Differential Radar
Under 10% owned, ranked by chance of a goal or assist.

Why him
0.53 xGI per 90 over his last 540 minutes (0.33 xG, 0.19 xA); the model uses 0.60 after blending with his 2025/26 numbers; at home to NFO; started 4 of his last 4, averaging 90 minutes; first choice on penalties.
The top 10 at a glance
9 more differentials are waiting
Top 10 by position
What a 50% chance looks like
A return in about 5 of 10 such games. A blank in the other 5.
Track record
Live picks, saved before each deadline
Backtest, GW1 to GW8 not live picks
Share of picks that returned a goal or assist. "Form" is simply the best-form players under 10% owned. Six gameweeks is a small sample and the model was adjusted after looking at them, so the live record is the one that counts.
How this works
- The chance comes from each player's expected goals and assists per 90 over recent gameweeks, blended with last season, and from the minutes he has been playing.
- Only regular starters who are not flagged as injured are considered.
- The opponent is not part of the number yet: in the gameweeks played so far, adjusting for the fixture made the chances less accurate.
- Not included: team news after the picks were saved, rotation that recent minutes do not show, and clean-sheet or defensive points.
- Data: official Fantasy Premier League API. Model 1.1.0. Picks saved 2026-10-31 09:04 Dubai (UTC+4).