This paper uses standard and penalized logistic regression models to predict the Great Recession and the Covid-19 recession in the US in real time. It examines the predictability of various macroeconomic and financial indicators with respect to the NBER recession indicator. The findings strongly support the use of penalized logistic regression models in recession forecasting. These models, particularly the ridge logistic regression model, outperform the standard logistic regression model in predicting the Great Recession in the US across different forecast horizons. The study also confirms the traditional significance of the term spread as an important recession indicator. However, it acknowledges that the Covid-19 recession remains unpredictable due to the unprecedented nature of the pandemic. The results are validated by creating a recession indicator through principal component analysis (PCA) on selected variables, which strongly correlates with the NBER recession indicator and is less affected by publication lags.