Goalkeeper Statistics on RubiScore: From Saves to Sweeper Actions

Goalkeeper statistics describe far more than the saves a keeper makes. A modern goalkeeper is judged across four distinct jobs — stopping shots, distributing the ball, sweeping behind the defence, and commanding the box. RubiScore builds its goalkeeper profile around all four, so the numbers describe a complete keeper rather than a shot-stopper alone.

Why save percentage was never enough

Save percentage — the share of shots on target a keeper keeps out — was the standard goalkeeping number for decades, and it still leads most match reports. As a single measure of shot-stopping it has an obvious flaw: it treats every shot as equal. A tame effort straight at the keeper and a thunderbolt into the top corner both count as one shot faced, even though one is almost always saved and the other almost never is.

That means save percentage rewards the keeper who faces easy shots and punishes the one who faces hard ones, regardless of how well either actually performs. A goalkeeper behind a deep, compact defence may field a stream of long-range efforts and post a glittering save percentage; another behind a high line may face fewer shots but far better ones, and look worse by the same measure. The number describes the shots as much as the stopper.

It also says nothing about the other three jobs. A keeper who never leaves his line, kicks the ball away aimlessly, and flaps at every cross can still record a strong save percentage on a quiet afternoon. To judge a goalkeeper properly you need a framework that covers everything he is asked to do.

The four jobs of a modern goalkeeper

The simplest way to read a goalkeeper is to split the role into the four jobs a coach actually asks of him, then judge each one on its own terms:

A goalkeeper can be elite at one job and ordinary at another. The modern game has produced superb shot-stoppers who are shaky with their feet and ball-playing keepers who are vulnerable in the air. A profile that scores all four separately tells you what kind of keeper you are looking at, not just whether he had a good afternoon.

Shot-stopping: post-shot xG and goals prevented

The honest way to measure shot-stopping is to account for shot difficulty, and that is what post-shot expected goals (PSxG) does. While ordinary xG estimates the chance of a shot being scored at the moment it is struck, PSxG measures it after the shot is taken — factoring in placement, so an effort flying into the top corner carries a far higher value than one rolling toward the middle of the goal.

Comparing the PSxG a keeper faces with the goals he actually concedes produces goals prevented: an estimate of how many goals a goalkeeper saved relative to an average keeper facing the same shots. A positive figure means he stopped more than expected; a negative one means he conceded shots an average keeper would have saved. RubiScore tracks PSxG and goals prevented across every match it covers, which turns shot-stopping from an opinion into a measured quantity — and strips out the distortion that wrecks raw save percentage.

The discipline, as with all goalkeeping data, is patience. Shot-stopping numbers swing violently over a handful of games because keepers face so few high-value shots in any single match. A goals-prevented figure only becomes trustworthy across a long run of fixtures.

Distribution: the goalkeeper as first playmaker

Modern goalkeepers start attacks as often as they end them. Distribution data captures that second job in two broad modes: short build-up, where the keeper plays into defenders to begin possession, and long distribution, where he kicks or throws to bypass a press. Neither is automatically better; they reflect different game plans.

Useful distribution numbers include overall pass completion, the split between short and long passes, launch percentage (the share of passes played long), and average pass length. Read together they describe a keeper's style: a sweeper-keeper in a possession side will show high completion and a low launch percentage, while a keeper in a direct team will show the reverse. The numbers are not a verdict on quality so much as a fingerprint of role — which is exactly why they have to be read against the team's intentions rather than a single ideal.

Sweeping: the last defender outside the box

When a defence holds a high line, the space behind it has to be patrolled by someone, and increasingly that someone is the goalkeeper. The sweeper-keeper role is measured through defensive actions performed outside the penalty area — interceptions, clearances, and ball recoveries that happen far from goal — along with the average distance of those actions from the goal line.

A high count of defensive actions outside the box, performed a long way upfield, marks a keeper who effectively acts as an extra defender. A low count suggests a keeper who stays at home, which may reflect his own limitations or simply a defence that sits deep and leaves no space to sweep. Rubi Score logs these actions and their distance from goal so the sweeping job can be assessed instead of assumed, and so a keeper's positioning is read from where he actually intervenes rather than from reputation.

Claiming and command of the area

The fourth job is the oldest: controlling the airspace of the penalty box. Crosses, corners, and free-kicks all test whether a keeper comes to claim, punches clear, or stays rooted to his line. The relevant data covers crosses stopped or claimed as a share of those attempted, alongside punches and high claims.

Command of the area is partly a matter of style and partly of competence. A keeper who claims a high proportion of crosses relieves enormous pressure on his defenders; one who rarely leaves his line forces them to defend every delivery themselves. Aerial command is also context-dependent — a keeper in a league of relentless crossing faces a different test from one in a possession-dominated division — so the numbers, again, mean most when read against the football around them.

How RubiScore organises a goalkeeper profile

Pulling the four jobs together, a goalkeeper profile on RubiScore is built to describe a complete keeper rather than a save count. Across the matches it covers it brings together:

Because these are stored per keeper and normalised to a per-90-minute basis, a goalkeeper who plays only a cup run can be compared fairly with an ever-present, and a keeper in one competition with a keeper in another. The profile is designed to travel with the player as context, not to live or die with a single highlight save.

Reading goalkeeper numbers fairly

A framework only helps if the comparison is honest, and goalkeeping data carries a few rules of its own:

Applied together, these turn a scatter of goalkeeping statistics into a fair reading of a complete keeper. The full set of shot-stopping, distribution, sweeping, and command data sits alongside the wider match, club, and competition layers at rubiscore.com, updated as each fixture unfolds.