Basin similarity is one of the important issues in hydrology. Thus, discovering a statistical technique to cluster basins that possess similar attributes helps hydrologists achieve better management of the basins for the sustainability of water resources. This study employed the affinity propagation (AP) technique for clustering 87 basins in the Kingdom of Saudi Arabia (KSA) which had areas ranging from 10.44 to 1364.77 km2. The cluster analysis was performed on nine scenarios comprising individual and combinations of indices based on climate, morphology, hydrology, land use, and soil type that cover many attributes of the basins. The results showed that neighboring basins were usually but not always have a high degree of similarity and belong to the same cluster, such as classification based on climate, land use, and soil type. We evaluated the clustering and similarity measures using percentile analysis of the attributes, Cramer's V, and Adjusted Rand Index (ARI). In general, the percentile analysis showed that most of the attributes are categorized by medium-to-high percentiles more than 33%. Moderate Cramer's V value indicated that the basins within each cluster were associated with eight geomorphic provinces in KSA except scenarios based on soil types which shared a weak Cramer's V value indicating high variability. The highest Cramer's V value was 0.48 which comes from the climatic scenario. ARI value showed a weak agreement between scenarios, especially since six pairs have negative values indicating that the clusters are extremely discordant. However, scenarios based on hydrology versus all indices shared the highest score of 0.979 indicating identical clusters. In general, the technique converges to a meaningful number of clusters when the number of indices is low. This implies the difficulty of coming up with global similarities between the basins.