Football Match Intelligence is a Streamlit app that converts SkillCorner tracking plus event data into interactive tactical analysis. You can explore match overviews (pass networks, shot maps, momentum and xG), dive into player profiling (heatmaps, speed zones and movement patterns), analyze team structure (field tilt, defensive organization and possession chains), compare players head-to-head with pizza charts, and reconstruct key events frame by frame with a tactical board that highlights passing options. Live app: https://lnkd.in/eChbEpud Repository: https://lnkd.in/e-drHUD2

𝐇𝐨𝐰 𝐈 𝐥𝐞𝐚𝐫𝐧𝐞𝐝 𝐭𝐨 𝐰𝐨𝐫𝐤 𝐰𝐢𝐭𝐡 𝐟𝐨𝐨𝐭𝐛𝐚𝐥𝐥 𝐭𝐫𝐚𝐜𝐤𝐢𝐧𝐠 𝐝𝐚𝐭𝐚 (𝐚𝐧𝐝 𝐰𝐡𝐚𝐭 𝐢𝐭 𝐥𝐞𝐝 𝐭𝐨) A few months ago, I realized something: If you want to understand football tactics, event data isn’t enough. You have to learn from space, motion, and interaction. That curiosity sent me deep into research on transformers, graph neural networks, and spatiotemporal modeling — and directly led me to build my first transformer-based similarity retrieval system for the SkillCorner × PySport Analytics Cup. The system learns from full-pitch scenes (22 players + ball) to: • retrieve tactically identical plays • cluster recurring patterns • automatically classify team actions and sequences • make visual insights easier for analysts and coaches to use Papers that shaped my thinking 👇 1️⃣ A Graph Neural Network deep-dive into successful counterattacks Joris Bekkers Amod Sahasrabudhe https://lnkd.in/gnDwBm5a The purpose of this research is to build gender-specific Graph Neural Networks to model the likelihood of a counterattack being successful and uncover what factors make them successful in professional soccer. 2️⃣ Graph representations for the analysis of multi-agent spatiotemporal sports data Dominik Raabe Prof. Dr. Daniel Memmert Reinhard Nabben https://lnkd.in/g36xemtf Proposed Tactical Graphs, an alternative graph-based format capable of producing integrative, contextualized models for machine learning applications 3️⃣ 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧 𝐨𝐟 𝐭𝐚𝐜𝐭𝐢𝐜𝐚𝐥 𝐩𝐚𝐭𝐭𝐞𝐫𝐧𝐬 𝐮𝐬𝐢𝐧𝐠 𝐬𝐞𝐦𝐢-𝐬𝐮𝐩𝐞𝐫𝐯𝐢𝐬𝐞𝐝 𝐠𝐫𝐚𝐩𝐡 𝐧𝐞𝐮𝐫𝐚𝐥 𝐧𝐞𝐭𝐰𝐨𝐫𝐤𝐬 Pascal Bauer Gabriel Anzer https://lnkd.in/gvy9VyBe Presenting practical applications of approach using the detection of overlapping runs as a showcase 4️⃣ 𝐒𝐞𝐥𝐟-𝐒𝐮𝐩𝐞𝐫𝐯𝐢𝐬𝐞𝐝 𝐑𝐞𝐩𝐫𝐞𝐬𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧𝐬 𝐟𝐨𝐫 𝐓𝐫𝐚𝐜𝐤𝐢𝐧𝐠 𝐃𝐚𝐭𝐚 Karun Singh https://lnkd.in/gsKbiCrF Building tools that leverage tracking data to help accelerate video analysis workflows. 5️⃣ 𝐓𝐞𝐦𝐩𝐨𝐫𝐚𝐥 𝐆𝐫𝐚𝐩𝐡 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐅𝐫𝐚𝐦𝐞𝐰𝐨𝐫𝐤 𝐟𝐨𝐫 𝐐𝐮𝐚𝐧𝐭𝐢𝐟𝐲𝐢𝐧𝐠 𝐏𝐚𝐬𝐬 𝐑𝐞𝐜𝐞𝐩𝐭𝐢𝐨𝐧 𝐏𝐫𝐨𝐛𝐚𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬 𝐀𝐠𝐚𝐢𝐧𝐬𝐭 𝐃𝐞𝐟𝐞𝐧𝐬𝐢𝐯𝐞 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐬 Pegah Rahimian Laszlo Toka https://lnkd.in/gpAW5DEC Proposed a framework for evaluating passing decisions against defensive structures using temporal graph networks