During the COVID-19 outbreak, the food delivery market in the United States began to thrive. However, Grubhub, one of the largest food delivery platforms, did not capitalize on this opportunity and experienced severe net losses and a significant decline in market share. Despite the popularity of research on the demographic factors affecting the food delivery market, geographic factors were poorly concerned. In this paper, more attention was paid to reveal the geographical factors that led to the recession of Grubhub under the pandemic. Four machine learning models, namely Linear Regression, Support Vector Regression, Bayesian Ridge Regression, and Elastic Net, were applied to identify the unusual decrease in the net income of Grubhub using Python. This paper then explore the geographical factors by visualizing the business and demographic data. The predicted results show that Grubhub's performance was far below its average over the past two years. Furthermore, by data visualization, it is found that a major geographical factor preventing Grubhub from capturing opportunities is its lack of business expansion into suburban and rural areas.