Predicted OBP for 2021 Season

Objective

The OBP Prediction project aims to predict the on-base percentage (OBP) for MLB players in the 2021 season using historical data from 2016-2020. The project utilizes player OBP, plate appearances (PA), and age to estimate future performance, applying a weighted average approach. The predictions are compared to the actual 2021 OBP to calculate total error.

Skills Learned

• Understanding of statistical methods for prediction.
• Experience with CSV data parsing and manipulation.
• Ability to apply weighted averaging to historical data.
• Proficiency in JavaScript for data processing and DOM manipulation.
• Incorporating real-world data, such as player age, into statistical models.

Tools Used

• JavaScript: To calculate OBP predictions, handle data processing, and manage interactions.
• PapaParse: For CSV parsing and data manipulation.
• HTML & CSS: For structuring and styling the webpage.
• GitHub: For version control and project collaboration.

Steps

Initial Setup

The project involves a simple web-based setup using JavaScript, HTML, and CSS. The CSV file with player stats is fetched and parsed using the PapaParse library. A list of predicted OBP values for each player is then displayed on the webpage.

Prediction Process

1. Fetching CSV Data:

   • Player data from 2016-2020, including OBP, plate appearances, and birthdates, is fetched and parsed from a CSV file using PapaParse.

2. Weighted Average Calculation:

   • Historical OBP from 2016 to 2020 is used, with more recent years and those with more plate appearances weighted more heavily. Missing data is handled by assigning zero weight.
   • An arbitrary 5% reduction is applied to players over the age of 30 to account for age-related performance decline.

3. Error Calculation:

   • The predicted OBP for 2021 is compared to the actual OBP for each player, and the total absolute error is calculated.

Running the Prediction

1. Fetch Player Data: Fetch the CSV file containing player OBP and PA data from 2016-2020.
2. Calculate Predictions: Apply weighted averaging to predict the OBP for the 2021 season.
3. Display Results: Display each player's predicted OBP and compare it to the actual OBP from 2021, alongside the total error.

Future Improvements

• Refine weight calculations with data-driven approaches, possibly using machine learning for better accuracy.
• Incorporate additional data, such as exit velocity and launch angle, to improve predictions.
• Explore advanced models that dynamically adjust weights based on historical trends.