Over the last decade, fire managers in Australia have embraced the concept of ‘fuel hazard’, and guides for its assessment have been produced. The reliability of these new metrics, however, remains to be determined. This study compared fuel hazard ratings generated by five assessment teams using two Australian hazard assessment methods, in two dry sclerophyll forest sites on Sydney’s urban fringe. Attributes that underpin hazard scores, such as cover and height of various fuel layers, were also assessed. We found significant differences between teams on most variables, including hazard scores. These differences were more apparent when fuel hazard assessments focussed on individual fuel layers than when teams’ assessments were summarised into an overall fuel hazard score. Ratings of surface (litter) fuel hazard were higher when one assessment method was used than when assessors employed the other; however, ratings of elevated (shrub) and bark fuel hazard were relatively consistent across assessment methods. Fuel load estimates based on the two hazard assessment methods differed considerably, with differences between teams also significant. Inconsistency in scoring fuel hazard may lead to discrepancies in a range of management applications, which in turn may affect firefighting safety and effectiveness.