Partial extractions have been employed and promoted as geochemical exploration tools, but assessment of performance is infrequently undertaken. Minimum probability statistics were used to quantitatively compare twelve extractions and document the level of exploration precision and accuracy. These methods were valuable in comparing the different digestion methods, identifying the best performance level and determining an appropriate geochemical threshold for future exploration. Eight ‘conventional’ reagents were tested, namely ‘four-acid’, aqua regia, bacterial leach (LocatOre®), Genalysis® proprietary leach (TL3), cold hydroxylamine hydrochloride, Mehlich I reagent, hydrogen peroxide, deionized water, together with four new digestible digests (Coke®, Pepsi®, Diet Coke® and a Tempranillo red wine). These tests involved analysis of thirty <63-μm soil samples from the Wartook region of Victoria, Australia, which exhibit Cu and Zn geochemical anomalies in six samples. All extractions successfully identified the anomalous zone; however, the digestible digests, Mehlich I, Genalysis® leach and the hydroxylamine hydrochloride leach performed the best in terms of both accuracy and precision. The new extractions were particularly effective for Cu, whereas the hydroxylamine hydrochloride leach was best for Zn. The Coke®, Pepsi® and Diet Coke® extractions exhibited some buffering effects. In contrast, the Tempranillo wine is exceptionally well buffered and the most robust of the new digestions. Results indicate that even over marginally anomalous soils, many partial extraction techniques will confidently identify the location of mineralization. The use of expensive and proprietary procedures may not produce any better result than using standard reagents or even common beverage solutions as soil extraction reagents.