While the popular means of enhancing the sensor's electroanalytical performance involves novel nanomaterial modifications and functionalization, these are multistep procedures which require several material/electrode characterization studies and are resource intensive. Alternatively, the sensor's performance can also be greatly enhanced by appropriately choosing the electroanalytical technique and by tuning its associated parameters. While such optimization has been popularly addressed via one-parameter-at-a-time strategies, this method demands greater number of experiments and fails to factor in the interaction between variables. Such issues can be addressed via the DoE multivariate statistical technique, which, however has not been widely employed in the field of electroanalysis. Moreover, there is a significant dearth in DoE based works for optimizing the LIG electrode's electroanalytical performance. Hence, in this work we have performed the DoE, FFD-based SWV technique optimization in bare LIG electrodes and have demonstrated a label-free affinity sensing application. While the optimized parameters significantly improved the output signals, we observed counterintuitive traits of increasing current values with protein loading. Hence, to explain these contrary trends we evaluate the electrochemical and physicochemical traits of model protein, BSA modified LIG electrodes, wherein we report the eccentric porous electrode behavior ensuant upon protein modification, associated with pore accessibility in these electrodes. Subsequently, we attribute the pore accessibility traits to hydrophilicity & steric hindrance aspects, and correlate it to the CPE effects, relaxation times and diffusion resistance. While few LIG-based works have reported such contrary trends in label-free affinity sensing, they have not provided the plausible explanation, and this is the first report to comprehensively elucidate the reasons.