
0/14 An Application of Stochastic Programming to FPL: robust planning under fixture uncertainty using Linear Optimisation with Recourse ⤵️

Probability nerd, building @SolioAnalytics @trout-post.com on bsky

0/14 An Application of Stochastic Programming to FPL: robust planning under fixture uncertainty using Linear Optimisation with Recourse ⤵️

🎥 #FPL OPTA-MISE [NEW OPTA FEATURES] We are very excited about our latest release, now live on Solio Analytics! In this video, @FPL_Spaceman, @FF_Trout and @robtFPL demo Solio's new Opta data toolkit to help you get started and show you around 📊 youtu.be/luyGYkqYSKo

This is the first thing we've built using this new foundation. It won't be the last!

It started from a vision @robtFPL had several months ago: Opta data is incredibly detailed, so it should be possible to use it to answer any question that we could ask. Over the last few months, we built an ambitious new data engine on top of Opta data to make this possible.

Our new Opta data interface is the biggest, coolest and certainly most complex thing we've ever built at Solio. I think it's the most capable and joyful interface for consuming football data. We hope Solio Pro users enjoys it!
We've built a completely new way to interact with the Opta data on Solio 👀 Detailed Player data and Team data sections, centred around one idea: the slice. Every stat, on every page, filtered to any football situation you choose 👇

New feature coming? Video out tomorrow... @FF_Trout is excited

Can The Eye Test Survive Against The FPL Solvers? 📝🎯 FPL 2026-27 Watch: youtube.com/watch?v=4fLGyp22… Going live RIGHT NOW with @FPLD_Unit and @FF_Trout


Running the calculation this year again, we have a fairly low value for team value increases in the current state of play - only ~0.21pts per gameweek for every extra £1.0m of team value. Historically we have seen this number often nearer ~0.3 If we were to assume Bruno Fernandes is never viable and Saka remains injured for the rest of the season (extreme assumptions, for illustrative purposes only), then this value does drop a little, but only to ~0.19 The point at which team value is meaningless tops out a little lower there though at 114.0m, instead of 119.6m
I posted this already. My question last year you made a graph for how "valuable" team value was. Do you have something similar for how this year is shaping up?

Technically if you want to increase your chances of getting luckier, then you want to widen your distribution of outcomes, such that the odds of getting really lucky increases. You can do that by sacrificing some projected points for a more differential team. Thus, you are more likely to get really lucky! (See how the yellow distribution in the image has a fatter right tail, depsite a lower mean score) The downside is that you have also increased your chances of getting unluckier.
How do I get luckier?

Do you have an #FPL question for GW9 or beyond? @FPL_Spaceman will be answering questions here on Twitter for two hours later today. Just reply to this post with your question and let Jonny help you out! Once Jonny has finished answering, we’ll randomly draw three winners, each receiving one free month of Solio Pro. This Q&A replaces this week's pod, which we've had to cancel due to scheduling conflicts.

On a simple 6GW sample, removing a player's best xG score decreases accuracy rather than improving it. There is some case to be made for down-weighting outliers, but throwing them out entirely is also throwing out useful signal, so be careful
I agree hes a great pick. But i do think the data needs to come with a caveat of the ipswich game inflating it. 8th for Xg per 90 for midfielders if you remove it

When I'm not doing FPL I'm fortunate to work on some extremely fun things! (@attio) Excited to show off all the really cool stuff we've been building 😁 (with an equally cool launch site) attio.com/architecture
A new type of user requires a new type of architecture. Legacy software wasn't built for agents and now they're breaking it. Today, we're launching the next generation of our platform powered by a hybrid architecture. Building this has been the most ambitious and challenging engineering challenge of my career. Attio now supports SQL - offering a PostgreSQL compatible connector and syntax that allows our customers and their agents to query all of their data like a database. It's transactionally consistent, horizontally scalable and allows agents to access the full context they require using a language and syntax that they're well trained on using. Whether you're accessing your data through our UI, our API, our MCP server or now our SQL Data Connectors: your team and your agents get the full context, every time. Learn more at attio.ai/architecture

Wildcard active? With the help of @SolioAnalytics we answer six of the biggest FPL questions ahead of GW6, including: ✅ DEFCON trends ✅ Set-piece importance ✅ Fixture swings ✅ Defensive rotations ✅ Too late to buy Pascal Gross? ✅ Budget forwards theanalyst.com/articles/defc…

Every FPL manager has agonised over a benching decision, and then end up starting the blank and benching the haul. So does rotation actually work? We checked @SolioAnalytics's deadline-locked projections for every 2025-26 gameweek against what actually happened 🧵

We spent the international break improving Solio's solver engine 🔧 Solves are now much faster, and they search deeper than ever. Plan several #FPL gameweeks ahead in seconds Throw your hardest plans at it & let us know if you can see the difference! 👇 fpl.solioanalytics.com