This study uses energy and greenhouse gas (GHG) balances to evaluate how the scale of sugarcane cultivation affects the performance of a sugarcane bioenergy system generating exportable electricity from bagasse. Small-, medium-, large- and miller-planter systems, with cane field areas of less than 10 ha, 10 – 42 ha, 42 – 2000 ha, more than 2000 ha respectively, were modelled. Each of them also has different combinations of manual and mechanical agricultural operations, resulting in different cane yields. Miller-planter system (fully mechanised) performs best with energy yield ratio of 10.99, GHG emissions in bagasse electricity of 0.0633 kg CO2eq/kWh and avoided life cycle GHG emissions of 82.07% when replacing electricity from coal, whereas small-planter system (fully manual) has the worst performance with energy yield ratio of 6.82, GHG emissions in bagasse electricity of 0.0881 kg CO2eq/kWh and avoided life cycle GHG emissions of 75.03% when substituting electricity from coal. Sensitivity analyses show that relative performances of all sugarcane planter systems both in terms of energy and GHG emissions are not significantly affected by variations in bagasse allocation factor, in sugarcane yield and in fertiliser input (the most energy-intensive and GHG-emitting component). Moreover, they confirm miller-planter system as the overall best performer and indicate that increasing small-planters’ cane yield is the critical measure to improve their energy analysis performance. In terms of the nature of agricultural operations, mechanical operations do not necessarily require more input energy than their manual counterparts, contrary to common belief. This is the case for fertilisation, irrigation and cane loading. Fully mechanised sugarcane production at miller-planter scale is therefore strongly encouraged.