Football hates VAR.

Half the country still isn’t sure about xG.
The other half pretends they understand it.

And yet football absolutely loves a new metric — preferably one that makes the game feel smarter, more modern, and just confusing enough to argue about on the internet.

xG. xA. xThreat. Field tilt. Shot-creating actions.
All very clever. All very serious. All loved and hated in equal measure.

But while football has spent the last decade measuring every blade of grass and every half-chance from 30 yards, one part of the game has remained completely unquantified:

The transfer rumour.

Every January and every summer, reality collapses.
Fans refresh timelines. Forums explode. “ITKs” emerge. Taxi sightings are analysed like Zapruder film. Sky Sports News stands outside empty stadiums telling us nothing is happening — which somehow makes it feel like something is.

And almost all of it is nonsense.

So rather than arguing about whether xG has ruined football, we decided to apply football’s favourite obsession — data — to the most ridiculous part of the sport.

Not to predict signings.
But to explain why we keep falling for rumours that were never happening in the first place.

Welcome to the TSBW Transfer Model™ — a purpose-built system designed to measure agent leaks, forum hysteria, hopium, and the overwhelming probability that nothing happens at all.

Finally, a stat that understands the transfer window.

Why the TSBW Transfer Model™ Exists

Traditional analytics assumes three things:

1. Information is reliable
2. People behave rationally
3. Deadlines mean something

Transfer windows politely ignore all three.

Instead, they operate on agent leaks, recycled reporting, forum hysteria, influencer engagement farming, and a collective willingness among fans to suspend disbelief because this one feels different.

The TSBW Transfer Model™ was built to reflect reality — not optimism.

The Core Metrics

These are the headline numbers. Your go to metrics – if you remember nothing else from this model, remember these!

xT – Expected Transfer

The probability the player actually signs for the club. A genuinely rare stat.

Anything above 0.6 usually requires:

• A credible source
• A logical squad fit
• Silence from everyone who normally talks too much (every club has them)

xL – Expected Loan

As above, but also measures how likely this ends in:

“Season-long loan with an option to buy.” or “loan til January, no recall”. We could break this down, but we really can’t be arsed. What we have done is over the top as it is.

High xL often coexists with optimism, before reality kicks in.

xB – Expected Bullshit

The most important metric in the model.

xB measures the likelihood that the rumour is complete nonsense.

Baseline values:

• Influencer source: 0.85+
• ‘ITK’ account like ‘agentFC48583939’ (EFL Transfers On Point) handle: 0.90+
• Screenshot of WhatsApp / Forum message: 0.95

If xB is elite-tier, everything else becomes academic.

NOTE: ITK = In The Know

Behavioural Indicators (Where the Damage Is Done)

These metrics explain why a rumour spreads — not whether it’s true.

xA – Expected Agent Leak

This measures the chance the rumour exists purely to:

• Force a new [or improved] contract
• Speed up another club
• Create interest where none previously existed

High xA usually means the player was never coming.

xI – ITK Reliability

An inverse metric.

The more:

• Emojis
• Capital letters
• “Trust me” language
• Often downplay the message with ‘so I’ve heard’

…the lower the reliability.

High follower count does not improve xI.

xE – Echo Factor

Tracks how often the same information is reposted, paraphrased, and reframed until it feels like momentum.

No new facts required. Just repetition.

xFF – Fan Forum Effect

The most powerful accelerant in the model.

xFF measures how much credibility a rumour gains simply because:

• A long forum thread exists
• Multiple posters say they’ve heard it
• Someone’s cousin works near the training ground

High xFF creates confidence without evidence.

This is where rumours stop being questioned and start being defended.

• “Someone on the forum said it”
• A 200-post thread exists
• Screenshots start doing the rounds
• “Too many people saying it for it not to be true”
• Same post quoted 12 times with added certainty
• Mild scepticism treated as heresy

High xFF = zero new information, maximum confidence.

xTx – Taxi Spotting Index

A false-positive classic. Any rumour involving:

• A taxi
• A train station
• A blurry photo

Could also include Flight Radar plane spotting.

Immediately inflates belief while doing nothing for xT.

xS – Sky Sports Inflation

Automatically adds credibility when someone is stood outside the ground with a microphone (see above image). Even if they’re saying: “There’s nothing happening at the moment.”

Add to ‘in studio’ analysis that stretches out airtime recapping the players full playing history, and how transfers actually work, every time, for every player transfer scenario.

xD – Deal Sheet Risk

Spikes sharply after 10pm on deadline day.

Usually followed by: “Paperwork couldn’t be completed in time.” or “fax machine not working”.

Emotional & Cultural Indices

These explain the fan reaction, not the rumour itself.

xH – Hopium Index

Calculated by dividing vibes by facts. xH rises sharply after:

• A good win
• A bad run
• A feeling that “we’re one player away”

xFM – Football Manager Bias

Measures the likelihood the rumour exists because:

• Someone signed the player on FM
• It worked once
• And it felt realistic

Yay, a new use for the acronym XFM!!

xGz – Gazump Risk

Activated when the phrase: “A bigger club is monitoring the situation” appears. Almost always terminal.

xC – Club Sense Check

If the rumour:

• Fits the recruitment model
• Makes financial sense
• Solves an actual problem

Fans immediately distrust it.

High xC paradoxically lowers belief.

xØ – Expected Nothing

The silent killer. xØ measures the probability that: Nothing happens at all.

Historically, this is the most accurate metric in the entire model.

How the Metrics Interact

The model becomes most accurate when you look at combinations:

• High xFF + High xB = Forum-fuelled nonsense
• High xH + High xE = Self-sustaining rumour loop
• High xA + Low xT = Agent working overtime
• High xFF, xH, and xE together = xØ quietly wins

Why This Model Works

The TSBW Transfer Model™ doesn’t try to predict football clubs.

It predicts behaviour.

• How rumours spread
• Why fans believe them
• And why disappointment feels inevitable

It accepts a simple truth: Most transfer rumours are content, not information.

Final Verdict

The Transfer Model™ does not promise signings.

It promises clarity. Sometimes that clarity is:

• Low xT
• High xB
• Elite-tier xØ

Which, statistically speaking, is called a transfer window.

The TSBW Transfer Model™
The most accurate transfer model in football.

If nothing else, it explains why our own xIRLP* is near 1.0 during the transfer windows.

Small print: Accuracy improves dramatically when nothing happens.

And finally, don’t expect us to remember all this. TSBW does not hold any responsibility for the success or failure of any transfer anywhere, ever.

*xIRLP = In Real Life Productivity

Full transparency, this whole idea started with the TSBW. AI helped create the initial text and flow, then it was checked, reworded, amended and checked again by a human, then published 👍🏼