Recent studies have demonstrated the effectiveness of switching bipolar radiofrequency ablation (bRFA) in treating liver cancer. Nevertheless, the clinical use of the treatment remains less common than conventional monopolar RFA – likely due to the lack of understanding of how the tissues respond thermally to the switching effect. The problem is exacerbated by the numerous possible switching combinations when bRFA is performed using bipolar needles, thus making theoretical deduction and experimental studies difficult. This article addresses this issue via computational modelling by examining if significant variation in the treatment outcome exists amongst six different electrode configurations defined by the X‐, C‐, U‐, N‐, Z‐ and O‐models. Results indicated that the tissue thermal and thermal damage responses varied depending on the electrode configuration and the operating conditions (input voltage and ablation duration). For a spherical tumour, 30 mm in diameter, complete ablation could not be attained in all configurations with 70 V input voltage and 5 minutes ablation duration. Increasing the input voltage to 90 V enlarged the coagulation zone in the X‐model only. With the other configurations, extending the ablation duration to 10 minutes was found to be the better at enlarging the coagulation zone.