Models
Companion Models to Published Articles and Presentations
The best way to learn about probability management is to explore interactively.
NOTE: to experiment with the models below, you will need to enable content, if asked, and make sure that Excel is in Automatic Calculation mode.
Decision Support for Security Projects in a VUCA Environment
These two models support Eng-Wee Ethan Yeo’s presentation at Risk Awareness Week 2020.
The first model uses a stochastic probability tree to calculate the expected value of the project across all risk scenarios in the main decision model.
The Flaw of Averages in Flattening the Curve
Blog Post by Sam Savage
Slowing the rate of contagion of an epidemic, or flattening the curve, is economically costly by keeping people at home. However, if it prevents limited critical care capacity from becoming overwhelmed, it may save thousands of lives.
A Combination of Coronavirus and the Flaw of Averages Can Drive You Nuts
Blog Post by Sam Savage
Datasaurus Arithmetic
Based on the Datasaurus data set of Alberto Cairo, which was based in turn on DrawMyData by Robert Grant. This model shows that SIPmath lets you do arithmetic with the most bizarre distributions imaginable.
Download Model (.xlsx)
None of My Successes Have Been Planned and None of My Plans Have Been Successful: Simulating Rags to Riches and Vice Versa
Blog Post by Sam Savage
Three Italian physicists created a simulation that shows how chance drives the disparity in the distributions of talent and wealth. Inspired by the physicists, this SIPmath model explores similar principles. The model shows that chance plays a role, but that disparity in income can arise without it.
The Sum of the Sandbags Doesn't Equal the Sandbag of the Sum: How Probability Management Helps Solve Age-Old Problems in Budgeting and Forecasting
Article: How to Profit from Probability Management by Matthew Raphaelson and Sam Savage, in Banking Strategies, Bankers Administration Institute
SandbagCalc.xlsx
Demonstrates basic sandbag math
BAI_Model.xlsx
Banking example with revenues and expenses
Rolling Up Operational Risk
Rolling up operational risk at PG&E by Jordan Alen, Christine Cowsert Chapman, Melissa Kirmse, Farshad Miraftab, & Sam L. Savage
From OR/MS Today, December 2016, Volume 43, Number 6.
Asset Level Model (.xlsx)
Consolidated Risk (.xlsx)
Operation Eagle Claw
Operation Eagle Claw was a failed special operations mission designed to rescue 52 Americans who had been taken hostage by Iranian revolutionaries in Tehran in November 1979. The Naval Postgraduate School uses it a case study in Flaw-of-Average based decision modeling. This is described in PowerPoint deck, and Helicopter.xlsx, which lets you experiment with the reliability and number of aircraft sent, then instantly simulates 10,000 missions to determine the chances of completion.
Eagle Claw Simulation (.xlsx)
SIPmath™ Flood Model
By Sam Savage
This is an active SIPmath™ simulation of the example in Figures 1.5 and 37.3 of Dr. Savage's book, The Flaw of Averages: Why we Underestimate Risk in the Face of Uncertainty. It demonstrates that the average damage from flooding is greater than the damage associated with the average crest.
Model (.xlsx)
Probability Management in Financial Planning
By Sam Savage and Shayne Kavanagh, Government Finance Review Magazine, February 2014.
These SIPmath™ models demonstrate the examples of scheduling, surplus and deficit, and reserves. The Dice calculator is available below.
Models:
Schedule (.xlsx)
Surplus Deficit (.xlsx)
Reserves (.xlsx)
Files from Dr. Savage's Stanford Webinar
The three SIPmath™ models included here demonstrate why so many things are Behind Schedule, Beyond Budget, and Below Projection. This is explained by the Flaw of Averages, a systematic set of errors that occur when single "average" estimates are substituted for uncertainties.
Behind Schedule (.xlsx)
Below Projection (.xlsx)
Beyond Budget (.xlsx)
SIPmath Dice Calculator
Here is a SIPmath™ calculator that simulates 10,000 rolls of a pair of dice before your finger leaves the key. Experiment with the formula in D10 to see the distributions of outcomes of different calculations based on the dice.
Model (.xlsx)
Further examples
The models below accompanied the original articles on Probability Management in OR/MS Today. They were created before the SIPmath standard, and were tedious to create, requiring thousands of formulas.
Probability Management, Sam Savage, Stefan Scholtes & Daniel Zweidler, OR/MS Today, Part 2, February, 2006
Probability Management, Sam Savage, Stefan Scholtes & Daniel Zweidler, OR/MS Today, Part 1, February, 2006
IMPORTANT: You need to enable macros to access the clickable scatter plot. Depending on the version of Excel, after unzipping the file, you may need to right click on it, go to Properties then click Unblock.
