How Data-Driven Football Clubs Find Undervalued Players

Data-driven football clubs find undervalued players by estimating what a player could contribute within their own system and comparing that potential contribution with the total cost of signing him. They combine statistical screening, contextual adjustments, video analysis, live scouting, medical evidence and financial modelling.

The objective is not simply to discover good players. Most good players are already visible and appropriately expensive. The real challenge is identifying players whose future value is greater than their current reputation, transfer fee or salary suggests.

This may mean finding a player in an overlooked league, recognising that poor team results are hiding strong individual performance, or identifying a specialist whose abilities are particularly valuable within the recruiting club’s tactical model.

What makes a football player undervalued?

An undervalued player is someone whose expected sporting or financial contribution is greater than the cost required to acquire and employ him.

That definition contains three separate estimates:

1. Expected contribution: how much the player could improve the team.
2. Total cost: the transfer fee, salary, bonuses, agent fees and opportunity cost.
3. Uncertainty: the probability that the player adapts, develops and remains available.

A player is not undervalued merely because he is inexpensive. A free transfer on excessive wages can be poor value. A £30 million player can be undervalued if his expected contribution and future resale value justify a significantly higher valuation.

Value is also specific to the buying club. A centre-back suited to defending a low block may struggle in a team that holds a high line and asks its defenders to progress the ball under pressure. The same player can represent excellent value to one club and an unsuitable investment for another.

This is why sophisticated recruitment begins with role, system and price rather than a universal ranking of the world’s best players.

Why the football transfer market can misprice players

The transfer market contains thousands of clubs, players and decision-makers. It is competitive, but it is not perfectly efficient.

Player valuations can be influenced by reputation, recent form, league visibility, international appearances, agent relationships, media narratives and the financial circumstances of the selling club.

Several recurring inefficiencies can create opportunities.

Players in less visible competitions

Major European leagues receive extensive scouting, media and data coverage. Players in smaller leagues may attract less attention even when their underlying abilities are transferable.

This does not mean that every successful player in a minor competition is undervalued. Statistical output must be adjusted for league strength, team quality and tactical context. The opportunity exists when the market discounts a competition more heavily than the evidence justifies.

Players performing for weak teams

Team quality has a substantial effect on individual statistics.

A midfielder playing for a relegation candidate may have fewer passing options, receive the ball under greater pressure and spend more time defending. His raw attacking output might look ordinary even if he consistently performs difficult actions well.

A strong recruitment model tries to distinguish the player’s contribution from the environment surrounding him.

Players used in the wrong role

A footballer may appear ineffective because his current team asks him to perform tasks that do not suit his strengths.

A wide forward who excels at attacking space could struggle in a team facing deep defensive blocks. A technically secure midfielder might look passive when instructed to hold a rigid position. A full-back’s progressive ability may be hidden if he is rarely allowed to advance.

The buying club may see a different application for the same abilities.

Players with unfashionable profiles

Recruitment markets can overvalue visible characteristics such as size, speed, goals or spectacular actions.

Less visible abilities may receive insufficient attention:

• finding space before receiving the ball;
• making reliable decisions under pressure;
• preventing opposition progression;
• creating space through off-ball movement;
• maintaining tactical discipline;
• and performing consistently without producing highlights.

These qualities are difficult to evaluate using simple totals, but they can be extremely valuable within a coherent team.

Contract and timing effects

A player’s market price is not determined by talent alone.

Contract length, age, selling-club finances, relegation, promotion clauses and the player’s willingness to move can all affect the transfer fee.

A strong recruitment department monitors these circumstances continuously. The same player may be overpriced in one transfer window and attractive six months later.

Step one: define the role the club needs

The recruitment process should begin before individual names are discussed.

The club must define the problem it is trying to solve. “We need a midfielder” is too broad. A more useful description might be:

• a defensively reliable number eight;
• comfortable receiving under pressure;
• capable of progressing possession through central areas;
• able to counter-press immediately after turnovers;
• physically suited to a high-intensity league;
• and young enough to retain potential resale value.

The required role can then be translated into observable actions and measurable characteristics.

This prevents a common recruitment mistake: signing a generally impressive player without establishing how he will be used.

Role definition should involve coaches, recruitment specialists, analysts and sporting leadership. If those groups have different ideas about how the team should play, even an excellent model may produce irrelevant recommendations.

Step two: build a broad player universe

Traditional scouting is constrained by time. A scout can watch only a limited number of matches in sufficient detail.

Data allows a club to begin with a much larger universe. Thousands of players can be screened across leagues, age groups and positions.

