We adapt the Life Cycle (LC) curve methodology to the problem of statistical anomaly detection in time series. LC-curves transform a raw non-negative time series into a cumulative share profile that highlights when, within a fixed observation window, most of the signal intensity is concentrated. Deviations of this profile from the uniform-accumulation baseline serve as simple anomaly indicators that require no distributional assumptions. As an application, we apply the method to 5-min GPS displacement time series from the Nevada Geodetic Laboratory, comparing 196 earthquakes with M ≥ 6 against 104 seismically quiet intervals, and evaluate performance with Probability of Detection (POD), False Alarm Ratio (FAR), and Critical Success Index (CSI). The method achieves POD ≈ 0.82 (CSI = 0.62, FAR = 0.29) at the lowest tested threshold. Pre-event anomalies are statistically distinguishable from calm-period anomalies, demonstrating that LC-curves can effectively convert high-frequency time series into actionable anomaly indicators.