How Does Jamestown Analytics Work?

Jamestown Analytics helps football clubs make better decisions by turning proprietary football data into club-specific assessments of players, coaches, teams and performance. Its work reportedly supports player recruitment, squad planning, head-coach selection and opposition analysis.

The precise algorithms remain confidential. Public evidence suggests that Jamestown starts with a large proprietary dataset, measures performance beyond familiar headline statistics, adjusts for competition and context, and then tailors its analysis to each client’s budget, squad and playing style.

The output does not automatically sign a player or appoint a coach. It helps a club identify what “good” looks like, narrow a large market into relevant candidates and challenge subjective opinions. Human decision-makers still assess personality, medical risk, tactical fit and affordability. Jamestown is therefore best understood as a football intelligence partner—not a machine that replaces scouts or guarantees successful decisions.

What Is Jamestown Analytics?

Jamestown Analytics is a specialist football intelligence company based in London. It emerged from the same analytical environment as Starlizard, the sports betting consultancy closely associated with Brighton & Hove Albion owner Tony Bloom.

The two organisations have related origins but different primary purposes:

• Starlizard analyses sporting events and betting markets.
• Jamestown Analytics applies proprietary football intelligence to club decisions.

Jamestown’s managing director, Justin Said, previously spent 13 years working at Starlizard. Public reporting indicates that Jamestown uses the same underlying supply of proprietary football data but refines and adapts it for individual clubs.

Tony Bloom is closely connected to the wider analytical ecosystem but is not reported to own Jamestown. The company has instead been identified as being owned by sports bettor and backgammon player Johan Moazed.

These organisational distinctions are explained further in GoalIQAI’s comparison of Starlizard and Jamestown Analytics.

What Is Publicly Known About the Process?

Jamestown does not publish its models, variables or player ratings. Any detailed explanation must therefore distinguish between three levels of confidence:

• Publicly confirmed: Information disclosed by Jamestown representatives, partner clubs or credible reporting.
• Reasonable inference: Processes commonly required to turn football data into recruitment decisions.
• Unknown: The exact algorithms, weightings, data definitions and internal ratings.

Public reporting has established several important features.

Jamestown uses proprietary data rather than relying only on familiar public statistics. It adapts its analysis to each club rather than distributing one universal list of players. It supports decisions involving both players and coaches. It also maintains an ongoing dialogue with client clubs rather than delivering a single static report.

Hearts sporting director Graeme Jones described this relationship as a constant dialogue. Jamestown analyses the club’s performances, provides feedback and helps define what good performance should look like.

What remains unknown is more technically significant:

• Exactly which events and player actions are collected.
• How each competition is rated.
• How player performance is adjusted for teammates and opposition.
• How tactical roles are classified.
• How future development is forecast.
• How much importance is assigned to each variable.

The company’s competitive value depends partly on keeping those answers private.

The Jamestown Process in Simple Terms

Although the internal models are secret, the publicly visible process can be understood as a sequence:

1. Collect detailed proprietary football data.
2. Convert match events into meaningful performance measures.
3. Adjust those measures for role, competition and context.
4. Define what the client club needs.
5. Search the player or coaching market for suitable candidates.
6. Estimate performance, fit, development and value.
7. Present evidence to the club’s decision-makers.
8. Combine the analysis with scouting and human assessment.
9. Monitor results and update the models.

This is a simplified reconstruction rather than Jamestown’s disclosed algorithm. Each stage may contain hundreds of variables and multiple analytical models.

Stage One: Building a Proprietary Football Dataset

The starting point is data.

Most professional football data records events such as:

• Passes.
• Shots.
• Carries.
• Pressures.
• Duels.
• Recoveries.
• Touches.
• Defensive actions.
• Player and ball locations.

Public platforms provide increasingly sophisticated versions of this information. However, the best-known metrics do not represent the complete analytical possibilities available to a private company.

A proprietary dataset may differ through:

• Greater competition coverage.
• Different definitions of football actions.
• More detailed positional information.
• Custom quality controls.
• Variables derived from combinations of actions.
• Longer historical records.
• Ratings that are unavailable to competitors.

The quality of the raw data matters because every later conclusion depends on it. A sophisticated model cannot fully compensate for incomplete, inconsistent or incorrectly classified inputs.

This is also why two analytics companies can study the same match and reach different conclusions. They may not be measuring the same events in the same way.

