A comparison score
The 0–100 number helps compare players using several signals at once. It is not a forecast of how many FPL points they will score next gameweek. That separate estimate is shown as xPts.
Most inputs are normalised within each position. A goalkeeper and a forward can both score highly without having similar raw attacking numbers. A player’s rank can also move when another player changes, even if the first player has not played.
The current weights
Recent form and total season points each contribute 25%. Expected goal involvement contributes 15%. Playing-time reliability, upcoming fixtures and value contribute 10% each. Availability contributes the final 5%. The availability status also determines which players are rankable.
These are the current baseline weights, not a claim that one formula will always identify the best transfer. The model is versioned and its separate points forecasts are measured against results once a pre-deadline snapshot has a completed gameweek to compare with.
A worked comparison: two midfielders
Imagine comparing Bukayo Saka and Cole Palmer in the Players table. Their names are examples; every value here is illustrative and says nothing about their current form or FPL status. Suppose one has a stronger recent form score while the other has more season points and a better five-week fixture run. The ranking combines those signals rather than allowing the latest return to decide the order on its own.
For a simple illustration, if their normalised form scores are 80 and 60, the form component contributes 20 and 15 points respectively because form has a 25% weight. A five-point lead can be offset by the season-points, expected-involvement, minutes, fixtures, value and availability components. The final 0–100 score is a weighted comparison after normalisation, not a sum of their raw FPL points.
Why normalise within positions?
Goalkeepers, defenders, midfielders and forwards earn different kinds of points and have different attacking baselines. A raw expected-goal figure that is high for a defender may be ordinary for a forward. FPL Intel scales most signals within each position so a player is judged against relevant peers rather than against a single all-position distribution.
This also explains why a score of 80 for a goalkeeper and 80 for a forward does not mean they are interchangeable transfers. They occupy different squad slots and their scores reflect different peer groups. Use the score to make a shortlist inside the position you need. Then compare actual forecast points, price, fixtures and your squad constraints before deciding.
What each part contributes
Recent form and season points each get a quarter of the score. Together they recognise both short-term returns and a longer record. Expected goal involvement gets 15% so repeated chances can be visible before they turn into FPL points. Minutes reliability gets 10% because an attractive player still needs time on the pitch.
Upcoming fixtures get 10%, helping to distinguish otherwise similar players with different schedules. Value gets 10%, recognising that points at a lower price can leave budget for improvements elsewhere. Availability receives 5% and the status also affects eligibility for ranking. These weights make the score explainable, but they do not include every consideration a manager may have.
A change in rank is not always a change in player
Normalisation uses the current group of players. If a peer’s form or price changes sharply, the scale can move and another player’s score or rank can change even if that player did not play. An injury flag can also remove a player from the rankable set. This is normal behaviour for a relative score.
Use ranks to find candidates, not to infer that a player became better overnight. When a ranking surprises you, inspect the underlying form, minutes, expected involvement and fixtures. A one-place movement is rarely enough on its own to justify a transfer, particularly if it would cost points or block a planned move later.
Ranking score and xPts answer different questions
The ranking score asks how strong a player looks across the weighted signals relative to peers. xPts estimates points for a particular gameweek or fixture horizon. A player with excellent season-long form can rank highly but have a modest next-round forecast because of an awkward opponent, reduced minutes or a blank gameweek.
For a one-week captaincy choice, look first at the upcoming forecast and its inputs. For a transfer, consider both ranking and the three- or five-week forecast. A ranking is particularly useful when your initial shortlist is large; xPts helps you examine the schedule ahead. Neither replaces checking team news near the deadline.
Build a shortlist you can act on
Filter the Players table to the position and price range your squad allows, then sort by score. Open the leading player pages and compare minutes, availability, upcoming fixtures and forecast points. If two candidates remain close, look at your planned future transfers and how each fits your team structure.
Remember that a free transfer has value beyond this week. An apparent score advantage may be too small to justify changing a player you already own, especially if your existing option has a good fixture run. The score is designed to make the first pass faster and more consistent. The final decision remains yours.
Use it with context
A high score is a sensible shortlist signal. Before acting, check the player’s expected minutes, next fixtures, price, squad fit and the length of time you would keep them. For a one-week captain choice, xPts may be more directly relevant than a season-wide ranking.
The “Why this player?” controls on featured cards show the selection rule behind each label, including the extra minutes threshold used for the best-value pick.
Put the idea into practice
Compare ranked players