Abstract

This study explores player performance across Europe’s top five football leagues (Premier League, La Liga, Bundesliga, Serie A, and Ligue 1) to identify undervalued players—those whose statistics suggest high-level performances but whose market values do not reflect their contributions. By analyzing key performance indicators such as goals (G), assists (A), progressive passes, and defensive contributions on a per-90-minute basis, we aim to highlight players who represent strategic investment opportunities in the transfer market.

Introduction

In modern football, market value does not always align with performance. While some players attract enormous transfer fees due to reputation and media exposure, others quietly deliver elite performances without comparable financial recognition.

This research investigates whether players performing at a high level remain undervalued compared to their counterparts. By analyzing advanced performance metrics across positions, we assess whether statistical output correlates with market valuation. Our goal is to uncover hidden gems in the football market—players who provide exceptional contributions but are overlooked in financial valuations.

Data & Methodology

This study uses player performance data from fbref.com and footballtransfers.com, focusing on footballers in the top five European leagues. The dataset covers:

• Attacking statistics (e.g., goals, assists, expected goals)
• Passing and possession-based metrics
• Defensive contributions
• Estimated market values

To ensure meaningful comparisons, we excluded players with fewer than 394 minutes played, which represents two standard deviations below the mean. This threshold removes fringe players and ensures we evaluate individuals with a more substantial impact on their teams.

To fairly compare players with varying game time, all performance metrics were normalized per 90 minutes. To select the metrics which most strongly influence market valuations, we applied SHAP (Shapley Additive Explanations) analysis. SHAP values quantify the impact of each feature on model predictions, allowing us to identify the most relevant variables.

To predict market values, we trained three machine learning models:

• Random Forest Regressor
• Gradient Boosting Regressor
• XGBoost Regressor

These models were chosen for their ability to capture complex, nonlinear relationships in player valuation, outperforming traditional linear models. Among them, Random Forest Regressor performed best, excelling in handling feature interactions and nonlinear dependencies.

Each model was evaluated using Mean Squared Error (MSE) and R² scores, selecting the best-performing model for final predictions. We then compared each player’s predicted market value with their actual transfer value, classifying them as:

• Undervalued: Predicted value > Actual value
• Overvalued: Predicted value < Actual value

Results

Market Value Distribution

Market valuations in football follow a highly skewed distribution, with most players valued below €20 million, while a small elite group exceeds €50 million. The market is dominated by a few high-value superstars who command extraordinary transfer fees, reinforcing the notion that financial valuation does not always reflect contribution.

To illustrate this, in Figure 1 we plotted mean and standard deviation markers:

• The mean market value is pulled upward by elite players.
• The majority of players fall within one standard deviation of the mean, while extreme outliers occupy the highest valuation tiers.
• The estimated market value at two standard deviations (57.44 million euros) highlights the extreme skewness of the market. Out of 1,421 players, 95% are below this threshold, while only 5% exceed it, emphasizing the concentration of high-value players in a small elite group.

Market Value vs. Position Analysis

Figure 2 provides insights into the distribution of player market values across different positions: defenders (DF), goalkeepers (GK), forwards (FW), and midfielders (MF). Similar to how certain metrics strongly correlate with a team’s success in the Champions League, this visualization highlights potential disparities in valuation based on position rather than individual performance.

The median market values for forwards (FW) and midfielders (MF) are noticeably higher than those of defenders and goalkeepers. This aligns with the general trend in football, where attacking and playmaking players attract more media attention and command higher transfer fees.

Despite their critical role in a team’s success, defenders (DF) and goalkeepers (GK) exhibit lower median values, with fewer extreme outliers compared to forwards. This suggests that elite defensive talents might be overlooked in valuation, similar to how metrics like “Opponent Goals” 

and “Total Saves by Team” negatively correlated with success in our previous analysis. While goalkeepers making more saves often indicates defensive vulnerability, their individual contributions may still be undervalued in the market.

Presence of High-Value Outliers Across All Positions

While the general trend shows lower market values for defenders and goalkeepers, the presence of several high-value outliers in these positions suggests that exceptional talents do receive recognition like Lucas Chevalier, David Raya for the goalkeepers or Pau Cubarsí, Aurélien Tchouaméni for the defenders. However, their valuation is less consistently high compared to attacking players.

Undervalued and Overvalued Players: Insights from Market Valuation Analysis

Players whose predicted value exceeds their actual market price represent undervalued transfer targets:

Undervalued Players by League and Position

To better understand where undervalued talent is concentrated, we analyzed the distribution of such players across both leagues and positions.

As shown in Figure 3, the Premier League dominates with the highest count of undervalued players, nearly 50 players. This could reflect both the league’s intense competition and inflated market valuations, making it easier for high-performing but overlooked individuals to emerge.

Following the Premier League are the Bundesliga and Serie A, each contributing around 20–23 undervalued players. La Liga and Ligue 1 trail with smaller counts, suggesting either better valuation alignment or less market attention on individual performances.

In Figure 4, we observe that midfielders (MF) and forwards (FW) make up the bulk of undervalued players, with 45 and 41 players respectively. These positions often contribute directly to performance metrics like goals, assists, and chance creation, which our model heavily relies on.

Defenders (DF) are the next largest group, indicating that some defensive contributions particularly progressive passing or carrying may be underappreciated in current market valuations.

