Abstract: To promote the sustainable development of paddy-upland rotation systems in the middle reaches of the Yangtze River, a long-term fixed-field experiment was conducted from 2017 to 2022. The study comprised of five typical triple-cropping patterns: Chinese milk vetch-early rice-late rice (CRR, CK), Chinese milk vetch-early rice-sweet potato || late soybean (CRI), rapeseed-early rice-late rice (RRR), rapeseed-early rice-sweet potato || late soybean (RRI), and potato-early rice-late rice (PRR). The experiment aimed to study the effects of different planting patterns on rice yield and soil nutrients. The results showed that: 1) Overall, over the 6-year period, the yield of late rice was generally higher than that of early rice; the yield of early rice showed a steady upward trend; and the total yield and late rice yield exhibited a trend of first increasing, then decreasing, and then increasing again. Analysis among treatments indicated that except for 2021, the late rice yield of the CRI and RRI treatments was higher than that of other treatments. Additionally, except for 2021, the total yield of these two treatments was the highest in different years, significantly 10.06%−30.23% higher than that of the RRR treatment (P<0.05). Stability analysis showed that during the 6 years, the yield stability of early rice in the CRI and RRI treatments was better than that of late rice, while the yield stability of late rice in the CK, RRR, and PRR treatments was better than that of early rice. 2) The RRI treatment significantly increased soil pH and alleviated soil acidification. The organic matter content of the CRI and RRI treatments was significantly higher than that of other groups in most years; for example, in the early and late rice seasons of 2020 and the early rice season of 2022, the organic matter content of the CRI treatment was 7.52%−23.56% higher than that of CK (P<0.05); in the early and late rice seasons of 2019 and the late rice season of 2017, the organic matter content of the RRI treatment was significantly 7.23%−20.98% higher than that of CK (P<0.05). Regarding soil total nitrogen, the effect of seasons on total nitrogen was significant. In the RRI treatment, during the winter crop seasons and early rice seasons of 2017 and 2019, as well as the winter crop season of 2020, the total nitrogen content was significantly 4.96%−24.64% lower than that of CK (P<0.05). The CRI, RRI, and PRR treatments all increased soil available phosphorus, among which the PRR treatment showed the most stable effect. In the winter crop seasons, early rice seasons, and late rice seasons from 2019 to 2022, the highest soil available potassium content in the PRR treatment reached 143.45 mg·kg⁻¹, which was 156.07% higher than that of CK (P<0.05). Compared with CK, the CRI treatment increased hydrolyzable nitrogen, while the RRR, RRI, and PRR treatments decreased hydrolyzable nitrogen. 3) Correlation analysis indicated that rice yield was significantly correlated with available phosphorus and hydrolyzable nitrogen (P<0.05), and significantly negatively correlated with pH (P<0.05). Therefore, in the triple-cropping paddy field areas of the middle reaches of the Yangtze River, the planting patterns of CRI and RRI are conducive to improving rice yield and soil fertility, and are suitable for popularization and application in the middle reaches of the Yangtze River.