The purpose of the initial model is not necessarily to choose the final signing. It is to reduce a large market to a manageable group requiring deeper investigation.

Basic filters might include:

• age;
• position;
• minutes played;
• contract status;
• estimated cost;
• league eligibility;
• physical requirements;
• and role-specific performance.

Clubs must be careful not to make the filters unnecessarily rigid. An unusually strict age limit could exclude a late-developing player. Position labels can also be misleading because two players listed as central midfielders may perform completely different functions.

The initial search should create possibilities, not manufacture certainty.

Step three: select metrics that describe the role

Recruitment analysis becomes more useful when metrics are selected because they relate to the required role.

A striker’s goals per 90 minutes may be relevant, but it does not explain how those goals were produced. Analysts might also examine:

• non-penalty expected goals;
• shots and touches inside the penalty area;
• runs behind the defensive line;
• pressing actions;
• link play;
• aerial ability;
• and the sustainability of finishing performance.

A progressive midfielder might be assessed using:

• passes and carries into advanced areas;
• ball retention under pressure;
• expected threat created;
• central progression;
• turnover frequency;
• defensive coverage;
• and the difficulty of attempted passes.

The relevant statistics depend on what the club wants the player to do.

Our guide to which football statistics actually matter explains why useful analysis must distinguish between descriptive numbers, contextual evidence and metrics with genuine predictive value.

Step four: adjust performance for context

Raw player statistics cannot be compared fairly without considering the environment in which they were produced.

A winger playing for a dominant team may receive the ball higher up the pitch and face isolated defenders more frequently. Another winger may start deeper, play with weaker teammates and spend long periods without possession.

Contextual adjustments may include:

• team possession;
• opposition strength;
• league quality;
• game state;
• tactical role;
• set-piece responsibility;
• penalty taking;
• player age;
• and the strength of surrounding teammates.

Suppose two midfielders complete six progressive passes per 90 minutes. The total appears identical.

The first plays for a team averaging 65% possession and attempts 70 passes per match. The second plays for a team averaging 42% possession and attempts only 38 passes. Their six progressive passes may represent very different levels of involvement and difficulty.

Possession-adjusted and opportunity-adjusted metrics can make the comparison more informative. They do not eliminate every contextual difference, but they reduce the risk of treating unequal situations as equivalent.

Step five: estimate league and team transferability

One of the hardest recruitment questions is whether performance will transfer from one environment to another.

A player moving between leagues may encounter differences in:

• pace and physical intensity;
• defensive pressure;
• tactical organisation;
• available space;
• refereeing;
• climate and travel;
• and the technical quality of teammates and opponents.

Clubs can study previous transfers between competitions to estimate how particular statistics tend to change. If forwards moving from one league to another generally experience a reduction in shot volume, the model can adjust its projection accordingly.

However, league adjustments should not be treated as universal conversion rates. Different skills transfer differently.

A player’s physical dominance may decline sharply against stronger opponents, while decision-making, movement or technical security might remain valuable. The destination club’s tactical system can also make the transition easier or harder.

Models provide a baseline forecast. Scouting and football expertise must investigate the mechanism behind that forecast.

Step six: project future performance rather than rewarding the past

Transfer fees are paid for future contribution, not historical statistics.

A club must estimate what the player is likely to become during the proposed contract. That forecast may consider:

• age and expected development;
• career minutes;
• injury history;
• physical maturation;
• technical trends;
• role stability;
• coaching environment;
• and comparable players.

Young players often command a premium because they may improve and retain resale value. Yet youth alone is not evidence of potential. Some young players have already developed close to their likely peak, while some older players possess skills that should decline relatively slowly.

Projection models can identify development patterns among comparable footballers, but every comparison contains uncertainty. Career paths are affected by injuries, opportunities, coaching, motivation and adaptation.

A responsible model should express a range of potential outcomes rather than a single precise forecast.

Step seven: use player similarity carefully

Player-similarity models search for footballers with comparable statistical profiles.

They can help a club find:

• alternatives to an expensive first-choice target;
• possible replacements for a departing player;
• players performing similar roles in less visible leagues;
• and footballers whose official position hides their tactical suitability.

However, similar output does not guarantee identical ability.

Two centre-backs may have comparable passing statistics because one breaks opposition lines and the other completes safe passes within a possession-heavy team. Two forwards may generate similar xG while relying on completely different movement patterns.

Similarity models should therefore generate questions rather than final answers. They identify players worth examining, after which video and scouting determine whether the similarity is real and relevant.