Notably, goalkeepers (GK) are largely absent from the undervalued group, with only a handful identified. This likely reflects both the specialized nature of goalkeeper metrics and the market’s cautious valuation approach for this position.

These distributions suggest that most undervalued opportunities lie in attacking and midfield roles, particularly within high-profile leagues like the Premier League. Clubs looking for under-the-radar talent may find the greatest return by scouting creative midfielders and efficient forwards who are overlooked due to age, club exposure, or injury history.

Overvalued Players by League and Position

In addition to identifying undervalued players, we also examined the distribution of overvalued players whose actual market values significantly exceed their model-predicted worth by league and position.

As seen in Figure 5, the Premier League once again leads, with over 65 overvalued players, reflecting the league’s inflated transfer market and media-driven valuations.
Serie A, La Liga, and the Bundesliga follow, while Ligue 1 shows the lowest number of overvalued players, suggesting more conservative market pricing or lower global exposure.

In Figure 6, midfielders (MF) are the most frequently overvalued group, followed by defenders (DF). This may be due to the market overemphasizing potential and brand value over measurable output, particularly in creative or central roles.

Interestingly, forwards (FW) often assumed to be the most overhyped appear less frequently overvalued in this analysis, possibly because their contributions (e.g. goals, as sists) align more closely with valuation expectations.

Goalkeepers (GK) show the fewest overvalued cases, likely due to the specialized nature of their metrics and the comparative market stability for this position.

These results reinforce that market hype, especially in high-profile leagues and midfield roles, can significantly skew valuations. Clubs aiming to avoid overpayment should scrutinize midfield and defensive signings from top leagues, where market value may be influenced more by perception than actual performance.

Top Over- and Undervalued players

These players have significantly higher predicted values than their actual market prices, suggesting they may be undervalued by the market.

Player	Actual Value (€M)	Predicted Value (€M)	Valuation Difference (€M)	Age
Mohamed Salah	30.2	78.7	+48.5	32
Romano Schmid	4.5	49.7	+45.2	25
Ante Budimir	4.9	41.7	+36.8	33
Moses Simon	11.4	38.7	+27.3	29
Lucas Da Cunha	7.8	32.8	+24.9	23

• Mohamed Salah is a standout case (+€48.5M difference) despite being 32, his goal contributions (1.54 G+A per 90), attacking touches, and dribbling metrics remain elite, yet his market value does not reflect this.
  
• Romano Schmid (+€45.2M) and Moses Simon (+€27.3M) show strong progressive play and attacking involvement, suggesting they are underappreciated creators in the market.
  
• Veteran striker Ante Budimir (33) is a strong finisher, and despite his age, his goal-scoring efficiency per 90 minutes remains high.
  
• Lucas Da Cunha (23) and Paul Nebel (22) represent young undervalued talents whose high ball-carrying distances and attacking touches suggest significant upside.

Clubs looking for smart investments should target attackers with high output but lower market recognition, as they may deliver elite-level contributions at a fraction of the cost.

Overvalued Players 

These players have significantly lower predicted values than their actual market valuations, suggesting their price tags may be inflated.

Player	Actual Value (€M)	Predicted Value (€M)	Valuation Difference (€M)	Age
Martin Ødegaard	134.5	42.5	-91.9	26
Declan Rice	118.5	39.5	-79.0	26
Pedri	129.7	59.8	-69.9	22
Federico Valverde	111.9	43.9	-67.9	26
Bukayo Saka	113.1	47.8	-65.2	23

• Martin Ødegaard (-€91.9M difference) and Bukayo Saka (-€65.2M) appear heavily overvalued, but this may be partly explained by their injury-affected seasons. Both players have had reduced playing time and lower statistical output due to injuries, which impacts the model’s valuation.
  
• Declan Rice (-€79.0M) and Federico Valverde (-€67.9M) are elite midfielders, but their market prices may be inflated relative to their actual statistical contribution.
  
• Pedri (-€69.9M) remains a high-value prospect, but his predicted valuation suggests that his output—especially in attacking play—is not yet at the level of similarly priced stars.

Some high-profile midfielders and playmakers may be slightly overvalued due to their reputation rather than their current statistical output. However, injury-related performance dips may also be influencing these predictions. Clubs should account for player availability and fitness trends when making transfer decisions.

Market-Wide Insights & Trends

• The mean valuation difference across all players is €26.8M, indicating an overall market overvaluation bias
  
• Attackers with strong goal contributions are often undervalued, whereas creative midfielders and big-name prospects tend to be overvalued
  
• Age remains a key factor—experienced players like Salah and Budimir appear deeply undervalued despite strong stats, while young stars maintain high market valuations regardless of form.
  
• Injury history plays a crucial role in valuation gaps—Ødegaard and Saka, for example, have suffered injuries that lowered their statistical output, likely influencing their predicted value.

Conclusion

This study reveals that football’s transfer market often misaligns player value with on-field performance. Using data-driven modeling, we identified undervalued players with strong statistical output, and overvalued stars whose market prices exceed their recent contributions.

Our findings show that efficient attackers are frequently undervalued, while high-profile midfielders and young prospects often remain overpriced. Injury history also plays a key role in valuation gaps.

To improve predictions, future research should explore the weight of individual metrics and include external factors like tactics, league strength, or contract status. With further refinement, valuation models could help clubs make smarter transfer decisions and uncover market inefficiencies more accurately.

Data source: www.fbref.com, www.footballtransfers.com