Rainfall is an oscillatory rather than purely stochastic signal, whose variability reflects alternating hydrological regimes rather than long-term trends. Recognizing this regime-based nature marks a conceptual shift in the way climatology interprets rainfall variability. At the monthly to multiannual scale, precipitation evolves through irregular wet, dry, and stationary phases whose duration and intensity vary over time. Although trend analyses, anomaly-based metrics, and spectral methods may at times suggest contrasting interpretations – each being sensitive to different aspects of the signal – they capture only partial views of a shared underlying variability. Framing precipitation as a sequence of irregular regimes offers a unifying perspective that helps reconcile these approaches and clarifies how rainfall fluctuations actually unfold. Using the Po River basin (Northern Italy) as an illustrative case, we show that Fourier and wavelet analyses confirm the intermittent character of rainfall oscillations, with regular periodicities emerging only at limited intervals. The Cumulative Deviation from Normal (CDN), computed as the cumulative sum of standardized monthly precipitation (SPI1), provides a simple yet physically consistent framework to visualize these irregular regimes and to quantify the resulting changes in water availability driven by cumulative surplus or deficit.