Step eight: review the player on video

Data can identify unusual performance, but video helps explain how it was achieved.

Once a player passes the initial screening process, analysts can review specific actions rather than watching matches without a defined question.

For example, the data may suggest that a midfielder progresses the ball effectively. Video analysis can investigate:

• whether he receives under pressure;
• which foot and body orientation he uses;
• whether passes break defensive lines;
• how quickly he recognises forward options;
• whether progression depends on a particular teammate;
• and what happens when the preferred passing lane is unavailable.

This is more informative than confirming that the player has accumulated a high number of progressive passes.

Video also reveals important off-ball behaviours that basic event data may miss, including positioning, scanning, recovery effort, defensive communication and movement that creates space for teammates.

Step nine: add live scouting and human evidence

Live scouting remains important because not every relevant quality can be reliably measured in available data.

A scout at the stadium can observe:

• communication away from the ball;
• responses to mistakes;
• interaction with coaches and teammates;
• positioning beyond the camera frame;
• warm-up and physical movement;
• and behaviour during breaks in play.

Recruitment teams may also speak with former coaches, teammates and other trusted contacts. The aim is to understand professionalism, learning ability, resilience and likely cultural adaptation.

This information is valuable but potentially biased. A single opinion should not be converted into a permanent judgement without examining its source and context.

Structured scouting reports can reduce this risk by asking consistent questions and separating observations from interpretations.

Step ten: assess tactical fit

A player’s performance depends partly on the system around him.

Recruitment analysts must consider how the player’s role will change after the transfer.

A centre-back joining a high-possession team may face fewer defensive actions overall but greater exposure to counterattacks. He may also be required to receive the ball under pressure and defend large spaces behind the defensive line.

A forward moving to a stronger club may receive more chances but face deeper opposition blocks. If his main strength is attacking open space, improved teammates do not guarantee improved individual performance.

A tactical-fit assessment should ask:

• Which current actions will remain available?
• Which new actions will be required?
• How much time and space will the player receive?
• Can the player perform in different game states?
• Which weaknesses can the team structure protect?
• Which strengths will the new system amplify?

The objective is not to find a player with no weaknesses. Such a player is unlikely to be affordable. The objective is to determine whether the player’s strengths are valuable and whether his limitations are manageable.

Step eleven: estimate the player’s financial value

Finding a suitable player is only part of the recruitment problem. The club must establish what it should be willing to pay.

The total investment can include:

• transfer fee;
• salary;
• signing bonus;
• agent fees;
• performance bonuses;
• sell-on clauses;
• and the cost of using a squad place.

The club then compares that cost with a range of possible sporting and financial outcomes.

A younger player might offer substantial resale potential but carry greater performance uncertainty. An older player may provide more predictable short-term contribution with little future transfer value.

Contract structure can change the balance. A lower initial fee with performance-related payments may reduce downside risk. A sell-on clause may help complete the deal but limit future profit.

Football valuation is therefore similar to probabilistic investment. The club is purchasing an uncertain stream of future contribution, not a guaranteed level of performance.

Step twelve: compare multiple targets rather than falling in love with one

Recruitment decisions become dangerous when a club becomes emotionally committed to one target.

A player may fit the role, but the transfer stops representing value if the price rises too far. Strong recruitment teams preserve alternatives.

A shortlist might include:

• a high-cost player with a relatively reliable projection;
• a younger player with more upside and more uncertainty;
• an experienced short-term option;
• a player from an overlooked competition;
• and an internal academy candidate.

These options should be compared using consistent criteria.

The purpose is not to create a mechanical final ranking. It is to make the trade-offs visible. Decision-makers can then see what they are paying for and which risks accompany each option.

A practical example: finding an undervalued midfielder

Imagine a mid-table club needs a central midfielder who can receive under pressure and progress the ball through the centre of the pitch.

The club’s data model identifies a 22-year-old playing for a lower-ranked team in a less visible European league.

His raw statistics do not look exceptional:

• few goals or assists;
• moderate pass completion;
• and limited international recognition.

Closer analysis reveals a more interesting profile.

His team averages little possession, but he receives a high proportion of its passes under pressure. When opportunities are considered, he ranks strongly for central progression, carries through midfield and passes that increase possession value.

Video shows that he scans before receiving, turns away from pressure and finds forward options quickly. His lower pass-completion rate partly reflects the difficulty of the passes he attempts.

The recruitment team then investigates the remaining questions:

• Will his physical performance translate to a faster league?
• Can he defend within the club’s pressing system?
• Does he retain the ball because of genuine technical security or weaker opposition pressure?
• How quickly could he adapt culturally and tactically?
• What salary and fee would make the risk worthwhile?

