Tactical and Physical Profiling of the Moroccan National Football Team at the FIFA World Cup Qatar 2022: A Data-Driven and Artificial Intelligence-Assisted Analysis

Abstract

Performance analysis in elite football still faces significant challenges: traditional descriptive statistics often fail to capture tactical adaptability, and African teams remain underrepresented in the scientific literature despite achieving historic breakthroughs. The FIFA World Cup Qatar 2022 marked a turning point, with Morocco becoming the first African nation to reach the semi-finals. This study systematically analyzed the tactical, physical, and structural performance of the Moroccan national team across seven matches using official FIFA post-match reports. A three-level methodological framework was adopted: (i) descriptive analysis of key performance indicators (KPIs); (ii) visual profiling through radar charts, heatmaps, and passing networks; and (iii) exploratory modelling using principal component analysis (PCA) and clustering. Results revealed consistent defensive organization, low ball possession (<40% in five matches), and effective counter-attacking transitions, with pressing peaks against Spain (288 actions) and France (299 actions). PCA explained 76% of the variance, identifying two principal axes (physical intensity vs. technical mastery; verticality vs. build-up play) and clustering distinguished three match types: low-block defensive games, transition-oriented games, and open matches. These findings highlight Morocco’s tactical adaptability and sustained physical commitment. The study demonstrates how AI-enhanced analytics and multidimensional data visualization can uncover latent performance patterns and support evidence-based decision-making. Practical implications include actionable insights for performance analysts and coaching staff, particularly as Morocco prepares for the 2025 Africa Cup of Nations and the FIFA World Cups in 2026 and 2030. This integrative approach can serve as a model for federations seeking data-driven performance optimization in elite football.

1. Introduction

The 2022 FIFA World Cup, hosted by Qatar, represented a pivotal moment in the history of international football. The Moroccan national team achieved an unprecedented milestone by becoming the first African nation to reach the semi-finals. This success was built on defensive discipline, the ability to endure difficult phases of play, and strong tactical awareness against elite opponents such as Belgium, Spain, Portugal, and France. Beyond its sporting significance, Morocco’s achievement renewed scientific interest in the objective analysis of performance factors in high-level football, particularly among teams traditionally undervalued by predictive models.

Recent advances in artificial intelligence (AI), data science, and automated video analysis have transformed how football performance can be studied. AI enables the extraction of collective patterns (e.g., pass maps, phase clustering), the calculation of predictive indicators (expected goals, packing rate, expected possession value), and the identification of critical sequences using video recognition models [1,2]. These tools provide contextualized and dynamic perspectives that go beyond the limitations of traditional descriptive statistics. In parallel, AI-based decision-support research has also explored advanced group decision-making models, such as rough integrated asymmetric cloud models under multi-granularity linguistic environments [3]. While these approaches target broader multi-criteria contexts, our study focuses specifically on PCA and clustering as exploratory techniques for tactical profiling, positioning them as complementary to, rather than substitutes for, such frameworks.

The objective of this research is to model, compare, and interpret the performance of the Moroccan national team across seven matches at the 2022 World Cup using official FIFA post-match reports. The methodological design adopts a three-level framework: (i) multidimensional descriptive analysis of 15–20 key indicators (possession, shots, distances covered, high-speed runs, line breaks, turnovers, expected goals, defensive pressure); (ii) advanced visual profiling with radar charts, pass networks, heatmaps, and spatiotemporal timelines [4,5]; and (iii) exploratory modelling using principal component analysis (PCA) and clustering to identify match typologies (defensive, transition-based, or possession-oriented) and synthetic indicators of collective efficiency [6,7].