Stage Two: Measuring Actions That Contribute to Winning

Raw event counts have limited value without interpretation.

A midfielder completing 60 passes has not necessarily played better than one completing 35. The value depends on where the passes started, where they ended, what defensive pressure existed and whether they improved the team’s position.

Jamestown’s models are likely to ask questions such as:

• Did the action increase the probability of creating a chance?
• Did it move the ball through an important defensive line?
• How difficult was the action?
• What alternative options were available?
• Did the player retain possession under pressure?
• Did a defensive action prevent a dangerous progression?
• How did the action affect the team’s overall structure?

Expected goals assigns a value to shots. More advanced possession models attempt to value the actions that happen before a shot is taken.

GoalIQAI’s guide to expected threat explains one public method for estimating how passes and carries change the probability of scoring. Jamestown may use conceptually related ideas, but its exact measures are proprietary and should not be assumed to be standard xT.

The analytical advantage is unlikely to come from one remarkable statistic. It is more likely to emerge from the combination of better data, better contextual adjustments and better decision processes.

Stage Three: Adjusting Performance for Context

A player’s statistics are partly produced by his environment.

A striker playing for the strongest team in a weak league may receive more chances than an equally talented striker playing for a relegation candidate. A centre-back in a possession-dominant team may complete more passes because opponents rarely press him aggressively.

Meaningful player comparison therefore requires contextual adjustment.

Factors may include:

• League strength.
• Opponent quality.
• Team quality.
• Tactical role.
• Possession share.
• Game state.
• Home advantage.
• Age and physical development.
• Minutes played.
• Quality of teammates.

This is one of the most difficult parts of football analytics.

Suppose two wingers each create 0.25 expected assists per 90 minutes. Those figures appear identical. However:

• Player A plays for the league champion.
• Player B plays for a team that averages 39% possession.
• Player A takes corners.
• Player B creates almost everything from open play.
• Player A is 27.
• Player B is 20.

A raw comparison conceals information that could materially affect recruitment value.

Jamestown’s ability to adjust performance between leagues may be especially important. Clubs want to know not simply whether a player has performed well, but whether his skills are likely to survive a move into a faster, stronger or tactically different competition.

Stage Four: Defining What the Club Needs

A high-rated player is not automatically the right player for every club.

Jamestown has said that it refines and tailors its output for individual clients. This suggests that the process begins with a clear understanding of the club.

The relevant questions may include:

• What style does the team want to play?
• Which positions need strengthening?
• What responsibilities must each role perform?
• What transfer fees and wages are affordable?
• Does the club need immediate performance or future development?
• Which leagues are accessible under registration rules?
• What resale profile does the ownership model require?
• How does the existing squad affect the type of player needed?

A club trying to avoid relegation may prioritise immediate reliability. A development-focused club may accept greater short-term uncertainty in return for a younger player with a higher potential ceiling.

The same player could therefore be:

• A strong recommendation for one club.
• Too expensive for a second.
• Tactically unsuitable for a third.
• Too old for the trading model of a fourth.

This is why Jamestown’s service should not be understood as one global ranking of footballers. It is more likely to produce conditional judgements: good for which role, at which club, in which league and at what price?

Stage Five: Searching the Global Player Market

Once a club’s requirements are defined, data allows the search to expand beyond players already known to local scouts.

A traditional recruitment department may begin with personal knowledge, live scouting networks and agent recommendations. These sources remain valuable, but they are naturally limited by time and geography.

A database can search thousands of players across many leagues in seconds.

An initial search for a defensive midfielder might include filters for:

• Age.
• Minutes played.
• Ball recoveries.
• Pressure resistance.
• Progressive passing.
• Defensive positioning.
• Aerial ability.
• Estimated transfer cost.
• Contract status.

The important advantage is not simply speed. It is the ability to find players who do not match conventional scouting assumptions.

A player may be overlooked because:

• He plays in an unfashionable league.
• His team is performing poorly.
• His most valuable actions are not visible in headline statistics.
• He performs an unusual tactical role.
• His underlying development is stronger than his reputation.
• His club or league receives limited scouting coverage.

Jamestown’s reported success has often involved identifying ability before it becomes obvious to the wider transfer market.

Stage Six: Comparing Players by Role Rather Than Position

Traditional positional labels can be too broad for advanced recruitment.

