Clubroot (Plasmodiophora brassicae) is an important disease of canola (Brassica napus) and other brassica crops. Accurate estimation of inoculum load in soil is important for evaluating producer risk in planting a susceptible crop, but also for evaluation of management practices such as crop rotation. This study compared five molecular techniques for estimating P. brassicae resting spores in soil: quantitative polymerase chain reaction (qPCR), competitive positive internal control PCR (CPIC‐PCR), propidium monoazide PCR (PMA‐PCR), droplet digital PCR (ddPCR) and loop‐mediated isothermal DNA amplification (LAMP). For ddPCR and LAMP, calibrations were developed using spiked soil samples. The comparison was carried out using soil samples collected from a long‐term rotation study at Normandin, Québec, with replicated plots representing 0‐, 1‐, 2‐, 3‐, 5‐ and 6‐year breaks following susceptible canola infested with clubroot. CPIC‐PCR and ddPCR provided repeatable estimates of resting spore numbers in soil compared with estimates from qPCR or LAMP alone. CPIC‐PCR provided the most robust measurement of spore concentration, especially in the 2 years following a crop of susceptible canola, because it corrected for effects of PCR inhibitors. PMA‐PCR demonstrated that a large proportion of the DNA of P. brassicae detected in soil after the susceptible canola crop was derived from spores that were immature or otherwise not viable. Each assay provided a similar pattern of spore concentration in soil, which supported the conclusion of a previous study at this site that resting spore numbers declined rapidly in the first 2 years after a susceptible crop, but much more slowly subsequently.