Four models on six public datasets: a majority-class guess, logistic regression, default boosted trees and a 200-fit tuned search. The tuning barely moved the s