Two players may both be classified as full-backs while performing completely different jobs. One overlaps aggressively and delivers crosses. The other moves into midfield during possession and helps control transitions.

A recruitment model must therefore move beyond simple labels such as:

• Defender.
• Midfielder.
• Winger.
• Striker.

More useful role profiles might distinguish between:

• A possession-building centre-back and a penalty-area defender.
• A ball-winning midfielder and a deep progressor.
• A touchline winger and an inside forward.
• A penalty-box striker and a forward who links play.

The model can then compare players who perform similar functions, even if they play in different formations or are publicly classified under different positions.

This helps reduce a common recruitment mistake: signing a player because his overall numbers look strong without understanding why those numbers were produced.

Stage Seven: Estimating Future Performance

Football recruitment is forward-looking. Clubs are buying what a player may do next, not what he did last season.

Historical performance is evidence, but the real question is whether it can be sustained or improved in a new environment.

A development forecast may consider:

• Age.
• Career minutes.
• Physical development.
• Rate of improvement.
• Injury history.
• Positional experience.
• League difficulty.
• Similarity between current and proposed roles.

Age curves are particularly relevant. Different attributes tend to develop and decline at different stages of a career. Decision-making or positioning may improve after explosive speed begins to fall.

A 19-year-old and a 28-year-old producing identical performances do not carry the same future distribution.

The younger player may offer:

• More development potential.
• Greater uncertainty.
• A longer possible contribution.
• Higher resale value.

The older player may provide:

• More reliable evidence.
• Immediate readiness.
• Lower adaptation risk.
• Less future transfer value.

Neither profile is automatically superior. The correct choice depends on what the club requires.

Stage Eight: Estimating Transfer-Market Value

Identifying a strong player is not enough. A recruitment department must determine whether the cost represents value.

A footballer can be:

• Excellent but overpriced.
• Average but appropriately priced.
• Unproven but materially undervalued.

The relevant cost is broader than the transfer fee. It can include:

• Wages.
• Signing fees.
• Agent fees.
• Performance bonuses.
• Sell-on clauses.
• Contract length.
• Expected resale value.
• The cost of using a squad or registration place.

A player’s football value and financial value are related but separate.

For example, two candidates may be expected to improve the team by a similar amount. One costs £15 million and is approaching his peak. The other costs £5 million, is 20 years old and could retain substantial resale value.

The second player may offer the stronger overall proposition, even if his immediate performance is slightly less certain.

This resembles the logic of value betting. The objective is not to find the player most likely to succeed at any cost. It is to find situations where expected ability and market price appear misaligned.

Stage Nine: Human Scouting and Due Diligence

Data can reduce a large market to a manageable shortlist. It cannot fully explain the person behind the numbers.

Before completing a transfer, clubs still need to investigate:

• Personality.
• Professionalism.
• Communication.
• Adaptability.
• Family circumstances.
• Medical risk.
• Training behaviour.
• Response to pressure.
• Willingness to accept the proposed role.

A player may be statistically suitable but unwilling to relocate. He may struggle with the language, dislike the coach’s methods or possess an injury risk that is not visible in event data.

This is why serious data-led clubs do not eliminate scouting. They use data to make scouting more targeted.

A simplified division of responsibility is:

• Data asks: Who deserves closer attention?
• Video asks: How are the numbers being produced?
• Live scouting asks: What is visible beyond recorded events?
• Due diligence asks: Will the person succeed in this environment?
• Club leadership asks: Is the complete risk worth the price?

Djurgården chief executive Hampus Frisén made this balance explicit when discussing the Swedish club’s 2026 Jamestown partnership: analytical evidence would be combined with football’s softer human factors rather than treated as a replacement for them.

How Jamestown Supports Head-Coach Recruitment

Jamestown is reported to analyse coaches as well as players.

Managerial appointments are difficult because results are heavily affected by club resources, squad quality and opposition strength. A coach who finishes eighth with a small budget may have performed better than one finishing fourth with the league’s most expensive squad.

A coaching model may attempt to separate the manager’s contribution from the environment by examining:

• Performance relative to squad strength.
• Chance creation and prevention.
• Pressing style.
• Possession structure.
• Set-piece performance.
• Player development.
• Adaptability between game states.
• Consistency across clubs.

The next stage is matching the coach to the client club.

A highly rated manager may still be unsuitable if he requires players the club cannot afford, prefers a style inconsistent with the squad or has limited experience developing young talent.

