The objective of this paper is to compare four methods of estimating yield on a field-by-field basis using final plant map data. Three of the methods use increasingly detailed boll by position and branch number data, and the fourth is a multiple regression model that was originally developed for interpretive rather than predictive use. Of these, only the method using the most detailed boll retention data appears possibly useful. It appears, however, that to be effective as a predictive tool the method will have to be supplemented with boll size measurements.