The predictability of performance factors and costs affecting the cost of gain of fed cattle in Texas feedyards

Abstract

Data including actual measures and estimates of various performance and cost factors were collected from three commercial feedyards located in the Texas Panhandle. The effectiveness with which such performance and cost factors were estimated was evaluated. Correlation coefficients between estimated and actual performance measures ranged from a low of .43 to a high of .93 while correlations between estimated and actual cost measures ranged from a low of .21 to a high of .69. Statistical analyses were performed on the data to evaluate the effects of yard (YD), year (YR), sex (Sex), placement month (InMo) and initial weight class (IWtC) on average daily intake on an as-fed basis (Int), average daily gain (ADG), feed conversion (FC), percent death loss (%DL), veterinary medicine and services cost (VCst) and cost of gain (COG). To facilitate the development of predictive means, additional analyses were performed which evaluated the combined effects of sex, placement month and initial weight class (SMWt) on the various performance and cost measures. Least-squares means and standard errors for intake as a percent of average weight (%Int) by SMWt were used in conjunction with estimated average weight to predict Int. Predictions of ADG were obtained via the Lofgreen-Garrett equations to which an adjustment factor based on InMo was applied. Predictions of FC were mathematically calculated. Least-squares means for %DL by sex * placement month * initial weight class published by Underwood (1991) were used to predict %DL and associated costs (DLCst) and VCst. Feed costs, COG and breakeven costs were mathematically calculated. Though accuracy of prediction was improved, the new predictions were not substantially more accurate than original estimates for Int, ADG, FC, or COG. Predictions of %DL, DLCst, and VCst were substantially more accurate than original predictions. However, due to the increased variability of these traits and their potential impact on COG, predictive accuracy with respect to COG appears to be limited.