Effectors encoded by avirulence genes (Avr) interact with the Phytophthora sojae resistance gene (Rps) products to generate incompatible interactions. The virulence profile of P. sojae is rapidly evolving as a result of the large‐scale deployment of Rps genes in soybean. For a successful exploitation of Rps genes, it is recommended that soybean growers use cultivars containing the Rps genes corresponding to Avr genes present in P. sojae populations present in their fields. Determination of the virulence profile of P. sojae isolates is critical for the selection of soybean cultivars. High‐resolution melting curve (HRM) analysis is a powerful tool, first applied in medicine, for detecting mutations with potential applications in different biological fields. Here, we report the development of an HRM protocol, as an original approach to discriminate effectors, to differentiate P. sojae haplotypes for six Avr genes. An HRM assay was performed on 24 P. sojae isolates with different haplotypes collected from soybean fields across Canada. The results clearly confirmed that the HRM assay discriminated different virulence genotypes. Moreover, the HRM assay was able to differentiate multiple haplotypes representing small allelic variations. HRM‐based prediction was validated by phenotyping assays. This HRM assay provides a unique, cost‐effective and efficient tool to predict virulence pathotypes associated with six different Avr (1b, 1c, 1d, 1k, 3a and 6) genes from P. sojae, which can be applied in the deployment of appropriate Rps genes in soybean fields.