abasaltr / baseball_learning Goto Github PK

=======

HEROKU Live Dashboard Deployment - https://baseball-learning-g3.herokuapp.com

Finish Liners (Group 3)

Reza, Germaine, Nick, and Jeremy

Baseball has been dubbed as Americas past time. For our project we decided to take over a hundered years worth of baseball data and design an extensive website that will include links to data, visuals, and a dashboard. The dashboard will show extensive visuals to show our findings in a way that tells a story. The goal is to show stats up to the end of 2019 and we will show stats of not only players but of teams as well. When given the option a user will either pick his team or player and it will pull up stats by team and player. Additionally the user can navigate to the dashboard and then walk through the same path but with more options.

http://www.seanlahman.com/files/database/readme2017.txt

http://www.seanlahman.com/baseball-archive/statistics/

For our statistical analysis we will use a regression model to do a series of predictive analysis utilizing attributing stats to predict a team or players probablity to win or lose. With this, we can then ultimately build a super team based on the analytics that theoretically would have the most wins.

We will use the SVM model with GridSearchCV in this project.

• We will likely use Tableau's extensive library of visuals to create the visuals for this project.

• Python
• Flask
• Web Based UI (HTML/CSS/Bootstrap)
• Tableau
• Heroku