Players with more minutes score more points, but that does not tell you why. Correlation vs causation is the gap where many bad betting ideas begin. This lesson explains what a regression line says, using points = 4 + 0.5 x minutes, and what it does not say about any one player. You see three stories that could explain two numbers moving together, including a confounder like being the team's best player. You also learn about leakage, the time-travel mistake of using information you would not have had before the game, and why honest testing needs a cutoff. Then you run a small prediction by hand, so the idea sticks.
Learning objectives
You will learn what a regression line says and what it does not, why two numbers can move together without one causing the other, and how to spot hidden factors and time-travel mistakes in betting data. You will also run a small prediction by hand.
