FIFA Football Players Dataset

By shrijeetverma13 · May 3, 2026

This dataset contains information about 19,667 football players, including ratings, potential, and market value. Datasource :

The analysis identified the top countries producing highly-rated players (Overall Rating 80) from the dataset. A bar chart and supporting data tables were generated showing the top 15 nations ranked by their count of elite players.

The analysis identified the top 10 teams by average Overall Rating and total squad value. Two bar charts were generated showing these rankings visually, along with supporting data tables for deeper exploration.

The highest valued players in the dataset are dominated by elite attackers and key midfielders. A horizontal bar chart and a pie chart have been generated to visualize the top 10 players by market value and the position distribution among the top 50 most valued players.

Among players aged 23 and under, the biggest gaps between current Overall Rating and Future Potential belong to a group of 16-year-old talents. Five players share the top gap of +26 points: Emre Tezgel (Stoke City), De Pievre Ilunga (Degerfors), Dylan Lobban (Aberdeen), Yoni Gomis (Le Havre), and Kerr Smith (Dundee United). A horizontal bar chart and data tables have been generated showing the top 15 young players ranked by their potential gap.

Player value and overall rating vary noticeably across the 17 positions analyzed. Two charts were generated: a combined bar and line chart showing average value (bars) and average rating (line) per position, and a scatter plot mapping average rating vs. average value with bubble sizes representing player counts. CF (Centre Forward) tops the value chart at $4.28M average, while attacking positions like ST, LW, and RW also rank highly. Defensive and goalkeeper roles tend to have lower market values. Interestingly, the SW (Sweeper) position has the highest average rating (75.0) but essentially no market value, likely due to very few players in that role.

Players reach their peak Overall Rating at age 35 (average rating of 71.24) and their peak Market Value at age 30 (average of $4.92M). This means market value peaks about 5 years earlier than overall rating. Two charts were generated showing these trends across all ages, along with the distribution of players by age.

A dual-axis bar chart has been created showing the average Value Per M$ (blue bars, left axis) and average Total Stats Score (orange bars, right axis) for each player position. The positions are sorted by their average stats score, making it easy to compare which positions produce the most statistical output versus which deliver the best dollar value. The chart lets you visually identify positions where value and stats are aligned versus those where one metric significantly outpaces the other.

The Pearson correlation matrix for Age, Overall Rating, Future Potential, Value Per M$, and Total Stats Score has been computed and visualized as an annotated heatmap. The heatmap displays all pairwise correlation coefficients, color-coded from red (negative) to blue (positive), making it easy to spot strong and weak relationships at a glance.

The analysis successfully identified players whose Value Per M$ is significantly higher than expected for their Overall Rating bucket. Using z-scores within 5-point rating bins, players more than 2 standard deviations above their bucket's mean were flagged. A bar chart and data tables were generated showing the top 15 most overvalued players, with each bar colored by Overall Rating and hover details showing Country, Position, and Z-Score.

The Overall Rating distribution across all 19,667 players has been analyzed with two visualizations: a histogram with KDE curve and a rating tier breakdown bar chart. The distribution is approximately normal (symmetric) with a mean of 63.23 and median of 63.00 — very close together, confirming the near-symmetric shape. The standard deviation is 7.81, and the skewness of 0.350 indicates only a slight positive lean. The majority of players fall in the 60-79 rating range, representing the bulk of the dataset.