Stock Market Price Prediction plays a critical area of research where forecasting future stock prices can provide useful insights to the users of such models, like investors, analysts and others. Based on historical prices, the performance of three different regressors, namely Linear Regression, Random Forest and Gradient Boosting, has been predicted. The study utilizes historical data of Infosys Ltd. And NIFTY 50 over the past two years. Then, the two different types of train-test split strategies have been implemented to know the reliability of these models, with a focus on error distribution and comparison between actual and predicted values. Three performance metrics – Mean Absolute Error (MAE), Mean Squared Error (MSE) and R2 score are used to measure each model’s performance. Visualization tools like residual plots, histograms, and line graphs are used for effective analysis and interpretation of these models. Linear Regression model turns out to be a superior model as it outperforms other models, achieving lower error metrics and the highest R2 score, indicating a better fit across both data splits. While Gradient Boosting performed closely with low error but its overall variance was slightly higher. Random Forest, on the other hand, displayed higher error margins and greater inconsistency across both splits. The importance of selecting models based on their accuracy and consistency for reliable financial forecasting is emphasized in this study.