Proposal

Dataset: https://www.kaggle.com/datasets/maksbasher/ufc-complete-dataset-all-events-1996-2024

My final project will use this dataset that contains data scraped from http://ufcstats.com

The purpose of this app is to give an analytical overview of the history of UFC fights. My intent is to show information such as which stance matchups are favorable, distribution of winning age, stance popularity by time, countries of winning fighters over time, etc.

For user input, the user will be able to compare specific fighters; the user will be able to select stance matchups and compare them (Orthodox, southpaw, switch). The user will be able to select countries and see their fighters’ performances over time.

The app will produce plots, tables, and maybe, if it is within scope, A user enters name, weight, height, reach, stance, wins etc. for two fighters and the app will give a prediction for which fighter it thinks will win.

Tabs will be Fighter Comparison | Stance Matchups | Fight Predictor I anticipate the prediction will be the hardest aspect of the project. Personally, I wish the dataset I found had more information about fighters such as their known disciplines (say if the fighter is a world class BJJ specialist going against someone who never wrestled a day in their life). Martial art discipline and skill level play a huge role in match outcomes, especially in the modern day.

For instance, Conor McGregor, in his first 7 matches in the UFC, was knocked out (by punch) 6 of those fights. Some of the opponents were world class names like Jose Aldo and Dustin Porier. Then suddenly, in 2018, Khabib Nurmagomedov showed the world just how dominant an amazing wrestler can be in a fight. Nurmagomedov’s fights are frequently described as “maulings” because of how dominant and controlling his victories were. He frequently made other seasoned, world class fighters look helpless in the octagon.

Project

So

Shiny App