Frank HopkinsinTowards Data SciencePropensity Score Matching (PSM) for A/B Testing: Reducing Bias in Observational StudiesA comprehensive guide to implementing PSM with your experimental data, including Python code·12 min read·Apr 26, 2023----
Frank HopkinsinTowards Data ScienceDetecting Sample Ratio Mismatch in your A/B TestsThis post helps you identify whether you have unbalanced cohorts in your digital experiments and incorporates some basic Python functions…·6 min read·Oct 20, 2021----
Frank HopkinsScraping FBRef to perform comparative football player analysisThis article breaks down some Python functions I wrote in order to compare Arsenal prospects and their attributes, using FBRef·5 min read·Jul 20, 2021--2--2
Frank HopkinsWhy did Arsenal perform so badly in 2020/21?Here I use some basic machine learning techniques with 2020/21 English Premier League (EPL) data to try to get to the bottom of Arsenal’s…·11 min read·Jul 2, 2021----
Frank HopkinsinThe StartupPredicting Arsenal Fixtures with Elo ScoresHow Premier League Elo scores can be used to predict the outcome of fixtures, using classification models in Sklearn·8 min read·Jan 31, 2021----
Frank HopkinsinTowards Data ScienceBayesian A/B Testing with Continuous Variables — including Python CodeThis article discusses how Bayesian estimation can be implemented in your digital experimentation methodology, with a specific focus on…·8 min read·Nov 4, 2020----
Frank HopkinsinTowards Data ScienceCUPED R-Shiny ToolIntroduction to CUPED·4 min read·Oct 11, 2020--1--1
Frank HopkinsinBBC Data ScienceIncreasing experiment sensitivity through pre-experiment variance reductionThis article discusses how the Experimentation team have been accounting for pre-experiment variance in order to increase the statistical…·6 min read·Sep 24, 2020--4--4
Frank HopkinsinTowards Data ScienceUsing Dot Plots With Experimentation DataThis article will show how dot plots can be used to translate your experimentation findings in a digestible and jargon-free manner·5 min read·Apr 21, 2020----
Frank HopkinsinBetter ProgrammingHow to Build Recommendation Models With MyAnimeList and Sklearn (Part 2)Simple techniques that can be used to recommend anime content based on user-rating correlations and feature variables·9 min read·Apr 3, 2020--1--1