Modern Python Data Stack
Hands-on Pandas & Polars, tidy data patterns, vectorized transforms, and reproducible notebooks with environment pinning for reliable runs.
Python MachineZ — Machine Learning with Sports Data
Master Machine Learning with Python
Build ML skills that transfer across industries—finance, health, retail, media, and sports. Learn modern Python tooling, production-ready workflows, and model deployment patterns while working on real, portfolio-ready projects.
Pandas & Polars
Feature Engineering
Scikit-learn Pipelines
XGBoost & LightGBM
Model Ops & Evaluation
APIs & Deploy
What You’ll Learn
A practical, end-to-end machine-learning journey—from clean data pipelines to performant models and deployment. Each module feeds a portfolio-ready deliverable you can show to hiring managers and stakeholders.
Feature Engineering at Scale
Create robust features (lags, rolling windows, encodings), handle leakage, and package steps into scikit-learn Pipelines for repeatability.
Modeling & Tuning
Train and tune tree-based models (Random Forest, XGBoost, LightGBM) with cross-validation, early stopping, and hyperparameter search that respects time.
Evaluation that Stakeholders Trust
Pick metrics that match business goals (AUC, logloss, MAE), add calibration, lift/decile charts, and build concise model cards executives can act on.
Operational ML
Persist artifacts, track experiments, and expose predictions through fast APIs. Learn dependency hygiene and light-touch CI so models ship safely.
Capstone: Industry Portfolio
Ship a polished project (your choice: finance, health, retail, media, or sports). Publish a live demo and a read-me recruiters will actually read.
Feeling Overwhelmed by Machine Learning?
You’re not alone. We turn complexity into shippable, production-ready work that stakeholders can actually use.
Common Roadblocks We’ll Tackle Together
- ?
Applying ML techniques to messy, real-world data.
- ?
Selecting the right algorithms for business problems.
- ?
Tuning hyperparameters and evaluating models with meaningful metrics.
- ?
Automating data pipelines and shipping reliable deployments.
- ?
Digesting advanced methods (tree-based ensembles, neural nets, Bayesian ideas) without drowning in math.
WNBA Shot Probability (Tree-Based Models)
Data Source
2023 WNBA play-by-play via py_ball (stats.wnba.com). Coordinates normalized for standard court view.
Key Features
locX, locY, shot distance, angle, side, shot type (hook/layup/jumper). Feature engineering aligns axes and flips negatives.
Models
- Decision Tree (depth=3, min_samples_leaf=100) for clear, explainable rules.
- Random Forest (200 trees, sqrt features) for bagged, stable probabilities.
- Gradient Boosting (lr=0.01) for refined, mid-range estimates.
What Stakeholders Learn
- Probability heatmaps coaches & execs can read in seconds.
- Trade-offs: bagging vs boosting; calibration & log-loss evaluation.
- Gaps to improve: defender context, richer spatiotemporal signals.
What You’ll Learn
Advance from foundational ML concepts to production-grade models. Each section builds career-ready skills that help employers deploy robust analytics pipelines and make smarter, data-driven decisions.
From Foundations to Pipelines
- Master ML fundamentals and apply them to structured sports datasets.
- Build full preprocessing & modeling pipelines with scikit-learn.
- Leverage EDA & feature engineering to drive model-ready insights.
Impact: Reduce manual workflows and standardize modeling across teams.
Modeling Mastery
- Regression to predict PGA drive distance and real-world KPIs.
- Tree-based models (Decision Trees, Random Forests, Gradient Boosting).
- Neural networks applied to NBA/ WNBA win probability models.
Impact: Equip your team with advanced predictive tools for any domain.
Performance & Scale
- Cross-validation & hyperparameter tuning for optimal models.
- Evaluate with advanced performance metrics tailored to stakeholders.
- Automate pipelines with PyCaret & streamline deployment.
Impact: Deliver models that are not only accurate but production-ready.
Applied Analytics
- Case studies: WNBA shot probability, NBA win probability, NFL field goal models.
- Incorporate GLMs, GAMs, and Bayesian approaches.
- Translate sports-driven insights into frameworks for finance, marketing, and beyond.
Impact: See direct ROI by adapting sports analytics rigor to your industry.
Course Overview
Advance from core ML concepts to production-grade pipelines. This compact overview spotlights the business value your team gains—then lets learners expand for lesson-level detail.
▶
1) Introduction to Advanced Machine Learning in Sports Analytics
Understand ML’s role, data landscape, and how this course maps to business impact.
stakeholder impactML landscapecourse roadmap
1) Introduction to Advanced Machine Learning in Sports Analytics
Understand ML’s role, data landscape, and how this course maps to business impact.
stakeholder impactML landscapecourse roadmap
▶
2) Linear Regression & The Machine Learning Pipeline
Build predictive foundations and production-style pipelines that scale across projects.
PGA drive distanceregularizationR² / MSE
2) Linear Regression & The Machine Learning Pipeline
Build predictive foundations and production-style pipelines that scale across projects.
