Continued dependence on stocking of hatchery-reared lake trout, Salvelinus namaycush, in the Laurentian Great Lakes indicates recruitment bottlenecks still exist between spawning and the juvenile life stage (age-1). Possible impediments include poor habitat quality and an inability of hatchery-origin adults to select substrates that support successful incubation. Understanding of characteristics of suitable lake trout spawning habitat is limited but has centered on the cobble-contour framework, which identifies sloped reefs with cobble substrate and deep interstices to entrain eggs as spawning habitat. A novel sampling design based on acoustic-telemetry detections was used to evaluate relative importance of cobble-contour habitat characteristics (substrate size, homogeneity, interstitial depth, and slope) for predicting egg presence on two spawning reefs in northern Lake Huron. Fixed effect logistic regression models were used to relate egg presence to all possible combinations of measured physical characteristics. Physical characteristics of sites with naturally spawned eggs present varied, and no single best model predicted egg presence. Model selection and multi-model inference revealed slope to be the only informative predictor. Substrate size, homogeneity, and interstitial depth were not important. Embryo survival-to-hatch was also compared among sampling sites using in situ habitat bioassays seeded with hatchery fertilized embryos. No difference in embryo survival was observed between sites where naturally spawned eggs were present vs. absent. Over half of sites evaluated exceeded 75% survival. Naturally spawned eggs were observed at 25 of 57 sites. We suspect other characteristics such as interstitial flow could be helpful for predicting selection of lake trout spawning habitats.