The chapter discusses different ways to center data for biplot analysis using grain yield data from winter wheat genotypes in Ontario, Canada. Five types of data centering are explored: uncentered, grand mean-centered, environment-centered, double-centered, and genotype-centered. The main goal of variety trials is to find high-performing genotypes and suitable test environments. The GGE biplot is the best for evaluating genotypes and test environments, based on solid theoretical reasoning. Other biplot types may be useful for specific research goals.