iiinova — where your digital self keeps evolving

Martin Eastwood @martineastwood.co.uk · 09/06/2026

Get started with: pip install -U penaltyblog Read the full release notes and documentation: penaltyblog.readthedocs.io/en/latest/ As always, feedback, bug reports, PRs and feature requests are welcome.

penaltyblog.readthedocs.io

penaltyblog: Football Data & Modelling Made Easy — penaltyblog documentation

031 3

Martin Eastwood @martineastwood.co.uk · 09/06/2026

🚀 penaltyblog v1.11.0 is out. This release adds: ⚡ predict_many() for fast batch forecasting 🏟️ Neutral venue support 📈 A bundled pre-trained xT model 🚄 ~4× faster Weibull Copula fitting 🤖 Context7 MCP documentation support

151 5

Martin Eastwood @martineastwood.co.uk · 27/03/2026

Looking forward to the Field Of Play Conference today. If you’re there as well then come over and say hi 👋

020 2

Martin Eastwood @martineastwood.co.uk · 28/02/2026

🚀 penaltyblog v1.9.0 is live! 🐍⚽️ New in this version: ✅ create_dixon_coles_grid(): Use lambdas from external ML models. ✅ goal_expectancy_extended(): Infer rho/lambdas from odds. 🛠️ Improved quarter lines (2.25/2.75) logic for totals market. ⚡️ More optimisations to FootballProbabilityGrid

A Python code snippet demonstrating the new 'create_dixon_coles_grid' function in the penaltyblog library. The code shows how to initialize a probability grid using expected goals (lambdas) for the home and away teams, and then calculate home win probabilities, over/under 2.5 goal totals, and Asian Handicaps directly from that grid.

090 9

Martin Eastwood @martineastwood.co.uk · 08/01/2026

Why build a custom MCMC sampler from scratch? In my latest blog post, I dig into the "why" and the "how". From the decision to drop heavy dependencies like Stan for this use case, to using Cython for speed, and how to interpret those "fuzzy caterpillar" plots. Read it here: pena.lt/y/2026/01/08...

A technical MCMC Trace Plot with two horizontal subplots labeled "attack_Arsenal" and "home_advantage." Each plot shows four overlapping colored lines (Chains 0 to 3) creating a dense, horizontal "fuzzy caterpillar" shape. This visualization indicates that the Bayesian model has successfully converged and is stable across 2,000 iterations.

120 2

Martin Eastwood @martineastwood.co.uk · 08/01/2026

What can you do with v1.8.0? ✅ Bayesian Dixon-Coles: Quantify uncertainty, not just point estimates ✅ Hierarchical Models: Learn league-wide variance to handle small sample sizes ✅ Built-in trace plots: Ensure model convergence ✅ Familiar API: Works seamlessly with existing scrapers and weights

110 1

Martin Eastwood @martineastwood.co.uk · 08/01/2026

Bayesian goal models are back in penaltyblog v1.8.0 - and this time, they’re dependency-free! ⚽️📈

A dark-themed code editor window showing Python code using the penaltyblog library. It demonstrates initializing a HierarchicalBayesianGoalModel with home/away goals and teams, fitting the model using MCMC sampling with parameters like 2000 samples and 4 chains, and finally predicting match probabilities for an Arsenal vs. Manchester City fixture

1133 13

Martin Eastwood @martineastwood.co.uk · 29/12/2025

⚽ penaltyblog v1.7.1 is now available: Added params_array and param_indices functions to goal models to make it easier to work with the model's parameters. Thank you to Sebastian Velandia for this contribution! pip install --upgrade penaltyblog

010 1

Martin Eastwood @martineastwood.co.uk · 09/12/2025

📝 A couple of important notes: You'll need your own Opta subscription/credentials to use this (I provide the tools, not the data!) 🔑 As this is a v1 release of the integration, there's likely some rough edges. If you run into any, let me know - I'm happy to work with you to improve it. 🤝

000 0

Martin Eastwood @martineastwood.co.uk · 09/12/2025

⚽ penaltyblog v1.7.0 is now available: Direct integration with the Opta (Stats Perform) API. You can now stream matches & events lazily without downloading the JSON first. Includes helpers for human-readable filtering (no more memorising IDs). pip install --upgrade penaltyblog

Code example showing how the penaltyblog python package can now integrate directly with the StatsPerform Opta API to download data directly.

