Sunflower downy mildew is a disease of high global economic impact as well as a causal agent that is extremely difficult to eradicate. During the past decades, several approaches for the determination of Plasmopara halstedii (Ph) races have been used worldwide and are discussed in this review. Procedures of isolation, cultivation and maintenance of Ph isolates, as well as the screening of sunflower for resistance, are also critically reviewed. The predominant, globally used resistance screening protocol is a ‘whole seedling immersion’ inoculation. ‘Soil drench’ inoculation allows more precise control of the number of Ph zoosporangia applied to a single sunflower seedling. A detached leaf assay has been described, but it has been used mainly for Ph subcultivation and fungicide tests. For race determination, a differential set consisting of nine sunflower genotypes has been used since 1988, coupled with a numerical triplet code for virulence phenotyping of Ph. The increasing variability in global Ph populations has demonstrated the inadequacy of the current set of differentials, and several researchers have proposed additional public lines as new differentials. Furthermore, bulk isolates may show different results in repeated tests, as Ph may contain genetically distinct zoospores within a single zoosporangium. For precise race determination, single zoosporangia or single zoospore isolates are advisable. However, due to low success of isolation, approximately 1–2%, this method cannot be applied in routine Ph race screening. Methods surveyed in this review have a broad spectrum of applications, including taxonomic studies.