This study aims to examine the stabilizing role of the government in the Iranian stock market during the COVID-19 crisis using a Markov-Switching MIDAS (Mixed-Data Sampling) regime analysis approach. Applied in nature, the study employs a quantitative-inferential methodology based on mixed-frequency time series. Data were collected covering the period from Dey 1398 to Dey 1401 (January 2020 – January 2023) at both daily and monthly frequencies. The analysis utilized a MIDAS regression followed by a hybrid Markov-Switching MIDAS model to investigate the effects of economic uncertainty, government response, economic support, the COVID-19 growth rate, and healthcare expenditures on Iranian stock market returns and volatility, while distinguishing between "calm" and "turbulent" regimes. Model estimation was performed using EViews 14 software. The findings revealed that the Iranian stock market exhibited non-linear, regime-dependent behavior during the COVID-19 pandemic; the Markov-Switching MIDAS model identified two distinct regimes: "calm" and "turbulent." In the calm regime, government responses and economic support measures were largely sources of volatility or had minimal impact; however, in the turbulent, high-volatility regime, these interventions played a stabilizing role—specifically, the government response and economic support indices contributed to reduced market volatility, with economic support also aiding in the improvement of market returns. The COVID-19 growth rate exhibited dual behavior: it reduced returns during the "calm" regime but was associated with positive returns during the "high-stress" regime, reflecting market expectations of post-crisis supportive interventions. The economic uncertainty index had a significant positive impact on volatility across all models and regimes, while healthcare expenditures negatively affected both returns and volatility. Overall, the results indicate that the effectiveness of government policies in the Iranian stock market depends on the prevailing market regime and the time horizon; furthermore, the MS-MIDAS model provides a more accurate and realistic picture of the stabilizing role of government interventions during the COVID-19 crisis compared to the linear MIDAS regression.