Hearts used Jamestown during processes that produced the appointments of Neil Critchley and later Derek McInnes. The contrasting outcomes illustrate why analytics should be evaluated probabilistically. A good process can reduce uncertainty without eliminating it.

How Jamestown Analyses Team Performance

Jamestown’s work does not appear to end once a player is signed or a coach is appointed.

At Hearts, the company reportedly analyses every performance and provides ongoing feedback through sporting director Graeme Jones.

This can help distinguish results from underlying performance.

A team may win three consecutive matches while:

• Creating fewer good chances than its opponents.
• Scoring from low-probability shots.
• Relying on exceptional goalkeeping.
• Allowing increasing territorial pressure.

Alternatively, a team may lose several matches despite generating stronger expected goals and repeatedly reaching dangerous areas.

The analytical question is not merely whether the team won. It is whether its performances contain processes that are likely to remain successful.

This is the same distinction GoalIQAI applies when explaining how to analyse football form properly. Results matter, but they do not contain all the information needed to estimate future performance.

What Does “They Tell You What Good Is” Mean?

Graeme Jones summarised Jamestown’s role at Hearts with the phrase: “They just tell you what good is.”

This does not necessarily mean Jamestown gives every player or performance a simple score.

It suggests the company provides an objective reference point.

For example, Jamestown could help a club understand:

• What elite chance creation looks like for a particular role.
• Whether the team’s pressing is genuinely effective.
• Which player actions contribute most to winning.
• How a performance compares with relevant opponents.
• Whether a target has the attributes required to improve the squad.

Without a reliable benchmark, football decisions can become relative to the people already inside the club. A scout may consider a player impressive because he is better than the club’s current option, while the data may show that neither meets the required level.

Defining “good” creates a consistent standard against which players, coaches and performances can be assessed.

How the Feedback Loop Improves the Process

A serious analytical system should learn from its decisions.

After a player signs, the club and analytics provider can compare predicted and actual performance.

Questions may include:

• Did the player adapt as expected?
• Were his previous league statistics translated correctly?
• Was his role at the new club accurately predicted?
• Did physical or personal factors affect the outcome?
• Was the transfer price justified?
• Which warning signs were missed?

Successful predictions can strengthen confidence in the process. Failures can reveal missing variables or incorrect assumptions.

This feedback loop is essential because football changes. Tactical trends evolve, leagues improve or decline, and recruitment markets become more efficient as competitors adopt similar methods.

A model that worked five years ago cannot simply be preserved unchanged.

Why Can Jamestown Find Players Other Clubs Miss?

Jamestown’s potential advantage appears to come from the interaction of several factors rather than one secret formula.

• Proprietary data: Competitors may not have access to the same information.
• Global coverage: More competitions create more opportunities to find overlooked talent.
• Contextual adjustment: Performance can be compared across different environments.
• Role-based analysis: Players are judged according to the job a club needs performed.
• Price sensitivity: Ability is considered alongside transfer cost and future value.
• Selective clients: Jamestown can work with clubs willing to implement its recommendations seriously.
• Continuous feedback: Decisions and performances can be used to refine future analysis.

Other clubs may possess good data but fail to convert it into decisions. Internal politics, recruitment hierarchies or resistance from coaches can prevent analysis from influencing the final choice.

The implementation advantage may therefore be as important as the technical model.

Why Does Jamestown Work With a Selective Group of Clubs?

Jamestown has described itself as selective about its partners.

There are several possible reasons for limiting the client base.

Competitive conflict: Giving identical intelligence to direct rivals could reduce its value.

Implementation quality: The company may prefer clubs prepared to integrate evidence into real decisions.

Customisation: Tailoring analysis requires time and specialist attention.

Confidentiality: A small network may make proprietary information easier to protect.

Market exclusivity: A club may receive exclusive access within its domestic competition.

Hearts have been described as Jamestown’s exclusive Scottish partner. This could provide a meaningful advantage if rivals cannot purchase the same service.

The current public client network is examined in GoalIQAI’s guide to which clubs are connected to Jamestown Analytics.

What Jamestown Analytics Cannot Guarantee

No football model can guarantee that a signing, coach or tactical decision will succeed.

The company must operate with incomplete information. Its recommendations may be affected by:

• Random performance variation.
• Unexpected injuries.
• Personal adaptation.
• Changes in coaching.
• Incorrect role assumptions.
• Limited data from smaller competitions.
• Transfer negotiations.
• Other clubs competing for the same player.

