jats:titleAjats:scbstract</jats:sc> </jats:title>jats:pDue to the conformal factor problem, the definition of the Euclidean gravitational path integral requires a non-trivial choice of contour. The present work examines a generalization of a recently proposed rule-of-thumb [1] for selecting this contour at quadratic order about a saddle. The original proposal depended on the choice of an indefinite-signature metric on the space of perturbations, which was taken to be a DeWitt metric with parameter jats:italicα</jats:italic> = – 1. This choice was made to match previous results, but was otherwise admittedly jats:italicad hoc</jats:italic>. To begin to investigate the physics associated with the choice of such a metric, we now explore contours defined using analogous prescriptions for jats:italicα</jats:italic> ≠ – 1. We study such contours for Euclidean gravity linearized about AdS-Schwarzschild black holes in reflecting cavities with thermal (canonical ensemble) boundary conditions, and we compare path-integral stability of the associated saddles with thermodynamic stability of the classical spacetimes. While the contour generally depends on the choice of DeWitt parameter jats:italicα</jats:italic>, the precise agreement between these two notions of stability found at jats:italicα</jats:italic> = – 1 continues to hold over the finite interval (– 2, – 2/jats:italicd</jats:italic>), where jats:italicd</jats:italic> is the dimension of the bulk spacetime. This agreement manifestly fails for jats:italicα</jats:italic> > – 2/jats:italicd</jats:italic> when the DeWitt metric becomes positive definite. However, we also find dramatic failures for jats:italicα</jats:italic> < – 2 that correlate with breakdowns of the de Donder-like gauge condition defined by jats:italicα</jats:italic>, and at which the relevant fluctuation operator fails to be diagonalizable. This provides criteria that may be useful in predicting metrics on the space of perturbations that give physically-useful contours in more general settings. Along the way, we also identify an interesting error in [1], though we show this error to be harmless.</jats:p>