sports Have All the Luck
Skill vs. Luck: The Key Idea
The Results
Chapter 18 Gerrymandering
A Stylized Example
The Mathematics of Gerrymandering
Chapter 19 Evidence-Based Medicine
James Lind and Scurvy: The Birth of  Evidence-Based Medicine
The Randomized Streptomycin Tuberculosis Trial
Excel Calculations
Hormone Replacement: Good or Bad?
Chapter 20 How Do We Compare Hospitals?
Ratings Criteria
Conclusion
Excel Calculations
Chapter 21 What Is the Worst Health Care Problem in My Country?
Disability-Adjusted Life Years
Determination of Disability Weights
To Age Weight or Discount, That Is the Question
Key Facts About World Health
Part II What Will Happen?
Chapter 22 Does a Mutual Fund’s Past Performance Predict Future Performance?
Mutual Fund Basics
Morningstar Ratings
Risk-Adjusting Fund Returns
How Well Do Morningstar Star Ratings Predict a Fund’s Future Performance?
The Effect of Expense Ratio on Long-Term Performance
Excel Calculations
Chapter 23 Is Vegas Good at Picking NFL Games?
How NFL Betting Works
Bias and Accuracy
Vegas Forecasts Are Unbiased
Totals Predictions and Money Line Predictions Are Unbiased
NFL Accuracy: The Line vs. the Computers
A System Works Until It Doesn’t
Chapter 24 Will My New Hires Be Good Employees?
What Data Do We Need to Determine Attributes That Best Predict Employee Performance?
Besides GMA, Not Much Affects Job Performance
Excel Calculations
Chapter 25 Should I Go to State U or Princeton?
Analyzing Princeton vs. Penn State
Excel Calculations
Chapter 26 Will My Favorite Sports Team Be Great Next Year?
Francis Galton and Regression to the Mean
Regression to the Mean in the NFL and the NBA
Excel Calculations
Chapter 27 How Did Central Bankers Fail to Predict the  2008 Recession?
The Inverted Yield Curve
The Sahm Rule: Early Warning Signal for Recession
Control Charts and the Housing Price/Rent Ratio
Excel Calculations
Chapter 28 How Does Target Know If You’re Pregnant?
What Available Data Can Be Used to Identify Pregnant Women?
Problems Arise
An Example of a Pregnancy Prediction Score
Chapter 29 How Does Netflix Recommend Movies and TV Shows?
User-Based Collaborative Filtering
Item-Based Filtering
Chapter 30 Can We Predict Heart Attacks in Real Time?
Posterior Probability
Sensitivity and Specificity
ROC Curve
Back to the Apple Heart Study
AliveCor and KardiaBand
Chapter 31 Is Proactive Policing Effective?
Hot Spots Policing
Predictive Policing
CCTV
Stop and Frisk
Broken Windows
Excel Calculations
Chapter 32 Guess How Many Are Coming to Dinner?
Which Parameters Must Be Estimated?
The Data
The Results
Which Factor Really Matters?
Excel Calculations
Chapter 33 Can Prediction Markets Predict the Future?
Examples of Trade Contracts
Prediction Market Trading Mechanisms
Accuracy of Prediction Markets and Wisdom of Crowds
Chapter 34 The ABCs of Polling
Why Are 1,112 People Enough to Represent U.S. Voters?
Why Doesn’t a Larger Population Require a  Larger Sample Size?
So, What Can Go Wrong?
Rating Polls
Chapter 35 How Did Buzzfeed Make the Dress Go Viral?
Measuring Instagram Engagement
Tweets Do Not Always Go Viral Immediately
Do the First Few Days Predict the Future of a Meme?
Chapter 36 Predicting Game of Thrones TV Ratings
What Does Google Trends Tell Us?
Predicting the Present with Google Trends
Using Google Trends to Forecast GOT Ratings
Excel Calculations
Part III Why Did It Happened?
Chapter 37 Does Smoking Cause Lung Cancer?
Correlation and Causation Redux
The Key Evidence
Could Air Pollution Have Caused Lung Cancer?
The Cigarette Companies Hit Back
Excel Calculations
Chapter 38 Why Are the Houston Rockets a Good Basketball Team?
NBA Shooting Math 101
Zach LaVine Battles the Bulls’ Analytics Department
Conclusion
Excel Calculations
Chapter 39 Why Have Sacrifice Bunts and Intentional Walks Nearly Disappeared?
The Case Against Bunting
Bunting Against the Shift
Why Are Intentional Walks on the Decline?
Chapter 40 Do NFL Teams Pass Too Much and Go for It Often Enough on Fourth Down?