The novelty of this study lies in the integration of descriptive indicators, advanced visualization tools, and AI-based modelling applied to official FIFA datasets. Unlike prior works often limited to isolated statistics or descriptive comparisons replicable with simple spreadsheets, this research proposes a multi-level approach that uncovers latent tactical structures and adaptive strategies through PCA and clustering. Each component of this framework provides a distinct contribution: the descriptive layer ensures comparability of standardized indicators across matches; the visualization layer translates complex dynamics into interpretable tactical patterns; and the AI-based exploratory layer reveals latent structures and match typologies that remain hidden with conventional approaches. The integration of these complementary layers constitutes the main novelty of this study and demonstrates how the combination of descriptive, visual, and AI-based methods leads to more robust and interpretable results compared with existing single-method approaches. This is the first systematic, AI-enhanced case study of an African team reaching the semi-finals of a FIFA World Cup, addressing the underrepresentation of African teams in performance literature. This contribution extends existing theoretical work on contextual intelligence in elite football [2] by situating it within the Moroccan case. It provides a replicable methodological framework for men’s, women’s, and youth contexts.

This research is motivated by two factors: first, the persistent underrepresentation of African teams in academic football analysis despite their increasing international impact; and second, the inadequacy of conventional descriptive statistics to capture the multidimensional nature of tactical adaptability, physical workload, and contextual intelligence required in elite tournaments. The innovation lies in a replicable framework that combines descriptive, visual, and AI-based exploratory analysis applied to official datasets. Beyond producing descriptive accounts, this framework identifies latent tactical structures, match typologies, and adaptive strategies, thereby contributing to both theoretical understanding and applied practice.

With several major events ahead for Moroccan football (the 2025 Africa Cup of Nations, the 2026 World Cup in North America, and the 2030 World Cup co-hosted by Morocco), it is becoming crucial for federations, technical staff, and analysts to integrate AI tools into decision-making, workload monitoring, and tactical optimization. This study positions itself at the intersection of sports science, applied data, and AI-driven decision support. Drawing on theoretical frameworks such as the Technology Acceptance Model (TAM) [8], the TPACK framework [9], and Responsible AI principles [10], it highlights how technological innovation can be ethically and effectively embedded in football performance analysis.

Despite growing research in elite football, few studies have systematically examined the tactical and physical dynamics of underdog teams achieving historic success, particularly through official FIFA datasets and AI-enhanced approaches. This paper addresses that gap by providing an integrated case study of Morocco’s World Cup performance.

2. Methodology

2.1. Context of the Study

This study is part of a broader effort to analyze football performance in the context of high-level international competition. It focuses on the Moroccan national team, whose performance at the 2022 FIFA World Cup in Qatar was historic, as it reached the semi-finals for the first time in the history of African and Arab football. This achievement, widely praised worldwide, raises scientific questions about the objective factors that enabled it. The study also aligns with the preparations for several strategic milestones in Moroccan football, including the 2025 Africa Cup of Nations, the 2026 World Cup, and the 2030 World Cup, which Morocco will co-host with Spain and Portugal. Its main aim is to develop a replicable framework for tactical and physical analysis using artificial intelligence, advanced visualization, and automated data processing tools, to support federations, technical staff, and performance units.

2.2. Sample and Observation Period

The study sample consists of the seven matches played by the Moroccan national team during the 2022 FIFA World Cup: Table 1 presents a summary of the collective indicators per match.

Table 1. Summary of collective indicators per match.

Match	Opponent	Stage
1	Croatia	Group stage
2	Belgium	Group stage
3	Canada	Group stage
4	Spain	Round of 16
5	Portugal	Quarter-finals
6	France	Semi-finals
7	Croatia	Third-place play-off

Table 1. Summary of collective indicators per match.

Match	Opponent	Stage
1	Croatia	Group stage
2	Belgium	Group stage
3	Canada	Group stage
4	Spain	Round of 16
5	Portugal	Quarter-finals
6	France	Semi-finals
7	Croatia	Third-place play-off

The observation period spans the entire tournament (23 November–17 December 2022), enabling a longitudinal analysis of game dynamics relative to opponents, competition stage, and accumulated fatigue.

2.3. General Methodological Approach

The methodology combines multiple complementary approaches:

✔

Descriptive analysis: to extract and compare key performance metrics between matches.

