Definition
Plain language
A statistical way of figuring out how much each factor pushes a result up or down.
As stated in the literature
A model estimating how an outcome varies with one or more predictors; used here to separate domain-specific from baseline penalties in country-endorsement scoring.
Why it matters: It lets you untangle overlapping influences and quantify each one, which is essential for knowing which factor is really driving a result.
For example, a regression can estimate how much a house's price rises for each extra bedroom while separating out the effect of its neighborhood.
Heard on the show
“That gives them a smooth target curve to fit a regression model to.”Episode 008 — Why Long-Horizon AI Agents Get Stuck, and a Milestone-Based Fix That Helps