Measures of improvement in prediction ×100 when risk models are fit to a training dataset of 50 observations and assessed on a test dataset of 5000 observations where P(D = 1) = 0.5. The linear logistic regression models fit to the training data are (i) baseline logitP(D = 1|X) = α0 + α1X; (ii) model(X, Y1) : logitP(D = 1|X, Y1) = β0 + β1X + β2Y1; (iii) model(X, Y1, Y2) : logitP(D = 1|X, Y1, Y2) = γ0 + γ1X + γ2Y1 + γ3Y2. Data generation is described in Supplementary Materials. Shown are averages over 1000 simulations. Neither Y1 nor Y2 are informative — the true values of β2, γ2, and γ3 are zero.