AbstractThe primary objective of sugarcane (Saccharum spp.) variety testing programs is to identify cultivars that have high cane yields and good ratooning ability (RA). The purpose of this study was to evaluate the usefulness of six RA indices to estimate RA among sugarcane varieties and to identify an index well suited to predict RA in long‐term ratoon crop cycles. The six indices were as follows: ratoon crops yield means, second ratoon to plant cane yield ratio, final ratoon crop to plant cane yield ratio, mean of ratoon crops to plant cane yield ratio (RP), linear regression, and quadratic regression. Data from four variety trials consisting of eight post‐release varieties plus a control, harvested early, mid, and late in the cropping season for over six crop cycles, were used to estimate the RA indices. To identify indices that can predict RA in long‐term ratoon crop cycles, simple linear regression analyses were used to compare the RA indices derived from 6‐year crop cycles and similar indices derived from data from 111 commercial fields harvested over 11 successive crops. Coefficient of determination (R2) values plotted against their respective ratoon crop numbers were used to determine the minimum number of ratoon crops required to accurately predict RA indices for the 11‐year crop cycle. The six indices varied in their rankings of the RA of varieties, emphasizing the importance of identifying the most appropriate indices for different ratooning practices. Based on its highest R2 value of 0.92, the RP index was identified as the most accurate index to predict RA in long‐term ratoon crop cycles requiring a minimum of five successive crops.