Agentic AI · Full-Stack · Claude

Every FPL decision,
answered at your fingertips.

Eleven million people play Fantasy Premier League, and every week each one faces the same questions: who to transfer, who to captain, when to play a chip. Pocket Scout reads your actual squad and answers all of them — personalized advice for your team, your budget, your gameweek. And because every call is grounded in real FPL data, it's advice you can actually trust.

11M+
FPL managers
facing the same calls
0.53
Model rank correlation
(vs FPL's own 0.59)
10
Seasons of data
backtested
0
Invented stats —
grounded in real data
Why trust it

Advice is grounded in live FPL data through a deterministic scoring engine — no invented stats — and the model's own rank correlation is reported honestly: 0.53 vs FPL's 0.59.

The problem

Three hard calls, every single week.

Every gameweek, eleven million managers stew over the same three calls — who to transfer, who to captain, when to play a chip — armed with a dozen open tabs and a group chat full of opinions. Most FPL chatbots make it worse: they happily hallucinate prices, form, and fixtures. I wanted one advisor that knows my actual squad and budget, and never invents a stat to sound confident.


How it works

Hard numbers first. The pundit second.

Enter your FPL manager ID and Pocket Scout reads your real squad — players, bank, and the free transfers you're holding. The hard figures are computed before any language model is involved: a deterministic engine rates every player 0–10 and proposes transfers, captaincy, and chips straight from the FPL API, then Claude reasons over those numbers to explain the call — fetching real stats through tool calls, and saying so when the evidence isn't there. It was built eval-first: the ranking model was fitted on a ten-season backtest, reaching a rank correlation of 0.53 against FPL's own 0.59 benchmark — reported honestly, not rounded up — and a 331-test suite plus squad replays keep behaviour pinned, which is why a hold gate now banks a free transfer when no move clears the bar. The chat also treats every message as a question to answer, not an instruction to obey — it can't be talked off its own plan.

📊
Base phase — deterministic
Rates every player 0–10 and proposes transfers/captain instantly from FPL data. No LLM in the loop.
🧠
Insights phase — Claude
Reasons over the computed numbers to explain the advice, cached for speed and cost.
Personalized verdict
A single, actionable call on transfers, captaincy, and chips — tailored to your squad and bank.

Demo

The two-minute recording below is the whole loop: squad rated, verdict banner, a transfer plan across This Week, Long Term, and Chips, then a question to the Scout. No spreadsheets, no twelve open tabs.

Full walkthrough — rate your squad, plan transfers, ask the Scout▶ 2-min demo

Reflection
Behind the build

I built Pocket Scout because I'm one of those eleven million managers, and the weekly calls are genuinely hard and genuinely fun to argue about. It grew out of an earlier English Premier League quant analysis I built on Azure AI Hub Prompt Flow, pairing statistical analysis with LLM-generated SWOT reports. The part I'm proudest of is letting evaluation overrule intuition — more than once the data killed a change that felt obviously right, and the eval harness got the final word, not me. That's the same bar I hold every project on this site to.

See it live — or dig into the build.