Protozoans make up 50% of the biomass of the equine hindgut, but their role is relatively unknown in comparison to the more prominently studied bacterial community. Previous studies have found differences in the bacterial population between lean and obese horses, but similar trends have not been investigated in the protozoal population. The objective of this study was to characterize the equine hindgut bacterial and protozoal populations based on the body condition score (BCS) category of the host, to identify associations between bacterial and protozoal species with BCS category, and to determine the predictive nature of the populations with respect to host BCS. Equine fecal samples were collected from horses (n = 45) that were body condition scored using the Henneke scale resulting in n = 16 lean horses, n = 10 normal horses, and n = 19 obese horses for 16S and 18S rRNA profiling. QIIME2 (Quantitative Insights Into Microbial Ecology, v. 2020.2) and R statistical software were used for data analysis. 18S α and β diversity measures, observed species, Jaccard, and unweighted UniFrac distances did not show significance between the BCS categories in the protozoal population (P > 0.05). 16S α diversity with observed species, Chao 1, and Shannon diversity showed significant differences between the groups via a Shapiro test (P = 0.0322) and a further Tukey test found the significance between obeseand normal (P = 0.025), while the other categories were not significant. 16S β diversity measures, Weighted UniFrac and PCoA Bray Curtis, found significance between the categories with β disper (P = 0.011) for UniFrac and adonis and β disper (both P = 0.001) for Bray Curtis. Three taxonomic classifications from the protozoa family Blepharocorythidae significantly correlated (Spearman rho > 0.3) positively with the lean and obese categories and negatively with normal. The 3 bacterial taxonomic classifications that correlated with more than one BCS were the class Gastranaerophilales correlated positively with normal and negatively with obese, order Marinifilaceae correlated positively with lean and negatively with normal, and the order Lachnospiraceae correlated positively with normal and negatively with lean. The bacterial population had higher accuracy when modeling (Random forest) for the ability to predict BCS category using the population (lean 61%, normal 86%, obese 83%) than the protozoal population (lean 64%, normal 64%, obese 65%). The bacterial population had more significant differences between BCS categories and better predictive abilities. The protozoal population has limitations due to a lack of research on their role and genomes.