The relationship between extreme rainfall and temperature (‘scaling’) is complex, yet it is crucial to understand how extreme rainfall changes with temperature due to its impacts on human lives. In theory, extreme rainfall is expected to increase at a rate similar to Clausius-Clapeyron (CC) scaling, i.e. approximately ~7%˚C^-1. There are various methods available to calculate scaling, however, it is particularly difficult to apply these methods to Australia as a whole continent as rainfall in different regimes is dominated by different physical processes and this leads to large differences. There have been numerous attempts to investigate the relationship between extreme rainfall and surface temperature in regions elsewhere using the binning method. Previous studies show a strong negative scaling rate – decreases in extreme rainfall with increased temperatures – in northern Australia due to the strong cooling effect that occurs during an extreme rainfall event. However, in this study, we demonstrate that removal of the cooling effect does not result in a positive scaling relationship. We further show that the existing methods fail to produce robust results due to various factors such as the assumption of a linear relationship between extreme rainfall and temperature, local cooling effect, strong seasonal signal, and varying precipitation type. This study highlights the pitfalls of using existing scaling methods and expecting a scaling equivalent to CC relationship. Improvements around these methods are suggested for future research to understand temperature-extreme rainfall relationships.