Correlation-based speckle tracking methods are commonly used in elasticity imaging to estimate displacements. In the presence of local strain, a larger window size (in cross correlation calculation) results in larger displacement error. To reduce tracking error, we proposed using a short window followed by a correlation coefficient filter. Although simulation and experimental results demonstrated the efficacy of the method, it is not clear why correlation filtering reduces tracking error. In this study, we addressed this issue and analyzed the relationship between displacement error and tracking parameters such as window size and filter size. For simplicity, we focused on axial displacement estimation. Analytic forms for tracking without and with correlation filtering were derived to predict tracking error. For the former case, the expression shows increase of error with resolution. For the latter case, there exists one extra negative term so that tracking error decreases with resolution. Furthermore, given a fixed resolution, a smaller window together with a larger filter is preferred. Simulations were performed and the results match the theory well for strains up to 4%.