Abstract The emerald ash borer ( Agrilus planipennis ) is a destructive invasive beetle that has caused mortality of millions of Fraxinus spp. trees in North America. The extended interval between insect establishment, detection and management has allowed this pest to spread over large parts of North America. Artificial purple canopy traps are currently used in national detection surveys for this beetle. The effectiveness of purple canopy traps at low‐to‐moderate A. planipennis population densities in relation to road proximity was evaluated in 2013 and 2014. Transects of traps were established at set distances from roads in northern Michigan near an isolated A. planipennis infestation. It was hypothesized that trap effectiveness is influenced by road proximity, and that traps placed closer to roads were more likely to detect A. planipennis . A significant relationship was established between the mean number of A. planipennis captured on traps and road proximity (nearness), while no significant relationship was observed between detection success and road proximity. These findings suggest establishing traps further from roads (which can be less economically efficient) provides no greater likelihood of detection than establishing traps on or near the road edge. Basal area of non‐ash and ash species, beetle population density, vigour rating, ash tree species and sampling duration were shown to significantly influence the number of A. planipennis captured on traps.