Confidence Interval

A range of values that likely contains the true parameter being estimated. A 90% confidence interval means that in repeated sampling, 90% of such intervals would contain the true value.

Like a GPS showing your location within a circle of uncertainty — the circle gets smaller with better signal (more data).

Why it matters

It quantifies uncertainty in a projection — knowing a player's projected points is useful, but knowing the range (e.g., 18-26) tells you how much to trust it.

How DMP uses this

DMP's variance estimates effectively create confidence bands around projections, which feed into P(over) calculations — wider bands mean less certainty about clearing a line.

Common mistake

Interpreting a 95% confidence interval as "95% chance the true value is in this range" — it's about the method's long-run reliability, not a single prediction.