PGA drive distanceregularizationR² / MSE
- Linear Regression & Least Squares
- Predicting PGA Tour Drive Distance
- Dataset Splits, Normalizing & Regularization
- Model Construction & Performance + Quiz
Outcome: re-usable pipelines → faster iteration, less risk of ad-hoc analysis.
▶
3) Exploratory Data Analysis & Feature Engineering
Turn raw data into signal with systematic EDA and features that boost lift.
visual analyticsfeature creationdomain signal
3) Exploratory Data Analysis & Feature Engineering
Turn raw data into signal with systematic EDA and features that boost lift.
visual analyticsfeature creationdomain signal
- Motivating Data Exploration & Visualization
- Feature Engineering (one-hot, scaling, interactions, polynomials)
- EDA & FE Quiz
Outcome: better features → simpler models, stronger accuracy, clearer storytelling.
▶
4) Hyperparameter Tuning & Model Selection
Pick the best model under time & compute constraints with rigorous validation.
CV / stratifiedgrid/random searchtrade-offs
4) Hyperparameter Tuning & Model Selection
Pick the best model under time & compute constraints with rigorous validation.
CV / stratifiedgrid/random searchtrade-offs
- Hyperparameters vs Parameters
- Cross-Validation Strategies
- Grid/Random/Automated Tuning
- Model Selection + Quiz
Outcome: repeatable model selection → confidence in decisions presented to stakeholders.
▶
5) Model Performance Evaluation
Measure what matters for the business: calibrated, comparable, defensible KPIs.
R² / RMSE / MAEPrecision/Recall/F1ROC-AUC
5) Model Performance Evaluation
Measure what matters for the business: calibrated, comparable, defensible KPIs.
R² / RMSE / MAEPrecision/Recall/F1ROC-AUC
- Performance Metrics & Criteria
- Exploring Model Behavior (residuals, importance, sensitivity)
- Identify Research Directions + Quiz
Outcome: avoid metric traps; align evaluation with product & stakeholder goals.
▶
6) Automating Machine Learning Pipelines
Standardize & scale with sklearn Pipelines and PyCaret; add monitoring.
sklearn pipelinesPyCaretmodel monitoring
6) Automating Machine Learning Pipelines
Standardize & scale with sklearn Pipelines and PyCaret; add monitoring.
sklearn pipelinesPyCaretmodel monitoring
- Why Pipelines
- scikit-learn Pipelines
- PyCaret + PGA drive distance case
- Model Monitoring + Quiz
Outcome: faster time-to-insight, fewer production defects, easier hand-offs.
▶
7) Generalized Linear Models (GLMs)
Model probabilities and counts with the right distributions & links.
Gaussian/Poisson/BinomiallinksNFL FG probability
7) Generalized Linear Models (GLMs)
Model probabilities and counts with the right distributions & links.
Gaussian/Poisson/BinomiallinksNFL FG probability
- Outcome Distributions & Link Functions
- NFL Field Goal Probability case study
- GLM Quiz
Outcome: interpretable probabilistic models trusted by non-technical leaders.
▶
8) Generalized Additive Models (GAMs)
Capture nonlinear structure with splines/basis functions; keep interpretability.
B-splinesbasis functionsNBA win probability
8) Generalized Additive Models (GAMs)
Capture nonlinear structure with splines/basis functions; keep interpretability.
B-splinesbasis functionsNBA win probability
- Motivation, Splines & Basis Functions
- NBA In-Game Win Probability
- GAM Quiz
Outcome: explain complex effects clearly to product & ops partners.
▶
9) Tree-based Methods & Ensembles
Deliver strong baselines at scale: trees, bagging, boosting, RF, GBM.
feature importancebias-varianceWNBA shot probability
9) Tree-based Methods & Ensembles
Deliver strong baselines at scale: trees, bagging, boosting, RF, GBM.
feature importancebias-varianceWNBA shot probability
- Decision Trees
- Bagging / Random Forest
- Boosting / Gradient Boosting Machines
- WNBA Shot Probability + Quiz
Outcome: robust accuracy with interpretable drivers; faster stakeholder adoption.
▶
10) Neural Networks
Model complex patterns with NN architectures and best practices.
FFNbackprophyperparametersNBA win prob
10) Neural Networks
Model complex patterns with NN architectures and best practices.
FFNbackprophyperparametersNBA win prob
- Architecture & Feedforward Networks
- Backpropagation & Training
- Hyperparameters + other NN types
- Revisiting NBA Win Probability + Quiz
Outcome: apply deep learning where it adds value—without over-engineering.
▶
11) Introduction to Applied Bayesian Statistics
Reason under uncertainty with priors, MCMC, and PyMC—then apply it.
Bayes’ theoremMCMCPyMCWNBA revisit
11) Introduction to Applied Bayesian Statistics
Reason under uncertainty with priors, MCMC, and PyMC—then apply it.
Bayes’ theoremMCMCPyMCWNBA revisit
- Bayes’ Theorem & Sampling (MCMC)
- Intro to PyMC
- Revisiting WNBA Shot Probability + Quiz
Outcome: calibrated probabilities & principled updates stakeholders can trust.
