Introduced Australian brushtail possums (Trichosurus vulpecula) are a major pest in New Zealand. Alternatives to current management by trapping and poisoning are being actively explored. The most effective scenario for the deployment of fertility control vaccines is the use of bait stations to maintain possum populations at low densities following their initial knockdown from high densities using conventional control tools. However, three field trials that estimated population coverage achievable by differing station layouts using dyed baits have produced conflicting results. To assess optimal vaccine-delivery strategies adequately we need meaningful a priori expectations of the level of population coverage achievable by differing bait station layouts and densities. To this end, we collated data on field operations in which bait stations were used to deliver acute toxins for lethal control of possum populations and constructed an individual-based spatially explicit stochastic model to simulate encounter rates between possums and bait stations. The model provided a reasonable fit to the data from both the acute toxin operations and two of the three dyed-bait trials. The third trial likely overestimated typical possum movement distances. Simulating vaccine delivery in the model demonstrated that grid densities of 0.6 bait stations ha^–1 are predicted to be optimal, and prefeeding may not be necessary. The model strongly indicates that the effective delivery of fertility control will not be limited by our ability to deliver vaccine baits to possums. However, the exact strategy to employ will depend on three key characteristics of any vaccine – its expense relative to existing tools, its longevity in the field, and its efficacy at reducing female breeding success.