I'm glad to share that I've successfully defended my Ph.D. thesis in Artificial Intelligence, under the inspiring direction of @LukeBornn and the support of @FCBarcelona coaches.
Here's a thread with details of our quest for a comprehensive analysis framework (toolbox) in soccer
Javier Fernández
667 posts
@ChelseaFC. Formerly @ZelusAnalytics, @FCBarcelona. Views are my own. At BlueSky as @javiondata.bsky.social
Joined November 2016
- Last week, @LukeBornn, @dcervone0, and I uploaded to ArXiv the full technical version of our expected possession value (EPV) approach for soccer, using tracking-data. I start a thread highlighting the most relevant parts of the paper. arxiv.org/abs/2011.09426
- Good news! The full document of my Ph.D. thesis it's now available at hdl.handle.net/10803/673529 You can read more about the details of our work in this thread x.com/JaviOnData/sta… Below, a quick description of which chapters to read depending on your interest and time.I'm glad to share that I've successfully defended my Ph.D. thesis in Artificial Intelligence, under the inspiring direction of @LukeBornn and the support of @FCBarcelona coaches. Here's a thread with details of our quest for a comprehensive analysis framework (toolbox) in soccer
- 🚨 We're hiring at Chelsea FC!🔵 🚨 This is a unique opportunity to join our analytics team and be at the heart of decision-making at a fantastic club. Full Stack Web Developer: linkedin.com/jobs/view/4178… Data Engineer: linkedin.com/jobs/view/4178… Full post: linkedin.com/posts/javionda…
- Some personal news: I'm thrilled to announce that I'm joining the @ZelusAnalytics team. I feel honored to be part of one of the most exciting sports analytics' projects and groups of professionals I've ever seen.
- Next Tuesday 15th @LukeBornn and I will be presenting our latest paper at ECML-PKDD 2020 conference. We introduce SoccerMap, a flexible deep learning architecture for predicting full probability maps from raw tracking-data.
- We are excited to share our most recent work, where we propose a rich tactical framework for assessing the real impact of physical effort in soccer. arxiv.org/abs/2204.02313
- We are happy to see two of our papers accepted for this year's @SloanSportsConf. We continue to stand on past research to bring some new ideas: the concept of positional advantage and EVA, and the estimation of the potential impact of off-ball runs #SSAC20 @BarcaInnoHub.
- I'll be speaking tomorrow at @StatsBomb first conference, talking about "Finding the free man". In a nutshell: lots of videos, talking more about football than science, exploiting EPV to the fullest, and presenting some new off-ball stats. Talk is at 11:30am. (Stamford bridge 🙃)
00:00 - Join us this Friday at the Sports IDEAS symposium. I'll discuss with @StatsbyLopez @nickwan and @MMattredfieldhow how we use tracking data at FC Barcelona to drive impact: from in-depth game analysis to player-customized video feedback. Register here: us02web.zoom.us/webinar/regist…
- Today at 2:00 pm in room 311 #SSAC19 l'll be presenting our approach to add a qualitative approach to decision-making in soccer, and improve interpretability. In other words join if you would like to see some nice video overlays and discuss about decomposing soccer in parts
- If you are into Sports Analytics, here's an idea: join the bright side of constructive feedback. It is important to understand the limitations of a model, but don't invest so much time in pointing out those, and rather spend your energy in thinking: how can this be useful?
- We are happy to attend #SSAC18 this year and present our work on space creation in soccer. Take a look to the paper (sloansportsconference.com/wp-content/upl…). Maybe Messi wasn't walking after all, but adding spatial value. See you all there.

















