In a case-control genetic association study, Hardy-Weinberg disequilibrium (HWD), or deviation from Hardy-Weinberg equilibrium (HWE), in the control group is usually considered as evidence for potential genotyping error, and the corresponding SNP is then removed from the study. On the other hand, a truly associated SNP is expected to be out of HWE in both the case and control groups, even if it is in HWE at the population level. We develop a new case-control association test that (i) leverages HWD attributed to true association to increase power, (ii) is robust to genotyping error, and (iii) is easy-to-implement at the genome-wide level. The proposed robust allele-based (RA) joint test expands our recent RA-regression model, for conducting allelic association tests (Zhang and Sun, in press at Biometrics), to incorporate the difference in HWD between the case and control samples into the traditional association measure. We provide the asymptotic distribution of the proposed test statistic under the null and evaluate its type 1 error control at the genome-wide significance level of 5x10-8, in the presence of HWD attributed to factors such as genotyping error that are unrelated to phenotype-genotype association. Finally, we demonstrate the power of the proposed RA joint test through both simulation and application studies.