This study integrates purely statistical methods of Mann–Kendall (MK) and Spearman rho (SMR) with statistical-graphical methods of Onyutha trend (OT) test and innovative trend analysis (ITA) to examine annual and seasonal rainfall variations at 12 stations across the Shire River Basin (SRB) during 1976–2005. The results reveal a general decreasing trend for annual rainfall throughout the basin. At seasonal scale, the following trends were observed: an increase for the December-January–February (DJF) season, especially in the southern portion of the basin; a decrease for the March–April-May (MAM) and June-July–August (JJA) seasons; and inconclusive results for the September–October-November (SON) season. Despite nearly all time series indicating consistent trend direction as established by the four tests, the ITA identified the most significant rainfall patterns on both annual and seasonal basis. The performance abilities for the MK, SMR, and OT tests demonstrated the closest agreement at the verified significant level. In addition to the monotonic trend results obtained statistically, sub-trends are visually distinguished using the graphical features of the OT and ITA approaches. For the former, changes are seen as step jumps in the mean of the data, and for the latter, trends regarding high and low rainfall clusters are evaluated, hence offering more details regarding rainfall variability, such as the SRB’s sensitivity to both floods and droughts. Thus, the completely different aspects offered by the visually oriented methods complement the purely monotonic trend detection methods.