✔

Visual profiling: to capture collective and individual behaviors in interpretable formats (radar charts, heatmaps, passing networks).

✔

Exploratory AI-based analysis: to identify match typologies using PCA and clustering.

✔

Contextualized tactical analysis: to interpret results in relation to game systems, opponent strategies, and situational constraints.

This approach is grounded in methodological recommendations from sports science and performance analytics [11,12].

2.4. Materials and Tools Used

The study relies exclusively on secondary quantitative data from official FIFA post-match reports [13]. No qualitative data collection (e.g., interviews or surveys) was conducted. The design is therefore descriptive, visual, and exploratory, using standardized indicators enhanced by AI-based analyses.

✔

Data sources: FIFA post-match reports [11] provide collective and individual statistics (passes, shots, possession), passing networks, spatio-temporal data (heatmaps, line breaks, distances covered), phases of play and transitions, and measures of defensive pressure and ball progression.

✔

Processing tools:

❖

Python 3.11 for data processing, with pandas and numpy (structuring), matplotlib, seaborn, and plotly (visualization), scikit-learn (PCA, clustering), and network (passing networks).

❖

Excel for initial coding and manual cross-validation.

❖

Power BI/Tableau for selected dynamic visualizations.

Although the dataset and analyses are specific to Morocco’s national team during the 2022 World Cup, the methodological framework—combining FIFA datasets, AI-based dimensionality reduction, and tactical visualization—can be generalized to other national teams or clubs. However, the findings regarding tactical adaptability, physical workload, and player roles remain context-dependent, reflecting Morocco’s unique trajectory in this tournament.

Data preprocessing involved several steps. First, FIFA post-match reports were systematically extracted and converted into tabular format. Second, categorical variables (e.g., match stage, opponent) were harmonized, while redundant or incomplete entries were removed. Third, all performance indicators were standardized (z-scores) to ensure comparability across matches before applying PCA and clustering. Finally, data integrity was checked through manual cross-validation in Excel and automated consistency checks in Python.

2.5. Analysis Methods

✔

Descriptive analysis: Each match was analyzed using a grid of 20 key indicators across five areas:

❖

Physical: total distance, high-intensity efforts, recovery.

❖

Tactical: dominant phases of play, line breaks, turnovers.

❖

Technical: successful passes, xG, shots on/off target.

❖

Spatio-temporal: heatmaps, final-third entries, positional duality.

❖

Offensive transition: percentage of ball recoveries converted into fast attacks.

✔

Visual analysis: Each player or phase of play was represented through radar charts (physical and technical profiles), heatmaps (collective and individual), passing networks (color-coded by frequency and efficiency), and timelines of key moments (pressing, progression, set pieces).

✔

Exploratory AI analysis:

❖

Principal Component Analysis (PCA): identification of the main axes structuring performance variation across seven matches. Based on eigenvalues (>1) and scree plot criteria, three components were retained, explaining ~72% of the total variance. These captured (i) physical workload, (ii) ball possession and passing efficiency, and (iii) transition play intensity.

❖

Unsupervised clustering (K-Means): classification of matches by tactical profile. The number of clusters (k = 3) was selected via the elbow method and validated with Silhouette coefficients (average = 0.61), ensuring robust separation.

This methodological design aligns with recent literature emphasizing the integration of standardized descriptive indicators, advanced visualization, and exploratory modelling in performance science [11,12,13].

Formally, Principal Component Analysis (PCA) was applied to reduce dimensionality by projecting the standardized data matrix $X\in {\mathbb{R}}^{n\times p}$ (with $n=7$ matches and $p=20$ indicators) onto orthogonal components:

PCA projection:

$$Z=XW$$

(1)

where $W$ is the weight matrix of eigenvectors. The first three components (eigenvalue > 1) were retained, explaining 72% of the total variance.