160 6

Martin Eastwood @martineastwood.co.uk · 04/12/2025

Have been doing Advent of Code in Nim and Kotlin this year and enjoying learning both. I can see Nim becoming one of my favourite languages!

010 1

Martin Eastwood @martineastwood.co.uk · 24/10/2025

Thanks Nils, it’s a good point and is on my todo list to dig further into how long it takes for something like FSAA to stabilise to something useful early on in a player’s career. Even with a fairly wide HDI at that stage, there’s still potentially benefits to its use

010 1

Martin Eastwood @martineastwood.co.uk · 20/10/2025

New article: "Shrinkage, Uncertainty, and Son Heung-min: Using Bayesian Methods to Identify Finishing Ability" which discusses using a Bayesian hierarchical approach to quantifying player finishing ability, with credible intervals to express uncertainty. pena.lt/y/2025/10/01...

pena.lt

Shrinkage, Uncertainty, and Son Heung-min: Using Bayesian Methods to Identify Finishing Ability

Why most finishing metrics are flawed and how a Bayesian approach gives us a truer picture of a player's finishing ability...

3235 23

Martin Eastwood @martineastwood.co.uk · 17/10/2025

🎉 penaltyblog v1.6.1 is out! ✨ What's new: - Python 3.14 support - scipy 1.16+ compatibility - Better numerical stability for Negative Binomial model - New Colab notebook for implied probabilities example pip install --upgrade penaltyblog

043 4

Martin Eastwood @martineastwood.co.uk · 07/10/2025

Absolutely. Also, people started sharing big (for then anyway) pre-trained networks that made it much easier to get started. I built many models in my day job back then by fine tuning BERT and ImageNet that I would have struggled to train from scratch without investing massively in compute.

000 0

Martin Eastwood @martineastwood.co.uk · 07/10/2025

We can also split Massey Ratings into attack & defence: 🔴 LFC: best attack in the league paired with a mid-table defence ⚪️ ARS: Elite at both ends. They have the #1 defence and the #3 attack 🌳 Forest: A disaster at both ends of the pitch

0101 10

Martin Eastwood @martineastwood.co.uk · 07/10/2025

Here's how the Premier League table really looks according to Massey Ratings, which account for strength of schedule. 📈 Arsenal (+1.5) & Man City (+1.3) are clear strongest overall 😬 Man Utd (-0.2) rank in the bottom half 📉 West Ham & Forest (-1.2) are worst teams by far

Massey Ratings for the English Premier League

3144 14

Martin Eastwood @martineastwood.co.uk · 23/09/2025

Thanks to everyone who suggested features and reported issues. Your input shapes the package's development. Questions or feedback welcome at pena.lt/y/contact Install: pip install penaltyblog GitHub: github.com/martineastwo...

000 0

Martin Eastwood @martineastwood.co.uk · 23/09/2025

📚 Interactive Colab notebooks are available in the docs - experiment with real examples without any local setup. I'll be steadily expanding these over the coming weeks to cover all functionality in the package. Docs: penaltyblog.readthedocs.io

penaltyblog.readthedocs.io

penaltyblog: Football Data & Modelling Made Easy — penaltyblog documentation

100 1

Martin Eastwood @martineastwood.co.uk · 23/09/2025

🔧 Improved implied odds module: - New logarithmic overround removal method for better accuracy - Structured results instead of raw arrays - Better handling of edge cases Making it easier to work with bookmaker probabilities in your analyses.

100 1

Martin Eastwood @martineastwood.co.uk · 23/09/2025

💰 Expanded betting utilities: - Kelly Criterion for multiple outcomes - Arbitrage opportunity detection - Value bet identification - Hedge bet calculations - Odds format conversion (decimal/fractional/American) All functions now return structured outputs for easier integration.

100 1

Martin Eastwood @martineastwood.co.uk · 23/09/2025

⚽ penaltyblog v1.6.0 is now available! 📊 MatchFlow updates: - SQL-style joins for nested football data (left, right, outer, inner, anti) - Cloud storage support: read/write directly to AWS S3, Google Cloud Storage, and Azure Blob - Automatic type inference for join keys pip install penaltyblog

110 1

Martin Eastwood @martineastwood.co.uk · 18/08/2025

🚀 New article on my blog walking through the latest updates to the penaltyblog python package for football (soccer) analytics & betting ✅ New interactive pitch plots ✅ 5-10× faster goal models ✅ New Flow query DSL 👉 pena.lt/y/2025/08/14...

pena.lt

Penaltyblog v1.5.0: Faster Models, Smarter Queries, and a Sharper Edge

v1.5.0 delivers interactive charts, faster models, upgraded football probability grid, and a powerful Flow query language - all designed to make your analysis sharper and quicker...

