This study presents the importance of coherency in modeling thermal-hydraulics and mechanical behavior of a solid for an advanced prediction of cladding thermal shock fracture. In water quenching, a solid experiences dynamic heat transfer rate evolutions with phase changes of the fluid over a short quenching period. Yet, such a dynamic change of heat transfer rates has been overlooked in the analysis of thermal shock fracture. In this study, we are presenting quantitative evidence against the prevailing use of a constant heat transfer coefficient for thermal shock fracture analysis in water. We conclude that no single constant heat transfer could suffice to depict the actual stress evolution subject to dynamic fluid phase changes. Use of the surface temperature dependent heat transfer coefficient will remarkably increase predictability of thermal shock fracture of brittle materials. The presented results show a remarkable stress prediction improvement up to 80-90% of the actual stress with the use of the surface temperature dependent heat transfer coefficient. For thermal shock fracture analysis of brittle fuel cladding such as oxidized zirconium-based alloy or silicon carbide during LWR reflood, transient subchannel heat transfer coefficients obtained from a thermal-hydraulics code should be used as input for stress analysis. Such efforts will lead to a fundamental improvement in thermal shock fracture predictability over the current experimental empiricism for cladding fracture analysis during reflood.