Advanced methods exist to estimate dietary crude protein (CP) degradation in the rumen from the concentrations and/or proportions of chemical nutrient fractions. These methods have been developed primarily for feeds in temperate climates. Therefore, their accuracy and precision to predict proportions of rumen-undegraded CP (RUP) in tropical forages is uncertain. The objectives of the present study were (1) to assess the accuracy and precision of the estimates of in situ RUP proportions from CP fractions using established algorithms, and (2) to identify chemical nutrient fractions that accurately and precisely predict RUP proportions of tropical forages. The proximate nutrients, CP and fiber fractions of 23 unconserved forage grasses and 15 forage legumes commonly used as feeds for ruminants in the Tropics and Subtropics were analyzed. Their RUP proportions were determined at ruminal passage rates (Kp) of 2%, 5%, and 8%/hour using the in situ method (i. e., reference) and were also estimated from the analysis of the CP and fiber fractions using established equations. The RUP proportions determined with the in situ and the chemical method were evaluated using error index and dimensionless parameters. The nutritional composition variables that can predict RUP proportions of tropical forages (proximate nutrients, fiber, and CP fractions and their ratios) were selected using stepwise multiple linear regression models. The CP concentrations of our sample set ranged from 46 to 212 g/kg dry matter with in situ RUP proportions from 73 to 596 g/kg CP across all Kp. Previously established equations could not predict accurately and precisely RUP proportions of all forages. The concentrations of CP and fiber fractions and Kp were able to predict the proportion of RUP estimated in situ for tropical forages at Kp of 2%, 5%, and 8%/hour with an adjusted coefficient of determination ranging from 0.79 to 0.82 and a root mean square error ranging from 13% to 15% of the mean in situ RUP proportion. While equations established for temperate forages either over- or underestimate RUP proportions, algorithms developed in the present study might allow for an accurate and precise prediction of RUP proportions in forages commonly used in ruminant feeding in the Tropics and Subtropics.