FPL editorial experiment
I built the TallyKind FPL tools because I wanted a clearer way to test my own Fantasy Premier League decisions. This page shows how I use the tools together and what kind of impact they are designed to reveal.
Experiment premise
Follow David Slade's TallyKind FPL experiment, with weekly notes on decisions, transfers, captaincy and how Saka Potatoes is doing.
Available weeks
Why I built it
A lot of Fantasy Premier League advice gives a conclusion without enough context. I wanted tools that let me see the projected points, fixture shape, player role signals, expected minutes, form and risk notes behind a recommendation.
TallyKind's FPL tools are my attempt to make that process inspectable. They use public FPL data and modelled signals to create a shortlist, then leave room for human judgement, deadline news and strategy.
How I use it
I start with the Team Analyser to see where the squad stands, then use Team Planner for transfer and captain context. Best Squad gives a clean benchmark, and the Transfer Analyser looks backwards at whether previous moves actually helped.
That keeps the workflow practical: rate the squad, compare options, make a decision, then review whether the move added value.
Gameweek notes
The weekly notes are where the model output, real FPL result and decision review meet. The first published review looks at the move from GW4 to GW5.
Each review should stay grounded in actual tool output and public FPL results rather than turning into a generic gameweek diary.
Inputs and limits
The FPL tools use public manager, squad, player, price, availability, fixture, minutes, form, attacking, defensive, team-strength and market signals where available.
They cannot see private transfer plans, exact selling values, hidden chip intentions, late leaks, or whether I simply decide to ignore a sensible recommendation.
FAQs
The page is about using TallyKind's tools against real FPL decisions and reviewing the impact honestly, without inventing results or ranks.
No. The model creates recommendations and context, but final decisions can include deadline news, risk preference, chip strategy and human judgement.
Where a decision is reviewed, the misses matter as much as the wins because they show what the model or decision process did not capture.
Transparency
The useful part is seeing what the model emphasises, where human judgement changes the call, and what the outcome says afterwards.