Xinjiang is the largest arid inland region in China and its agricultural development plays an important role in the socio-economy of China. Drought is one of the major natural hazards in this region and has impacts on agriculture. In this study, spatiotemporal patterns of droughts are analysed based on monthly precipitation datasets from 52 meteorological stations covering the period of 1961–2008 using the Rotated Empirical Orthogonal Functions (REOF) technique and also the Mann–Kendall trend test (MK test). Results of the study indicate that: (1) a wetting tendency is observed in recent years, particularly in northern Xinjiang. Seasonally, the wetting tendency is found mainly in winter. The modes of REOF can be divided by the Tianshan mountain and the principal component (PC) of the northern Xinjiang mode in winter has a significant increasing trend. (2) The area of drought-affected regions is decreasing in northwestern Xinjiang and along the northern side of Tianshan mountain. Seasonally, the area of drought-affected regions is significantly decreasing in winter. (3) The wetting tendency of Xinjiang may alleviate the drought influence. However, uneven spatial patterns of precipitation changes may enhance drought impacts. Besides, drought-related impacts can be greatly alleviated where irrigation conditions are favourable.