The current antibiotic resistance crisis is causing a resurgence of interest in phage therapy (PT). However, bacteria carry a huge diversity of anti-phage defence systems (DSs) that may cause treatment failure. It is currently impossible to predict phage resistance/infectivity patterns based on the DS content of a strain, because DS phenotypes have been characterised under different culture conditions, diverse experimental models and non-representative phage panels. To address this issue, we produced a strain of Pseudomonas aeruginosa that lacks known defence systems, into which we introduced individually ~30 of the most common DS. We then performed tens of thousands of infection assays using a functionally diverse and representative panel of ~60 lytic and ~40 temperate phage. Using this “common garden” experimental design, we were able to quantify the relative contribution of individual DS to phage resistance phenotypes, both in terms of resistance, strength and range. Generally, DSs that act against phage DNA provide broader resistance ranges than late-acting abortive infection systems, which specifically act against certain phage clades. By varying DS expression with inducible promoters, we confirmed the robustness of our findings and found that higher expression levels amplify both the fitness costs of DS carriage and, in some cases, the magnitude or breadth of phage resistance. Our high-throughput data on the genotype-phenotype relationship of DS and phage-resistance provides a step-change in our ability to predict resistance patterns of P. aeruginosa clinical isolates from genomic data alone, with the overall goal of advancing PT development.