The Ascent of Passing
Fourth Down Strategy
New Data Partially Vindicates the Coaches
Teams Should Go for Two More Often
Chapter 41 What Caused the 1854 London Cholera Outbreak?
Cholera
Snow and the Broad Street Pump
Snow’s Randomized Controlled Trial
Conclusion
Excel Calculations
Chapter 42 What Affects the Sales of a Retail Product?
Painter’s Tape
Estimating the Model Parameters
Excel Calculations
Chapter 43 Why Does the Pareto Principle Explain So Many Things?
Power Laws
Why Do Incomes Follow the Pareto Principle?
Why Do a Few Websites Get Most of the Hits?
Excel Calculations
Chapter 44 Does Where You Grow Up Matter?
Quasi-Experimental Design vs. Randomized Controlled Trials
What Drives Neighborhood Differences in  Upward Mobility?
How Can We Make Things Better?
Chapter 45 The Waiting is the Hardest Part
Which Factors Influence the Performance of a  Queueing System?
Operating Characteristics of a Queueing System
How Does Variability Degrade the Performance of a Queueing System?
Calculating the Operating Characteristics of a  Queueing System
Excel Calculations
Chapter 46 Are Roundabouts a Good Idea?
What Is a Roundabout?
History of Roundabouts
Benefits of Roundabouts
Disadvantages of Roundabouts
Roundabout Capacity
Roundabouts and Revolutions
Chapter 47 Red Light, Green Light, or No Light?
What Causes Traffic Jams?
How Should We Set the Lights?
Ramp Meters and Equity
Measuring the Impact of Ramp Meters
The Twin Cities Metering Holiday
Part IV How Do I Make Good Things Happen?
Chapter 48 How Can We Improve K–12 Education?
Tennessee’s STAR Study on K–2 Class Size
Cost–Benefit Analysis
Can Predictive Analytics Increase Enrollment and  Performance in Eighth-Grade Algebra I?
Excel Calculations
Chapter 49 Can A/B Testing Improve My Website’s Performance?
Improving Obama’s Fundraising in 2008
The Mechanics of Resampling
Excel Calculations
Chapter 50 How Should I Allocate My Retirement Portfolio?
The Basic Portfolio Optimization Model
The Efficient Frontier
Difficulties in Implementing the Markowitz Model
Excel Calculations
Chapter 51 How Do Hedge Funds Work?
Growth in Hedge Funds and Hedge Fund Fee Structure
Shorting a Stock
Long/Short and Market-Neutral Strategies
Convertible Arbitrage
Merger Arbitrage
Global Macro Strategy
Hedge Fund Performance
The George Costanza Portfolio
Excel Calculations
Chapter 52 How Much Should We Order and When Should We Order?
The Economic Order Quantity Model
Reorder Points, Service Levels, and Safety Stock
Excel Calculations
Chapter 53 How Does the UPS Driver Know the Order to Deliver Packages?
Why Is the Traveling Salesperson Problem So Hard?
Solving the Traveling Salesperson Problem
The Traveling Salesperson Problem in the Real World
Excel Calculations
Chapter 54 Can Data Win a Presidential Election?
Democratic Presidential Analytics
The GOP Strikes Back
Cambridge Analytica and the 2016 Election
Excel Calculations
Chapter 55 Can Analytics Save Our Republic?
Arrow’s Impossibility Theorem
It’s Not Easy to Pick a Winner!
Ranked-Choice Voting
Approval Voting
Quadratic Voting
Excel Calculations
Chapter 56 Why Do I Pay Too Much on eBay?
How Many Pennies in the Jar?
The Importance of Asymmetric Information
The Winner’s Curse and Offshore Oil Leases
Sports Free Agents and the Winner’s Curse
Can You Avoid the Winner’s Curse?
Excel Calculations
Chapter 57 Can Analytics Recognize, Predict, or Write a Hit Song?
How Does Shazam Know What Song You Are Listening To?
How Did Hit Song Science Know Norah Jones’s Album Would Be a Smash?
Can Artificial Intelligence Write a Good Song?
Chapter 58 Can an Algorithm Improve Parole Decisions?
An Example of Risk Scores
ProPublica Criticizes Risk Scores
Skeem and Lowenkamp and PCRA
Machine Learning and Parole Decisions
Chapter 59 How Do Baseball Teams Decide Where to Shift Fielders?
The Debut of the Shift
The Return of the Shift
Empirical Evidence on the Shift
Why Not Just Beat the Shift?
Excel Calculations
Chapter 60 Did Analytics Help the Mavericks Win the 2011 NBA Title?
How Can You Evaluate a Basketball Player?
From Player Ratings to Lineup Ratings
Chapter 61 Who Gets the House in the Hamptons?
The Basic Idea
What Asset Division Is Best?
Excel Calculations
Index
EULA