<p>Residence time (or water age) is often used as a surrogate indicator of water quality in water distribution networks. Water age depends on the system design and associated demands, which are stochastic in nature, so the identification of critical areas (areas prone to high water ages) must be addressed from a probabilistic perspective. A systematic approach for the STATistical characterization of Water AGE (STATWAGE) has been recently presented in the literature. It has proved that Bottom-Up (BU) demand allocation strategies, which acknowledge the pulsating nature of water demands, are associated with greater water age variance than more conventional Top-Down (TD) approaches, which rely on average consumption patterns. The aim of this paper is (i) to assess if there are any laws of behavior that can explain the difference in variance between STATWAGE-BU and STATWAGE-TD by relying on temporal variability levels, and (ii) if such behavioral laws exist, to propose a systematic approach to simplify the estimation of this additional variance and thereby complement STATWAGE-TD results. This paper demonstrates that these differences are related to the short-term variance of network flows, and a new complement (Short-Term Incremental effect over Water AGE Variance, STIWAGEV) is here presented to consider their effect. The STATWAGE-TD + STIWAGEV approach is a potentially meaningful contribution for effectively quantifying the statistical properties of water age without having to rely on complex BU models and/or high-resolution data. Therefore, this framework constitutes the first step towards developing a systematic methodology for accurately and affordably identifying critical areas.</p>