When to take a hit

An extra transfer costs 4 points. The squad page prices this in: the Net column is gross projected gain minus the hit.

The bar

A −4 needs the incoming player to out-score the outgoing one by more than four points over the horizon you are planning. The transfer plans use a three gameweek horizon, so a hit that gains 4.5 points over three weeks is barely worth it: that is 1.5 points a week for a permanent 4-point loss.

What the numbers do not include

• Price movement. Selling a rising player and buying a falling one costs real squad value over a season, and none of that shows up in the projection. The Δ£ risk column on the stats table is the signal to check.
• Rank context. A hit that gains 2 points is bad in isolation, and fine if it moves you onto a player 60% of your league owns and you do not. Check the league page's Gaps section before deciding.
• Injuries you already know about. The model sees status and chance_of_playing, but not a press conference from an hour ago.

Rule of thumb

Take the free transfer every week it gains anything. Take a hit when the net number is comfortably positive and the player is someone you would have wanted anyway next week.