Suspicion of malignant change within a lipoma is a common and increasing workload within the UK Sarcoma multidisciplinary team (MDT) network, and a source of considerable patient anxiety. Currently, there is no lipoma-specific data, with regard to which clinical or radiographic features predict non-benign histology, or calculate an odds-ratio specific to a lipomatous lesion being non-benign. We performed a 9-year, double-blind, unmatched cohort study, comparing post-operative histology outcomes (benign versus non-benign) versus 15 signs across three domains: Clinical (size of tumour, depth, growth noticed by patient, previous lipoma, patient felt pain), Ultrasonographic (size, depth, vascularity, heterogenous features, septae) and MRI (size, depth, vascularity, heterogenous features, septae, complete fat signal suppression). Receiver operating characteristic (ROC) analysis, odds ratios and binary logistic regression analysis was performed double-blind. When each sign is considered independently, (ROC analysis, followed by binary logistic regression) only Ultrasound depth is a significant predictor (p = 0.044) of a histologically non-benign lipoma. Ultrasonographically determined vascularity and septation were not statistically significant predictors. None of the clinical signs were statistically significant (p > 0.05). Of the MRI signs none was statistically significant (p > 0.05). However, heterogeneous MRI features fared better than MRI depth. Ultrasound signs (Pseudo R-Square = 0.105) are more predictive of the post-operation histology outcome than Clinical signs (Pseudo R-Square = 0.082) or MRI tests (Pseudo R-Square = 0.052) Ultrasound and Clinical tests combined (Pseudo R-Square = 0.147) are more predictive of the post-operation histology outcome than MRI tests (Pseudo R-Square = 0.052). This work challenges the traditional perception of “red-flag” signs when applied to lipomatous tumours. We provide accurate data upon which an informed choice can be made, and provides a robust bases for expedited risk/benefit. The importance of an experienced and cohesive MDT network is emphasised.