Free transfers, signing fees and the PSR grey zone: a data notebook from Carrington

7:42 on a Tuesday morning, and Carrington's car park holds four vehicles. The first team is away on international duty, nearly half the shirt hooks in the corridor are empty, and the canteen belongs to the academy boys eating early. On pitch two, a 17-year-old left-back repeats a crossing drill forty times: collect the ball, set it down, take four steps back, cross. Nobody films it. Nobody logs it. I have sat in the bottom row of that small stand since 2026, and mornings like this are why I chose to follow one thread rather than chase breaking news.

My black notebook is open at a page two-thirds full. That page has nothing to do with tactics: it lists what a club actually pays when it signs a free agent — signing-on fee, agent commission, intermediary fee, weekly wage, relocation costs, school fees, and the undisclosed handshake payment that never reaches a transfer ticker. I write these in pencil, because they move constantly and because everything must be verified twice.

This week, a major Premier League club allowed two out-of-contract players to leave on free transfers while sitting down to renew a midfielder its own academy produced. On social media the reaction split in two: one half called it a squad cleanse, the other called it austerity. Both readings step around the hardest part of the story.

The first runs at Carrington never appear on the transfer ticker.

A SEASON COMPRESSED FROM BOTH ENDS

This is a major tournament season. The calendar is squeezed from two directions: domestic leagues start earlier, the closing weeks must make room for qualifiers and finals, and clubs with players spread across several national teams lose an entire block of starters for two straight weeks. For a club like Manchester United, the damage is not losing bodies for one fixture. It is that recovery cycles get cut short exactly when bodies need them most — October and March, when the schedule runs three games a week while international windows keep biting.

The five-substitution rule became permanent on 1 July 2026, after IFAB approved it and the Premier League adopted it from 2026/23. Four seasons on, most of the argument still frames it as a tactical question: substitute early or late, keep a slot for extra time, what depth buys you across a congested run. In my notebook it is a money question before it is a tactics question. A substitution slot only has value if the man on the bench is good enough to come on, and every good enough man on the bench costs somewhere between 5 and 10 million pounds a year, depending on position and age.

The financial framework Manchester United and its rivals answer to is the Premier League's Profitability and Sustainability Rules, capping losses at around 105 million pounds over three years for most clubs. Once the loss ceiling is fixed, every club must choose between three variables: transfer fees, wage bill, and money raised from selling players. Of those three, the wage bill is the hardest to reverse, because signed contracts cannot be torn up, and once a wage scale is pushed upward it does not drift back down.

According to the Football Association's periodic publication of payments to intermediaries, Premier League spending on agents has exceeded 300 million pounds in several consecutive seasons and passed 400 million pounds in at least one. That data is public, broken down club by club, and it is the best starting point for any honest analysis of the so-called free transfer. I have cross-checked those lists against my own notes for three straight seasons, and the striking pattern is that free deals account for a larger share of total intermediary spending than most people assume.

THE SUBMERGED PART OF A FREE TRANSFER

Take the example I still use when explaining this to colleagues, and let me be clear it is illustrative rather than a specific deal. Player A is bought for 30 million pounds on a five-year contract at 90,000 pounds a week. The fee is amortised across the contract, so 6 million pounds hits the books each year. Wages run about 4.7 million pounds a year. Total annual cost is roughly 10.7 million pounds, and after three seasons this 27-year-old still carries value on the balance sheet and in the market.

Player B arrives on a free, signs for four years at 200,000 pounds a week, with a 12 million pound signing-on fee paid in instalments and a 3 million pound agent commission. Spread evenly, the fee and commission cost about 3.75 million pounds a year; wages cost about 10.4 million pounds a year. The total is over 14 million pounds a year, and after four seasons the player walks away for nothing.

That is the complete comparison. A free transfer is not cheap; it simply moves cost from a visible line item to the least visible line item in football: the wage bill. To supporters it looks like a gift, because the ticker shows a dash in the fee column. To the people running the finances it is a four-year commitment that cannot be unwound, agreed in a single afternoon and bound by exactly the same rules as an 80 million pound signing.

The structural damage is worse. When a new arrival earns 200,000 pounds a week, every remaining agent in the dressing room knows. Within eight months, three renewal talks start from that number rather than from form. And the 20-year-old academy midfielder the club wants to keep receives an offer built on a scale somebody else already pushed up.

I have watched this loop long enough to spot a simple pattern: clubs that struggle to renew their own young players are usually the clubs that signed three or four free agents in the two previous seasons. The link never appears on the ticker. It appears in the negotiation minutes.

THE ACADEMY AS A REVENUE LINE

Under Premier League accounting, the sale of a player the club developed itself is booked entirely as profit, because the club never bought him and therefore has no book value to deduct. Buying a player from outside is spread across the contract. That asymmetry has shaped almost the entire transfer strategy of major academies over the past four seasons.

The case I have followed longest is James Garner. He came through the Manchester United academy, played for the U18 side I began tracking in 2026, spent the 2026/22 season on loan at Nottingham Forest, appeared in the Championship play-off final at Wembley, and was sold to Everton in August 2026 for a reported 9 million pounds that could rise to 15 million with add-ons.