Meet Your Instructor
Patrick McFarlane — Director of Predictive Modeling
As the Director of Predictive Modeling for an MLB team, I build models that inform in-game strategy, player acquisition, and front-office decision-making. I’m bringing that same rigor to your learning journey in Python MachineZ.
- Engineer-turned-data scientist: BS Aerospace Eng. (Notre Dame), MS (MIT). Quantitative training that translates directly to reliable models and clear communication.
- Built predictive systems across aviation safety, finance, consulting, and now baseball—so examples go beyond sports and apply to any industry.
- We’ll move from clean data pipelines to tuned models and executive-ready results—so you can deliver value your stakeholders trust.
MLB Predictive Modeling
MIT — MS
Notre Dame — BS Aero
Sports & Industry Use-Cases
After jump-starting your Python journey with Evan in FoundationZ of Data Science, join me in Python MachineZ to take your machine learning skills to the next level.
What Industry Experts Are Saying About AthlyticZ
Discover what respected leaders in sports analytics and performance have to say about their experience with AthlyticZ:
"Here’s your chance to learn future data gurus!"
AthlyticZ has completely transformed the learning approach to data science through the use of sports-based problems. The course structure is intuitive, the content is comprehensive, the instructors are the best of the best, and the practical projects have immediate impact to students.
Bill Geivett, M.Ed.
President @ IMA Team
Author: Do You Want to Work in Baseball?
Sr. VP, Colorado Rockies (2001–2014)
Asst. GM, Los Angeles Dodgers (1998–2000)
Author: Do You Want to Work in Baseball?
Sr. VP, Colorado Rockies (2001–2014)
Asst. GM, Los Angeles Dodgers (1998–2000)
"Empowering and Engaging"
I've always been intrigued by the intersection of sports analytics and data science. AthlyticZ provides students with the perfect platform to explore this passion. The hands-on projects are particularly empowering, allowing them to apply theoretical concepts to real-world scenarios.
Brad Smith, PhD
Senior Performance Analyst @ University of Nebraska
Sports Analytics Professor, Data Science Program @ Northwestern
Player Development Analyst @ New York Yankees (2018–2021)
Sports Analytics Professor, Data Science Program @ Northwestern
Player Development Analyst @ New York Yankees (2018–2021)
How We Conduct Our Course
At AthlyticZ, we design for flexibility and impact—mixing clear instruction, continuous assessment, and applied projects so you can learn at your pace and show real results.
Prerecorded Video Lessons
Learn on your schedule with concise, high-quality videos you can pause, replay, and revisit anytime.
Assignments & Exercises
Reinforce concepts with practical exercises that build intuition and confidence in real workflows.
Continuous Assessment
Low-friction quizzes and checkpoints give you immediate feedback and keep you on track.
Hands-On Projects
Tackle real analytics problems—from quick wins to capstones—so you graduate with portfolio-ready work.
Supplementary Resources
Get curated readings, slides, and short references to deepen understanding and speed up execution.
Interactive Features
Use lesson-embedded notes, preloaded IDEs, and live widgets for an engaging, hands-on experience.
Python MachineZ
Course Information
- 50+ Lessons of engaging content
- Asynchronous, self-paced learning
- Pre-recorded video lessons for flexible scheduling
- Interactive Features: Quizzes throughout the course to reinforce learning
- Supplementary Resources: Access to downloadable slides, readings, and additional materials
- Hands-On Projects: Dive into major case study projects tailored to your interests in sports
- Virtual Machines: Immerse yourself in our customized IDE solutions for real-world programming
START FOR $139
Frequently Asked Questions
Answers to the most common questions about Python MachineZ.
1. Can I expense this course with my employer?
Absolutely. Employers benefit directly when their team members master machine learning. Completing this course helps you:
- Streamline Data Analysis: Build essential Python workflows that accelerate insights and improve decision-making.
- Optimize Workflows: Design tailored data pipelines that reduce manual effort and increase efficiency.
- Cost-Efficient Solutions: Replace expensive software with powerful open-source libraries for analysis and visualization.
If your employer is looking for team-wide upskilling, consider starting with our FoundationZ of Data Science course as a primer before advancing into Machine Learning.
2. Do I need access to any commercial software?
No. Everything in this course uses open-source Python tools and libraries—no additional software licenses are required.
3. Does the course assume extensive knowledge of Python?
This is an intermediate-to-advanced course. We recommend that students have completed our FoundationZ of Data Science course or have equivalent Python experience.
4. Is the course graded? What defines successful completion?
Evaluation is based on completion of coursework rather than “correct answers.” To successfully complete the course:
- Achieve 80% or higher on all quizzes
- Complete 100% of course modules
Upon meeting these requirements, you’ll receive a certificate of completion.
5. What if I need to drop out? Is there a refund policy?
Yes. Our refund policy is as follows:
- You may drop the course within 3 days of commencement for a full refund.
- No refunds are provided after 3 days.
- No refunds are granted if you complete 25% or more of the course material, regardless of timing.
We’re committed to supporting your success every step of the way.
© 2025 AthlyticZ
Designed With ❤️ By Jackson Yew
Privacy Policy · Terms of Use · Terms of Sale · Email questions to : [email protected]