050 5

Martin Eastwood @martineastwood.co.uk · 15/08/2025

📚 Docs: penaltyblog.readthedocs.io/en/latest/in... 💻 GitHub: github.com/martineastwo... 🐍 pip install penaltyblog Feedback welcome, let me know what you build!

penaltyblog.readthedocs.io

penaltyblog: Football Data & Modelling Made Easy — penaltyblog documentation

010 1

Martin Eastwood @martineastwood.co.uk · 15/08/2025

🔍 New: Flow Query DSL Filter datasets with safe, Pythonic expressions: - AST-parsed (no eval) - Variables, regex, dates - Access nested fields

Image of code showing examples of penaltyblog Flow's new DSL

100 1

Martin Eastwood @martineastwood.co.uk · 15/08/2025

📈 Goal models are now 5-10× faster - Cython-powered analytical gradients for speed + stability - Fine-tune with minimizer_options:

Image of code demonstrating the the features in penaltyblog's goals models

110 1

Martin Eastwood @martineastwood.co.uk · 15/08/2025

⚽ New: Pitch Plotting API Build interactive football visualisations with: - Multiple layouts & themes - Scatter, heatmaps, arrows, comets - Custom hover tooltips

GIF demonstrating the interactive tooltips in penaltyblog's new Pitch plotting API

100 1

Martin Eastwood @martineastwood.co.uk · 15/08/2025

🚀 penaltyblog v1.5.0 is here! ✅ Interactive pitch plots ✅ 5-10× faster goal models ✅ New Flow query DSL What’s new 👇

142 4

Martin Eastwood @martineastwood.co.uk · 16/07/2025

📊 New article on my blog: How Accurate Are Soccer Odds? A Data Dive into 250 Million Betting Lines 🔍 How sharp are different bookmakers? 📈 How accurate are bookmaker's odds? 🎯 Are the odds well-calibrated? ➡️ pena.lt/y/2025/07/16...

pena.lt

How Accurate Are Soccer Odds? A Data Dive into 250 Million Betting Lines

A data-driven deep dive into how accurately bookmakers price global soccer markets...

0164 16

Martin Eastwood @martineastwood.co.uk · 19/06/2025

🚀 penaltyblog v1.4.0 is out! Now includes a query plan optimiser for smarter Flow pipelines: • Optional FlowOptimizer for smart rewrites (optimize=True) • New .plot_plan() for pipeline viz • .with_schema() for field validation • Rolling- and time-based summaries github.com/martineastwo...

github.com

GitHub - martineastwood/penaltyblog: ⚽ High-performance football analytics toolkit: build data pipelines, scrape data, model matches, rank teams, and bet smarter | Powered by pena.lt/y/blog 🚀

⚽ High-performance football analytics toolkit: build data pipelines, scrape data, model matches, rank teams, and bet smarter | Powered by pena.lt/y/blog 🚀 - martineastwood/penaltyblog

140 4

Martin Eastwood @martineastwood.co.uk · 25/05/2025

If you're interested in finding out more: Why I built it 👉 pena.lt/y/2025/05/25... Docs 👉 penaltyblog.readthedocs.io/en/latest/in...

pena.lt

Introducing MatchFlow: a JSON-native query engine for football data.

MatchFlow is a JSON-native query engine for football data - no flattening, no fuss...

010 1

Martin Eastwood @martineastwood.co.uk · 25/05/2025

At the heart of matchflow is the Flow class - a lazy, composable way to work with nested football data. You can filter, group, and summarize JSON-like data without flattening or loading everything into memory. 👇 Example:

Python code showing how to use the Flow class from penaltyblog.matchflow: it loads a folder of JSON files, filters for shots, groups by player name, and summarizes shot counts.

110 1

Martin Eastwood @martineastwood.co.uk · 02/05/2025

Thanks! Just to remove another variable - by simulating the result I definitely know which forecast is better as the simulated result is sampled from that particular forecast. When using the actual result you could argue that the true probabilities it's sampled from are unknown.

