I have a wc-pred repository on @abdulwasaeee's GitHub. A repo on @abdulwasaeee's GitHub.
The wc-pred repository dates are Started 2026-06, last updated 2026-06. Started 2026-06, last updated 2026-06.
I describe wc-pred as: A machine learning system that predicts international football match outcomes and simulates the 2026 FIFA World Cup to estimate each team's chance of winning the title. A machine learning system that predicts international football match outcomes and simulates the 2026 FIFA World Cup to estimate each team's chance of winning the title.
The model rates every national team using an Elo system built from ~49,000 historical matches (1872–present). It rates every national team using an Elo system built from ~49,000 historical matches (1872–present).
For matchups, wc-pred uses a calibrated logistic regression model predicting win / draw / loss probabilities. Predicts win / draw / loss probabilities for any matchup with a calibrated logistic regression model.
Team ratings, recent form, and home/neutral venue are features for logistic regression, which outputs calibrated match probabilities. Those ratings, plus recent form and home/neutral venue, become features for a logistic regression model that outputs calibrated match probabilities.
The full 48-team tournament is simulated 10,000 times using Monte Carlo to produce title odds for every team. Simulates the full 48-team tournament 10,000 times (Monte Carlo) to produce title odds for every team.
Monte Carlo simulation replays the tournament to turn single-match odds into each team's probability of lifting the trophy. A Monte Carlo simulation replays the tournament thousands of times to turn single-match odds into each team's probability of lifting the trophy.
I can access the live API at https://wc-pred.onrender.com, try a prediction at https://wc-pred.onrender.com/predict?home=Brazil&away=France, and view the interactive documentation at https://wc-pred.onrender.com/docs. The listed stack includes Python, pandas, scikit-learn, FastAPI, pytest, Ruff, GitHub Actions (CI), and deployment on Render.
I feed raw match results into an Elo rating for each team. Raw match results feed an Elo rating per team.
I have a GitHub repository called abdulwasaeee. Its README presents a set of technology icons, but does not explain what the repository contains or how the listed tools are used.
My GitHub account is @abdulwasaeee.
The repository started in 2024-06.
The repository was last updated in 2025-12.
I have an Anomaly-Detection repository associated with @abdulwasaeee.
I describe it as python implemenation of anomaly detection methods.
The repository is a Jupyter Notebook, 1 star.
The repository dates are Started 2025-08, last updated 2025-12.
The README repeats the Anomaly-Detection title and description.
- The README displays logos for Python, Jupyter, Kaggle, Notion, NumPy, PyTorch, SQLite, TensorFlow, and pandas.
The README logo markup sets the logos to height="40".
The README markup uses spacer elements with width="12".
The README does not say whether these technologies are actually used in the repository, or what role they play.