nmarek1269 / scrape_ufc_stats

Scrape UFC stats

Scrape UFC Stats

Scrape all available UFC events data, fights stats, and fighter details and from ufcstats.com and save in CSV format.

Prerequisites

Python 3

Install requirements with pip install -r requirements.txt

Introduction

Data for all events, fights, and fighters have scraped and saved as the following data files:

ufc_events.csv
ufc_fight_details.csv
ufc_fight_results.csv
ufc_fight_stats.csv
ufc_fighter_details.csv
ufc_fighter_tott.csv

To download the CSV files without running any code, click Code > Download ZIP or clone the repo.

You can also scrape the all data for fight stats again using the notebook scrape_ufc_stats_all_historical_data.ipynb, and all data for fighter tale of the tape again using the notebook scrape_ufc_stats_fighter_tott.ipynb. Do note these will each take a few hours to complete.

Once you have the up-to-date historical data for fight stats, you can run the notebook scrape_ufc_stats_unparsed_data.ipynb or the script scrape_ufc_stats_unparsed_data.py to scrape the only latest fights and refresh the data.

The notebook scrape_ufc_stats_working_example.ipynb can be used for testing or debugging. The code here is broken down into sections which can be executed to scrape single data points, e.g. scraping stats for one fight only.

Do feel free to reach out with any comments, suggestions, or issues. 😃

Data Refresh

The script scrape_ufc_stats_unparsed_data.py has been added to PythonAnywhere and will run daily to check and scrape new fights and fighters and push the refreshed data files to this repository.

This keeps the data files up to date with the latest fight and fighter stats and you can quickly download and use the CSV files above without running any code.

Next Steps

Data Refresh for Fights - Parse and refresh new fight data programmatically when there are new events

Scrape Fighter Details - Get all fighter's tale of the tape (e.g. height, weight, etc.)

Data Refresh for Fighters - Parse and refresh new fighter details programmatically when there are new fighters added to the roster

Clean Data - Clean data to be used for further analysis

Data Exploration - Plot and visualise stats

Prediction - Build a machine learning model to predict the outcome of fights