Manny @TheNumbers_Game

Mateus Fernandes: Built For A Bigger Stage backseatballer.com/2026/06/10/mat…

⭐ Why Mateus Fernandes is Ready for a Big Move ⭐ - Actions across the pitch - High completion rates IP and OOP - An interesting stat about his importance to West Ham - New viz! New Article up now on the BackseatBaller website! Read below 👇🏾

With this, the player season dashboard and the key actions, I'm now much better placed to return to article writing. The end goal is still to make the generation of all of these visualisation accessible to all via web applications, but for now, I can focus on getting content out.

Finally got round to putting together passing dashboards, with: - Pass distribution densities - Pass Completion by length - Player percentiles (with distributions for added context) - Passing Sonars - Pass Recipients - Pass Map Hardest thing about this was tweaking the plot locations/sizes to make it all work together. Here's Matheus Fernandes, as an example:

Thoughts on van Hecke: 🟩 Strong in buildup. Probably the best in the league for first phase passing and carrying progression, and under De Zerbi specifically, this quality will be a fine asset to the team. Also useful in the final third as an extra support through marauding runs, which is nice. 🟥 I'm not too keen on his OOP numbers. His duel success isn't really where I'd want it to be (especially ground duels). His speed (or lack thereof) has been mentioned by others as well. Spurs would need to pair him up with a strong duelist, or at least someone quick, to cover his weaknesses (I'm not 100% sure on VDV) In any case, this signing makes sense, at the right price.

keyboard_arrow_left Previous keyboard_arrow_right Next

David Ornstein @David_Ornstein

a week ago

🚨 Tottenham Hotspur submit bid to sign Jan Paul van Hecke from Brighton & Hove Albion. #THFC offer for 25yo centre-back below #BHAFC valuation so rejected, but talks continue in bid to strike agreement. Dutch int’l close to Roberto De Zerbi @TheAthleticFC nytimes.com/athletic/73315…

1K 3K 35K 8.7M 598

Getting his key actions has made me adapt my clustering approach. I'm using silhouette scores to find the optimal cluster number to get good separation between clusters, avoiding having a large number of groups for the sake of it. Clusters will be larger, but more informative.

I'm a bit late to the party here, and it's unsurprising that I also think Fernandes is a Good Footballer™. If United want him, they should sign him quickly before shifting their focus onto Anderson/Tonali (who we probably don't need if we get Ederson)/Baleba and a LW.

Mateus Fernandes vs Leeds: West Ham will be playing Championship football next season, but Fernandes definitely won't be playing with them.

Finally got something going for player match reports. Here's Bruno Fernandes vs Brighton yesterday: (Also, more excitingly for me, I've managed to integrate xT into my dataset, giving me the scope for more in depth analysis in the future!)

after finishing university, i’m now seeking out opportunities in football! i’m open for roles in recruitment, opposition analysis and player identification. if you know someone who might be interested, or just want to connect, my DMs are open! see pinned for my portfolio!

I'm now looking at adding things like xThreat and VAEP to my data (depth), as well as expanding to other leagues across Europe (breadth). I may write an article about the work I've had to do, but for now, I'm excited to be back sharing analysis (hopefully with more frequency)!

Doing this was a massive learning opportunity, and allowed me to incorporate some of the things I've learnt since starting my full time job a couple months ago. Heck, I even tried a bit of prompt engineering and LLM work.

Well, after a couple of months deriving a bunch of metrics from first principles and building my own xG model, I think I can finally say I've managed to overcome the loss of FBref. Building this from scratch has also given me much more flexibility. More to come soon!

@chadha_ishdeep was who I was referring to in the above tweet (Twitter likes to mess up threads for me, sorry)

As bad as the situation is, I feel even more motivated to put more work in to provide a similar level of analysis as always, hopefully with even more frequency, and to branch out to more academic, data science heavy, projects. BackseatBaller isn't dead, is what I'm saying, haha.

For people scrambling for an alternative solution for the time being, I can point you towards @BeGriffis as well as @pranav_m28 and @chadha_is

That isn't to say that the democratisation of data isn't a key issue in the football analysis realm. The challenge for most aspiring analysts is "where do I get data?". This is especially the case when online portfolios are just as important as traditional education for clubs.

I see this development as a similar event to the cancellation of FM25 for FM content creators. The loss of one's main source of content leaves a lot of people in the wilderness. However, it provides the opportunity for innovation and creativity. I'll be working even harder now.

Best of Spalletti ball so far x.com/jaraaxz_comps/…

Diogo Dalot (24/25) vs Patrick Dorgu (24/25) - Player Radars #MUFC

keyboard_arrow_left Previous keyboard_arrow_right Next