030 3

Martin Eastwood @martineastwood.co.uk · 16/04/2025

Thanks Paul. Good question, in theory with enough cross league connectivity you have one global Pi system and it all converges. Ratings are then relative to average team in your global pool so can compare across leagues. Is on todo list to see how well this performs compared to other approaches.

100 1

Martin Eastwood @martineastwood.co.uk · 14/04/2025

New on my blog: Pi Ratings — a football-first alternative to Elo. ✅ Handles draws ✅ Uses score margins ✅ Beats Elo on predictions Includes code, benchmarks & more: 👉 pena.lt/y/2025/04/14...

pena.lt

Pi Ratings: The Smarter Way to Rank Football Teams

A smarter, football-focused alternative to Elo — using Pi Ratings to track team strength and predict matches...

1174 17

Martin Eastwood @martineastwood.co.uk · 10/04/2025

penaltyblog v1.2.0 is released! ✅ Elo Ratings now football-specific (home advantage + draw probabilities) ⚡️ Fast Cython implementations of Ignorance Score & Multiclass Brier Score 📝 Updated docs / examples for Elo, Pi Ratings, RPS & more github.com/martineastwo...

github.com

GitHub - martineastwood/penaltyblog: ⚽ High-performance football analytics toolkit: scrape data, model matches, rank teams, and bet smarter | Powered by pena.lt/y/blog 🚀

⚽ High-performance football analytics toolkit: scrape data, model matches, rank teams, and bet smarter | Powered by pena.lt/y/blog 🚀 - martineastwood/penaltyblog

0133 13

Reposted by Martin Eastwood

Nils Mackay @nilsmackay.bsky.social · 01/04/2025

[NEW] First blog in a long time! In which I show a pretty straightforward way to derive teams' attacking and defensive strengths from bookmaker odds. This method is not only easy to implement, but also very accurate compared to results-based models. mackayanalytics.nl/2025/04/01/g...

2347 34

Martin Eastwood @martineastwood.co.uk · 18/03/2025

Looking forward to the Field Of Play Conference today. If you’re there as well today then come over and say hi 👋

010 1

Martin Eastwood @martineastwood.co.uk · 16/03/2025

⚽ New blog post - which football prediction model is the best? Poisson, Weibull, or Negative Binomial — each has its strengths and weaknesses. Learn how to optimise them for accurate match predictions using my penaltyblog Python package. pena.lt/y/2025/03/10... #FootballAnalytics #DataScience

pena.lt

Football Prediction Models: Which Ones Work the Best?

Comparing football goals models — Poisson, Dixon-Coles, and more — to see which predicts best and how to optimize them...

0257 25

Martin Eastwood @martineastwood.co.uk · 16/03/2025

That wasn't the answer I'm looking for, stop encouraging me 😅

010 1

Martin Eastwood @martineastwood.co.uk · 16/03/2025

I may just create a separate package for them for people willing to fight with the dependencies. I'm also trying really hard to talk myself out of building my own lightweight Bayesian framework for them 😆

100 1

Martin Eastwood @martineastwood.co.uk · 16/03/2025

Main issue is that pymc and stan are such heavy dependencies that all the Bayesian stuff became too difficult to share as users struggled to install them. Plus, pymc and Stan's dependencies were then so strict that it was difficult to install other packages alongside them. Wasn't worth the hassle.

200 2

Martin Eastwood @martineastwood.co.uk · 15/03/2025

Also, v1.1.0 is the first Cythonized penaltyblog release using wheels across multiple platforms. This should now make installation smoother on Windows, macOS, and Linux. However, if you run into any issues installing, let me know so I can look into it. Thanks!

000 0

Martin Eastwood @martineastwood.co.uk · 15/03/2025

penaltyblog v1.1.0 is released! 🔥 250x faster models with Cython 📊 New models: Negative Binomial, Weibull, ZIP & Bivariate 🔢 Pi Ratings System for ranking teams 🔗 Full changelog: penaltyblog.readthedocs.io/en/master/changelog/index.html #Python #SportsAnalytics #MachineLearning

penaltyblog.readthedocs.io

Changelog — penaltyblog documentation

2222 22

Martin Eastwood @martineastwood.co.uk · 27/02/2025

@pysport.org Looks like you've got the wrong Martin on your website as the contributor for my penaltyblog python package - opensource.pysport.org/project/pena...

opensource.pysport.org

000 0