Abstract: Heavy mineral identification and chemical analysis are two commonly used methods for calculating the geological grades of coastal metallic placer deposits. However, significant discrepancies often exist between the results obtained from the two methods, leading to exploration risks and hindering the development and utilization of critical metal resources such as zirconium and titanium in China's coastal and shallow marine areas. Using the surface and core samples collected in the Liaodong Bay offshore area, we conducted heavy mineral identification, chemical analysis, and grain-size analysis, from which the zirconium grades were calculated using the two methods and the results were compared. Results show that the zirconium grades calculated by the heavy mineral identification had a narrower range and lower average values but a higher coefficient of variation than those of chemical analysis. Additionally, the spatial distribution correlation of high-grade points between the two methods was weak. We believed that the discrepancies were due mainly to the differences in sample selection, preconditions and assumptions, and application conditions. Specifically, the heavy mineral identification is limited by grain-size ranges, calculation errors due to different presumptions, and increased grade variability caused by parameter interactions. On the other hand, the chemical analysis method faces several other problems, namely, the unavailability of fine-grained zircon, incorrect or missing analysis parameters, and high background values and errors caused by the presence of zirconium in light minerals. To enhance the reliability of the conversion results between the two methods, we recommend selecting testing and calculation methods based on specific objectives, combining multiple analytical approaches to verify the results, and introducing new technologies and improved models to optimize the accuracy and efficiency of coastal metallic placer deposit exploration.