Energy systems models are diagnostic tools, not projection machines
Energy systems models (ESMs) are a leading tool to guide the energy transition. They have been influential in supporting national decarbonisation strategies and regional system planning, but these complex models can be opaque. Their results are often presented predictively or prescriptively, with little exploration of uncertainty and without clear discussion of limitations. Consequently, projections of ESMs have often been given more authority than their evidence can bear, undermining their contributions to energy transition policy. We argue that ESMs should instead be applied as explanatory, diagnostic tools. Model studies should explore uncertainty to find robust insights and define limitations, interrogate model behaviour to find testable real-world explanations, and communicate these explanations plainly and responsibly. This approach shifts the evidentiary burden from plausible projections to real-world insights that can be broadly understood and debated, and considered alongside other forms of evidence. Used this way, ESMs can support robust, justifiable, and pluralistic decision-making.