Machine learning has revolutionized materials property prediction, yet can exhibit substantial performance degradation when extrapolating beyond training distributions—precisely the capability required for discovering unprecedented materials. In the end-to-end graph neural networks (GNNs) baselines examined here, joint optimization couples learned representations to target property distributions, and predictions tend to remain near the training-property range. We demonstrate that decoupled transfer learning—separating pretrained GNN feature extractors from simple regressors—addresses this challenge. Pretrained features provide transferable structural knowledge, while simple regressors enable smooth extrapolation by maintaining learned trends beyond training boundaries. Benchmarked on layered intercalation compounds through four rigorous extrapolation scenarios and a temporal Materials Project split, our framework achieves 68% error reduction (RMSE: 0.881 vs. 2.778 eV/atom) versus end-to-end CGCNN for extrapolation. Failure analysis reveals extrapolation succeeds for continuous chemical space but fails for discontinuous space, establishing clear design principles. Validated on Fermi energy prediction, this framework is immediately deployable using existing pretrained models, requiring no architectural innovations—transforming ML-driven materials discovery.