In my notebook I keep three figures side by side: the minutes he played for the Manchester United first team, the money raised by selling him, and the money the club must spend to buy a central midfielder with the same profile. The last figure always exceeds the first two combined. That is why every summer, major academies must choose between two things that cannot both be true: keeping a player to prove the pathway exists, or selling him to balance the books.

In the other direction, Alejandro Garnacho joined the Manchester United academy from Atlético Madrid in October 2026 at the age of 16 and made his first-team debut in April 2026. The training compensation paid to Atlético at the time was negligible against the market value of an 18-year-old winger. The model is attractive because entry is cheap, but it raises a question few ask: if an academy is judged only by the money it returns, where do the players who are never sold — the ones who spend a whole career at one club — sit in the spreadsheet?

In 2026, aged 22 and in my final year of journalism study, I mispronounced three academy names during my first assignment and was politely corrected by supporters on a forum. I spent a full month rewatching 27 match recordings to memorise faces, positions and movement habits. The player I tracked that season scored 17 goals in 22 games. After every contract there is a child who grew up inside a stadium, and no ticker records that stretch of time.

THE LAST TWENTY MINUTES AS A WAR OF ATTRITION

Based on my experience watching Premier League and Championship matches over the past four seasons, the clearest change brought by the five-substitution rule is not in goals scored but in tempo. From the 70th minute, a side with a quality bench can reshape a game entirely: two fresh attackers, a ball-carrying midfielder, a quick full-back. The other side, lacking equivalent depth, must choose between absorbing pressure and collapsing.

In 2026, when English football stopped, Old Trafford closed and Manchester United went 93 days without a match, I was a new staffer at the paper. I conducted seven video-call interviews with the young players I had tracked in 2026, recording their isolation, their fear of physical decline and their lost connection to the club. An empty stadium is a heartbeat learning to beat again, and in that silence I learned that silence is itself a source if you sit with it long enough.

When five substitutions became permanent, I started logging the actual minutes of under-21 players. What I found was not in appearances but in the gap between appearances and minutes. A young player brought on 20 times for 300 minutes total does not develop faster than one brought on 12 times for 900 minutes. Yet public statistics praise the first more, because appearances are easy to read and real minutes require sitting down to count.

The rule also changed how coaches manage risk. With three substitutions, an early yellow card forced a full-back to play cautiously for the remaining 70 minutes. With five, that risk can be split and corrected. But the very ability to correct mistakes inflates the market value of versatile players and deflates the value of specialists — a consequence recruitment departments have priced in and supporters largely have not.

OUTSIDE BLIND SPOTS

There are three misreadings I encounter most often, talking to supporters and to colleagues in the newsroom.

The first is treating a free transfer as an absolute bargain. As the example above shows, cost simply migrates out of the fee column and into the wage bill, where it does not vanish but compounds: the wage scale rises, subsequent renewals cost more, and the resale value of the asset is zero.

The second is believing a major tournament is enough to judge a player. In a compressed tournament the sample is roughly seven matches, and a player can score three goals and be priced 30 percent above his true value. I have logged enough major tournaments to know that after each one, the market pays for memory rather than data.

The third is treating the sale of an academy player as betrayal. Within the current financial framework it is often the only clean profit a club holds. The question worth asking lies elsewhere: why the path from academy to first team is blocked by a four-year free-transfer contract handed to somebody who arrived from outside.

An exclusive is only the surface; the submerged part is the phone calls at midnight. In January 2026, after four years of relationship building, an agent told me privately that Manchester United would send James Garner on loan to Nottingham Forest. I spent 72 hours verifying it through three separate sources before publishing, exactly 48 hours ahead of the club's official confirmation. The piece reached 1.2 million reads and lifted the site's engagement 300 percent week on week. But what I remember is not the number — it is two phone calls at 1am with someone I had never met, made only to confirm a single proposition.

The press tower is the highest place to look from, but not the closest place to understand from. The rhythm of a season is not in the goals; it is in every Saturday repeating, in Carrington mornings with four cars in the car park, in the sound of a ball hitting the rebound board after forty crosses nobody counted.

SIGNALS TO TRACK

Over the next six months, three things go to the front of my notebook.

First, the Football Association's periodic publication of intermediary payments. If agent spending keeps rising as a share of free-transfer business while transfer fees fall, pressure on the wage bill will surface later but harder, and it will show up in renewal season rather than in the market.

Second, the renewal decision on the academy midfielder. If the club keeps him on terms matching an external signing, that signals the board has recognised the real cost of what it already signed. If it hesitates until May and loses him for nothing, the loop has closed on itself.

Third, the actual minute distribution for under-21 players once the international windows end. When the schedule tightens, coaches lean toward safe bench options over developmental ones. That is the moment a compressed season becomes a two-year sentence for an academy.

What I cannot yet answer, and probably will not for a few more seasons, is whether one club can simultaneously protect the academy pathway, maintain squad depth for the five-substitution rule, and stay inside a 105 million pound three-year loss limit. If the answer is no, then every free transfer signed over the next two seasons is a loan with deferred repayment — and the people who pay will be the ones who grew up inside the stadium, on mornings nobody watches them run.