We study the query performance prediction (QPP) task for fusionbased retrieval. Within such a retrieval setting, several ranked lists, each one retrieved by a different method, are combined into a single (fused) ranked list. A common prediction approach is to treat the (base) ranked lists as reference lists and combine those lists' QPP estimates according to their similarity with the fused-list. Yet, we identify a gap in the way that relevance-dependent aspects of inter-list relationships are modeled within such an approach. Aiming to address this gap,we derive an enhanced estimation approach which results in a more accurate prediction.