FPL Expected Points (xPts) Explained

I built FPL Copilot's expected points model because FPL's built-in "form" stat kept misleading me. Form is backward-looking. xPts is forward-looking. That distinction changed how I make every FPL decision.

What Are FPL Expected Points (Predicted Points)?

FPL expected points (also called predicted points, projected points, or xPts) project how many fantasy points a player will score in a given gameweek. They are the output of an FPL points predictor: a model that combines expected goals (xG), expected assists (xA), clean sheet probability, minutes likelihood, and bonus point patterns, all converted into FPL scoring.

"Expected points," "predicted points," and "projected points" mean the same thing in FPL. xPts is the shorthand. Whichever name you search for, this is the stat that turns gut feel into maths.

I update the model multiple times a week as new match data comes in. It processes 50+ data points per player per gameweek across all 686 Premier League players.

Why I Built It This Way

Most xPts models just look at player-level stats. Ours factors in the opponent. A midfielder with 0.4 xG per 90 will produce very different numbers against the league's worst defence vs. the best. Fixture difficulty is the variable that most models underweight.

The model also accounts for rotation risk. A player with 6.0 xPts who only has a 60% chance of starting is really worth 3.6. The solver and the EP table both factor this in so you're not blindsided by Pep roulette or cup rotation.

How to Get the Most Out of the EP Page

Captain picks: Sort by xPts for the current GW. The top player gets the armband. Over a season, this alone gains you 10-20 points compared to gut feel or following the crowd.

Spot fixture runs early: Switch to multi-GW view and look 3-5 weeks ahead. Players with great upcoming runs show up before the FPL community catches on. Early movers get better prices.

Transfer targets: Compare cumulative xPts across your planning window, not just this week. The better fixture run almost always wins over a single-week spike.

Bench order: Put your highest-xPts bench player first. Over a season, good bench ordering is worth 5-10 points from autosubs alone. Most managers ignore this.

The EP page lets you override projected minutes for any player. If you know something the model doesn't - a manager quote, a training photo, a confirmed absence - adjust the minutes and the xPts recalculate instantly. This is the fastest way to get an edge when news breaks.