Player Radar

What it shows

Six axes, one for each of the ways a player can add value to their team: shooting, dribbling, passing, receiving, interrupting and drawing fouls. The radar plots a percentile — 90 on the passing axis means the player passes better than 90% of the players they're being compared with, not that they scored 90 out of 100.

Underneath, the same six numbers appear as goals added per 96 minutes, which is the part the percentile hides. Two players can both sit in the 70th percentile for dribbling while one of them is worth three times as much.

You can put a second player on the radar, move the minutes floor to change who counts as a peer, and switch the comparison between a player's exact position, their position group, or the whole league.

Who a player gets compared to

By default, other players in the same position and the same league. When that pool is too small to mean anything — the USL Super League has four attacking midfielders past the minutes floor — the comparison widens one step to the position group, and the page tells you it did. It never quietly ranks somebody against a handful of peers and prints the number as though it were solid.

Leagues covered

MLS, NWSL, USL Championship, USL League One, MLS Next Pro and USL Super League — 2,046 outfield players.

Outfield only: goalkeepers are modelled through a different set of actions, so a keeper on these six axes would be six axes of nothing. They're left off rather than faked.

Where the numbers come from

Goals added (g+) from American Soccer Analysis. It puts a value on every action a player takes, based on how much that action moved their team's chance of scoring and of conceding, then reports it against an average player in the same position. Everything here is per 96 minutes and covers the season so far.

A note on what this isn't: clubs at the top of the game build radars like these on optical tracking data — high-speed runs, sprint distance, top speed. Nobody publishes that for American soccer, so it isn't here. What is here is the ball-in-play half of the same question. See the methodology & disclaimer.