A model may correctly identify a talented player who then suffers a serious injury. It may recommend a suitable coach who loses key players shortly after arriving. It may value a transfer accurately before an auction pushes the price beyond the point of value.

The correct standard is not perfection. It is whether the process produces better decisions, on average, than the alternatives.

Why Public Success Stories Can Be Misleading

Jamestown is frequently associated with successful signings such as Moisés Caicedo and Kaoru Mitoma. These examples demonstrate the possible upside of the model, but they do not reveal its complete record.

Public analysis faces several limitations:

• Unsuccessful recommendations may remain confidential.
• A successful player may have been identified independently by club scouts.
• The model’s preferred target may not have been signed.
• A rejected player may later succeed elsewhere.
• The contribution of coaching and development is difficult to isolate.

This creates survivorship and attribution problems.

To evaluate Jamestown properly, an outsider would need access to all recommendations, rejected candidates, valuations and predicted outcomes. That information is not public.

The available evidence supports the conclusion that Jamestown is influential and respected. It does not allow an independent calculation of the model’s precise accuracy or financial return.

How Is Jamestown Different From Public Scouting Platforms?

Public and commercial scouting platforms can provide video, event data and player-search tools to many clubs. Jamestown appears to offer something more customised.

A platform can show a club what happened. A football intelligence partner attempts to explain what matters, how it should be interpreted and what the club should investigate next.

What Can Football Bettors Learn From Jamestown?

Bettors do not have access to Jamestown’s data or algorithms, but its apparent process contains several transferable principles.

Start with a clear question. Do not collect statistics without knowing what decision they are intended to support.

Measure underlying performance. Results alone can conceal whether a team’s process is sustainable.

Adjust for context. Statistics are shaped by opponents, teammates, tactics and game state.

Compare like with like. A player or team should be assessed relative to the role and situation being analysed.

Always include price. A strong team, player or idea can still be overvalued.

Combine evidence. Data, tactical analysis and contextual information answer different parts of the question.

Review decisions over a portfolio. One outcome cannot validate or invalidate a probabilistic process.

This resembles the professional framework explained in how professional football bettors analyse a match.

A Simplified Jamestown-Style Recruitment Example

Imagine a club needs a young central midfielder who can receive under pressure and progress the ball through compact defensive structures.

A simplified analytical process might be:

1. Define the tactical responsibilities of the role.
2. Identify relevant performance measures.
3. Search multiple leagues for players meeting minimum thresholds.
4. Adjust output for possession, opponent strength and league quality.
5. Remove players who exceed the club’s budget or age profile.
6. Compare expected performance in the destination league.
7. Review video to understand how the actions occur.
8. Conduct live scouting and character checks.
9. Estimate the transfer’s sporting and financial value.
10. Make the final decision through the club’s recruitment structure.

The result may be a player whose traditional statistics appear unremarkable but whose contextual profile matches the club’s requirements.

This is the essential logic behind finding hidden value: not discovering that a famous player is good, but identifying useful ability before the wider market prices it correctly.

Key Takeaways

• Jamestown Analytics turns proprietary football data into club-specific assessments of players, coaches and performance.
• The exact algorithms, variables and weightings remain confidential.
• Its process appears to combine data collection, contextual adjustment, role profiling, market search and valuation.
• Jamestown does not simply distribute one universal ranking of players to every client.
• Recommendations are tailored to a club’s squad, tactics, budget and development model.
• The system can support player recruitment, coach selection, squad planning and ongoing performance analysis.
• Human scouting, medical work and character assessment remain essential.
• The company’s advantage may come from implementation and decision discipline as much as from its proprietary models.
• Public success stories do not reveal Jamestown’s complete recommendation record.
• The correct measure is whether the process improves decisions over time—not whether every signing succeeds.

Related Guides

• Inside Jamestown Analytics
• Which Football Clubs Are Connected to Jamestown Analytics?
• Starlizard vs Jamestown Analytics
• What Football Statistics Actually Matter?
• What Is Value Betting?

Understand How Football Intelligence Creates Value

GoalIQAI explores how professional analysts use data, probability and market thinking to make better football decisions. Subscribe to the newsletter for new guides covering football intelligence, recruitment analytics and evidence-based betting analysis.