Hainan state farms are the largest production base of natural rubber in China. The nutrient status of rubber plantations has changed greatly since the second national soil survey that was published in 1982, but its specific conditions are not clear at present. Therefore, the soil productivity of the main rubber plantations in Hainan state farms was systematically evaluated by applying the evaluation methods of cultivated land productivity. The objective of this study was to provide theoretical support for the improvement of soil productivity of rubber plantations and sustainable development of natural rubber. For this study, the rubber plantations of Hainan state farms were split into eight different types on the basis of soil parent material and rainfall. The granite rainy area, which refers to the soil parent material for the granite and the average monthly rainfall for more than 100mm, accounted for 56.97% of the total area of rubber plantations in Hainan state farms and was therefore selected as a representative assessment area. A total of 7204 soil samples of the 0-20 cm tillage layer were collected from this area to evaluate soil productivity grades. We applied the Delphi method, the fuzzy mathematics theory and the analytical hierarchy process method. This would enable the determination of 11 assessment indexes and their weights based on the six principles of necessity, importance, diversity, stability, ease of access and simplicity. The soil productivity of the rubber plantations in the assessment area was then split into five grades. The results showed that the soil productivity of the assessment area was mainly in the second, third and fourth grades. This accounted for 85.85% of the total rubber plantation area in the granite rainy area, and was widely distributed in 15 cities or counties: Qiongzhong, Tunchang, Chengmai, Danzhou, Wanning, Baoting, Ledong, Sanya, Lingshui, Dingan, Qionghai, Wuzhishan, Changjiang, Wenchang and Dongfang. Soil that was assigned to the first productivity grade was found in a few, scattered places, whereas soil in the fifth productivity grade was not common but was found in a relatively concentrated area. Soil organic matter contents, soil pH and the concentrations of available phosphorus (P) and potassium (K) were found to be important impact factors affecting the soil fertility level of the rubber plantations. The soil pH was determined to be suitable for growing rubber trees in the assessment area. However, the content of soil organic matter and the concentration of available P were below the value suitable for growing rubber trees. The concentration of available K was suitable for growing rubber trees in the first and second productivity grade soils, only just reached the required level for rubber tree growth in the third and fourth productivity grade soils and was not suitable for normal growth of rubber trees in the fifth productivity grade soils of the assessment area. According to these results, it is suggested that green manure should be vigorously intercropped, and that fertilizing with N, K, Ca and Mg simultaneously with increasing organic fertilizer will strengthen the nutrition management and improve the utilization rate of fertilizer in rubber plantations.