For decades, models have predicted that gas giant planets that come too close to their host stars will be destroyed, with most of their mass accreted onto the host star.  However, recent work modeling Roche lobe overflow from close-in gas giants now indicates that angular momentum exchange between the escaping gas envelope and the remaining planet core can cause the core to migrate outwards, narrowly avoiding destruction. At the same time, new observational work suggests that high density planets located in the ‘Neptune desert’ (a well-studied deficit of Neptune-sized planets on close-in orbits) may in fact be the tidally stripped cores of Jovian-mass progenitors. In my talk I will describe two separate tests of this hypothesis. First, I will show how surveys of atmospheric mass loss in the high Roche filling factor regime can be used to quantify the effect of this mass loss on the long-term orbital evolution of tidally decaying hot Jupiters. Second, I will describe how the population-level properties of planets currently located in the Neptune desert can be used to investigate their origin.