Lessons in Risk – Foundations

This series focuses on the foundations of risk assessment—what risk scores represent, how they are constructed, and how they should (and should not) be interpreted.

Before getting into tools, applications, or policy decisions, the goal here is to establish a clear understanding of the core concepts that everything else depends on.


What this series covers

  • Scores as probabilities
  • How scores are built
  • Same score ≠ same person
  • Separation (accuracy)
  • What AUC measures
  • Why AUC is not enough
  • Calibration (predicted vs. observed)
  • Base rates
  • Cut points in practice

Modules

Start Here

Why Risk Scores Are Misunderstood

An introduction to a basic but often overlooked issue: not all risk scores mean the same thing, even when they appear similar.

Module 1

What a Risk Score Represents

What risk scores actually mean—and how to interpret them in practice.

Module 2

How Risk Scores Are Built

How risk assessment tools are constructed.

Module 3

Same Score, Different Person

Different pathways can lead to the same score.

Module 4

What "Accuracy" Really Means

Accuracy is about how well a tool separates outcomes; not just how often it's "right."

Module 5

What AUC Actually Measures

AUC measures how often a tool correctly ranks people; not how well it estimates risk.