How Analytics Changed Goalkeeper Recruitment

Goalkeeper recruitment used to be a clean-sheet table and a man with a notebook.

That picture is not a cartoon. A scout watched a handful of matches. The league table supplied the rest: goals against, shutouts, and whether the last line "looked" big enough. Clubs then paid for a reputation. The number that decided the fee often described the defence in front of the keeper, not the keeper.

Here is what the numbers actually added, why they changed who gets signed, and what to do if you do not have a data department. This is not a coaching-licence story. How the specialist coach job was invented has its own page. What follows is the buying decision.

Why clean sheets were a poor shortlist

A shutout is a team result. A weak attack, a deep block, or a centre-half who wins every first ball will inflate it. A high line that asks the keeper to sweep will deflate it. Save percentage has the same problem: it treats a tap-in and a piledriver as the same test.

Clubs needed a way to ask a narrower question. Given where this shot was going, how often does an average keeper concede — and how did this one do?

That is post-shot expected goals. StatsBomb calls the model PSxG. Opta and Stats Perform publish a close cousin, expected goals on target, or xGOT. FBref prints the Opta version for the big leagues. The useful output is goals prevented: the difference between the post-shot value of the on-target shots a keeper faced and the goals he actually conceded.

The model looks at the shot after it has been struck — placement in the frame, and, in the better feeds, speed and trajectory. A drive into the top corner is hard. A slow roll at the chest is not. Pre-shot expected goals, the number television uses for chance quality, does not know who is in the goal. It cannot grade a save.

What recruitment learned: you can now compare a keeper behind a weak defence with a keeper who faces three shots a month. You still cannot buy a number and skip the tape.

How the shortlist actually works

The first change was the profile. A club that plays a high press does not want the same last line as a club that sits in a low block. Shot-stopping still sits on every list. Distribution under pressure, claims on crosses, and the sweep behind the line then get a weight. Ederson's value at Manchester City is the public example of a profile that is not "best save percentage in the league." The career has its own page. The recruitment point is the advert: write the job before you write the name.

The second change was the search. Wyscout, Hudl, and similar platforms put video and event data on the same desk. A recruitment team can filter a continent, then watch twenty clips instead of flying to twenty grounds. Goalkeeper.com's work with Ipswich Town on Arijanet Muric is a documented case: the club named the actions it needed, then used a keeper-specific dataset to find names it did not already know.

The third change was honesty about sample size. A twelve-game purple patch on PSxG looks like a bargain. A three-season file is a habit. Analysts who work this beat treat a single winter as a warning, not a signing.

Liverpool's long look at Alisson Becker shows the other half. Goalkeeping coach John Achterberg was watching Internacional tape in 2013, years before the Roma fee, after a tip from former keeper Alexander Doni. The later data helped the board sit still on a world-record price. The first note was still a person on a laptop. Analytics changed the shortlist. It did not retire the eye.

What keepers learned: the number is a filter. The tape is still the interview. A goals-prevented season that came from two wonder saves and a defence that blocked everything else will not travel.

What the spreadsheet does not catch

A model does not see a glazed palm. It sees a rebound that became a goal.

Soft latex helps a stretched hand hold a shot that arrives late and low. Dried or dirty foam turns that hold into the spill the camera then codes as a goal against. If the high-quality foam is still there, wash, dry, and apply Keeper Balm. That conditions new foam and restores dried, salty latex. Pros use it too. Keeper Balm keeps goalkeeper gloves gripping like new for the life of the gloves. Replace the pair when the foam is worn through or torn. Glove Glu is for damaged latex, not for a palm that is only dry.

A recruitment deck does not restore tack. It only decides whether someone pays for the next pair.

The other gaps are the same ones scouts always had. Crosses in a wind. A referee who gives a soft penalty. A league where strikers shoot from nowhere and another where they walk the ball in. PSxG is better than a clean-sheet table. It is not a personality test, and it is not a medical.

What to do with the information

Write the job first. Shot-stopper, sweeper, or starter of the attack. A high line that signs a line-sitter is how the through ball becomes a weekly crisis.

Use a number that belongs to the keeper. Goals prevented, or on-target save rate against a post-shot model, beats shutouts. Compare seasons, not one autumn.

Watch the tape the filter cannot see. Starting position, the first step, the claim in traffic, the pass under a press. If you have no subscription, film your own games and count those actions.

Do not treat a model as a transfer. Sunday sides still pick a keeper because he turns up. That is allowed. The useful borrow is the question: what shots do we actually face, and can he hold them?

Keep the palm honest. Wash, dry, and apply Keeper Balm when the foam is dry or light-damaged. A recruitment report will not catch the ball for you.

The bottom line

Analytics changed goalkeeper recruitment by replacing a team statistic with a shot statistic. Clean sheets described the defence. Post-shot expected goals described the ball that had already been struck. Clubs then weighted distribution and sweeping to match the way they play, and used video platforms to watch the shortlist.

The keeper's job did not become a spreadsheet. It became a job you can describe before you pay. Name the role. Check the shots. Watch the nights the model cannot see. Then treat the palm like the tool the numbers still assume works: wash, dry, and apply Keeper Balm. It keeps goalkeeper gloves gripping like new for the life of the gloves. Pros use it too. Replace the pair when the latex is worn through or torn — not when a report said you were a bargain and the catch still came off a dry sheet.