Electrocatalytic nitrogen oxide reduction (NO x RR) emerges as an effective way to bring the disrupted nitrogen cycle back into balance. However, efficient and selective NO x RR is still challenging partly due to the complex reaction mechanism, which is influenced by experimental conditions such as pH and electrode potential. Here, we have studied the enzyme-inspired iron single-atom catalysts (Fe–N4–C) and identified that the selectivity roots in the first step of the nitric oxide reduction. Combining the constrained molecular dynamics (MD) simulations with the quasi-equilibrium approximation, the effects of electrode potential and pH on the reaction free energy were considered explicitly and predicted quantitatively. Systematic heat maps for selectivity between single-N and N–N-coupled products in a wide pH-potential space are further developed, which have reproduced the experimental observations of NO x RR. The approach presented in this study allows for a realistic simulation of the electrocatalytic interfaces and a quantitative evaluation of interfacial effects. Our results in this study provide valuable and straightforward guidance for selective NO x reduction toward desired products by precisely designing the experimental conditions.