Understanding soil organic carbon (SOC) sequestration is important to develop strategies to increase the SOC stock and, thereby, offset some of the increases in atmospheric carbon dioxide. Although the capacity of soils to store SOC in a stable form is commonly attributed to the fine (clay + fine silt) fraction, the properties of the fine fraction that determine the SOC stabilization capacity are poorly known. The aim of this study was to develop an improved model to estimate the SOC stabilization capacity of Allophanic (Andisols) and non‐Allophanic topsoils (0–15 cm) and, as a case study, to apply the model to predict the sequestration potential of pastoral soils across New Zealand. A quantile (90th) regression model, based on the specific surface area and extractable aluminium (pyrophosphate) content of soils, provided the best prediction of the upper limit of fine fraction carbon (FFC) (i.e. the stabilization capacity), but with different coefficients for Allophanic and non‐Allophanic soils. The carbon (C) saturation deficit was estimated as the difference between the stabilization capacity of individual soils and their current C concentration. For long‐term pastures, the mean saturation deficit of Allophanic soils (20.3 mg C g−1) was greater than that of non‐Allophanic soils (16.3 mg C g−1). The saturation deficit of cropped soils was 1.14–1.89 times that of pasture soils. The sequestration potential of pasture soils ranged from 10 t C ha−1 (Ultic soils) to 42 t C ha−1 (Melanic soils). Although meeting the estimated national soil C sequestration potential (124 Mt C) is unrealistic, improved management practices targeted to those soils with the greatest sequestration potential could contribute significantly to off‐setting New Zealand's greenhouse gas emissions. As the first national‐scale estimate of SOC sequestration potential that encompasses both Allophanic and non‐Allophanic soils, this serves as an informative case study for the international community. A quantile regression model was developed and used to estimate the soil C stabilization capacity and saturation deficit for a wide range of New Zealand pasture and cropping soils. The overall mean saturation deficit for these soils was 15.1 ± 0.8 mg C g−1 soil. Accounting for differences in bulk density and land area, the mean saturation deficit of each major soil Order was used to estimate a national C sequestration potential of 124 ± 37 Mt C.