The player is not undervalued because a model produces a high rating. He becomes a potential value opportunity because the club has developed a better explanation of his current performance and future role than the wider market appears to possess.

How specialist football intelligence companies contribute

Not every club can build extensive data, modelling and scouting infrastructure internally.

Specialist football intelligence companies can provide data, recruitment models, player evaluation, opposition analysis and decision support.

Public information about proprietary systems is necessarily limited. However, reported services associated with organisations such as Jamestown Analytics illustrate how data and scouting can be connected across recruitment workflows.

Our guide to how Jamestown Analytics works examines the publicly reported process while separating confirmed information from reasonable inference.

External analysis does not remove the club’s responsibility for the final decision. The club must still determine whether the recommended player fits its manager, squad, finances and long-term strategy.

Why data-driven recruitment still produces failures

No recruitment system can eliminate uncertainty.

A well-researched transfer can fail because of:

• injury;
• poor tactical fit;
• managerial change;
• limited playing time;
• difficulty adapting to a new country;
• unexpected physical development;
• or normal variation in performance.

Models can also be wrong. The historical data may be incomplete, the league adjustment may be unreliable or the player’s role may be misunderstood.

The relevant test is not whether every signing succeeds. It is whether the process produces better decisions than realistic alternatives across a sufficiently large sample.

This is one reason clubs associated with professional betting experience often emphasise portfolio thinking. Individual outcomes are uncertain, but a repeatable valuation process can still create an advantage across many decisions.

Our analysis of why clubs owned by professional bettors often overperform explores how probability, valuation and organisational discipline may transfer into football operations.

How clubs should evaluate recruitment decisions

Recruitment should not be evaluated only by whether the player became a star.

Clubs should compare the original forecast with what subsequently happened.

Relevant measures may include:

• minutes played;
• availability;
• contribution within the intended role;
• development against the original projection;
• team performance with and without the player;
• salary relative to contribution;
• and eventual resale value.

The club should also review the reasoning behind the decision.

Was the role correctly defined? Did the model identify the right strengths? Was a risk known but accepted, or completely missed? Did the player fail because the environment changed after the transfer?

This distinction prevents the organisation from learning the wrong lesson.

A successful signing can result from a weak process and good fortune. An unsuccessful signing can follow a reasonable decision affected by an unpredictable injury. Both cases deserve careful review.

The real competitive advantage is the recruitment system

Public discussion often focuses on whether a club has discovered a secret statistic or algorithm.

Individual models can provide an advantage, but sustainable recruitment performance usually depends on the whole system:

• clear tactical requirements;
• broad and reliable data coverage;
• context-sensitive models;
• structured scouting;
• independent medical and character assessment;
• financial discipline;
• fast execution;
• and honest post-transfer evaluation.

Data has little value if the manager refuses to use the player. Scouting reports have limited value if they cannot be compared consistently. Accurate valuations do not help if executives repeatedly exceed them under pressure.

The strongest data-driven clubs create an organisation capable of turning information into decisions.

This is also the central lesson from what football clubs can teach bettors about finding value: an advantage comes from disciplined valuation and execution, not simply from knowing which player or team is likely to perform well.

Key Takeaways

• An undervalued player is worth more to a particular club than the total expected cost of signing and employing him.
• Data-driven recruitment begins by defining a tactical role rather than ranking players in the abstract.
• Statistical screening helps clubs search a much larger player universe than traditional scouting alone.
• Raw performance must be adjusted for team quality, possession, opposition, league strength, role and game state.
• Recruitment models must project future contribution rather than simply reward past output.
• Video and live scouting help explain the mechanisms behind a player’s statistics.
• Tactical suitability can make the same footballer valuable to one club and unsuitable for another.
• Transfer fees are only one part of the investment; salary, bonuses, agent fees and resale potential also matter.
• Strong clubs maintain several recruitment options and remain willing to abandon a target when the price removes the value.
• No model eliminates uncertainty, so the recruitment process must be evaluated across many decisions rather than one successful or unsuccessful signing.

Related Guides

• What Football Clubs Can Teach Bettors About Finding Value
• Recruitment Models vs Betting Models
• How Does Jamestown Analytics Work?
• Why Clubs Owned by Professional Bettors Often Overperform

Understand the process behind football decisions

Subscribe to the GoalIQAI newsletter for evidence-based guides to football analytics, recruitment, probability and betting-market intelligence. Learn how data-driven organisations evaluate performance, price uncertainty and make better long-term decisions.