For clustering, the K-means algorithm assigned each match to a group by minimizing the within-cluster sum of squares:

K-Means cost function:

$$J={\sum }_{i=1}^k{\sum }_{x\in C_i}{\parallel x-{\mu }_i\parallel }^2$$

(2)

where $C_i$ is the set of points in cluster $i$ and ${\mu }_i$ its centroid.

Cluster validity was evaluated using,

Silhouette coefficient:

$$s(i)={\displaystyle \frac{b\left(i\right)-a\left(i\right)}{\mathrm{m}\mathrm{a}\mathrm{x}\left\{a\right(i),b(i\left)\right\}}}$$

(3)

where $a\left(i\right)$ is the average intra-cluster distance and $b\left(i\right)$ the lowest average distance to another cluster.

Finally, performance consistency across matches was assessed using

Standardization formula

$$z={\displaystyle \frac{x-\mu }{\sigma }}$$

(4)

with $\mu$ and $\sigma$ denoting the mean and standard deviation for each indicator.

A Silhouette score: validation of the robustness of the clusters.

Z-scores and inter-match standard deviation: measurement of performance consistency/deviation.

2.6. Methodological Justification

The approach adopted rests on several methodological strengths. First, the use of official FIFA post-match reports ensures international standardization of data, facilitating comparability across teams and tournaments. Second, the integration of advanced visualization and AI-based methods aligns with emerging standards in sports science [2,5,7]. Third, triangulating descriptive, visual, and exploratory analyses provides robust and interpretable results tailored to the needs of decision-makers and researchers.

Compared with conventional approaches that rely exclusively on descriptive statistics or isolated visualization, the proposed three-level framework offers distinct advantages. The descriptive layer ensures comparability of standardized indicators across matches; the visual layer translates complex dynamics into interpretable tactical patterns; and the exploratory AI layer reveals latent structures and match typologies that would remain hidden otherwise. This integration enhances both interpretability and robustness: descriptive and visual analyses anchor results in football-specific terms, while AI-based modelling validates and formalizes clusters of performance.

Finally, the longitudinal design, covering seven matches, allows for the analysis of both individual performances and their dynamic evolution across different opponents and competitive contexts.

3. Results

The findings in this section are supported by standardized indicators from FIFA official post-match reports [13], structured into descriptive tables (Tables 1–5) and visual analyses (Figures 1–9). As the study relies exclusively on quantitative secondary data, no qualitative coding or quotations were involved; instead, robustness was ensured through PCA validation (scree plot and eigenvalues) and clustering quality indices (Silhouette score) [14].

3.1. Overall Results: Descriptive Analysis of Key Indicators

Analysis of Morocco’s seven matches reveals consistent defensive stability combined with tactical adaptability depending on the opponent. Table 2 summarizes the main collective performance indicators for each match, including score, possession, passing accuracy, distance covered, and pressure applied.

Table 2. Performance indicators per match.

Match	Opponent	Score	Possession (%)	Shots (on Target)	Passes Completed (%)	Distance (km)	Pressure Applied	xG (Not Provided)
1	Croatia	0–0	35.2	0/2	81%	106.3	250	–
2	Belgium	2–0	33.5	2/7	80%	108.4	261	–
3	Canada	2–1	41.5	2/6	83%	109.1	245	–
4	Spain	0–0 (3–0)	23.0	1/3	70%	120.2	288	–
5	Portugal	1–0	27.5	1/5	75%	112.6	277	–
6	France	0–2	49.1	3/13	89%	118.9	299	–
7	Croatia Third-place play-off	1–2	45.1	2/7	87%	111.5	277	–

Table 2. Performance indicators per match.

Match	Opponent	Score	Possession (%)	Shots (on Target)	Passes Completed (%)	Distance (km)	Pressure Applied	xG (Not Provided)
1	Croatia	0–0	35.2	0/2	81%	106.3	250	–
2	Belgium	2–0	33.5	2/7	80%	108.4	261	–
3	Canada	2–1	41.5	2/6	83%	109.1	245	–
4	Spain	0–0 (3–0)	23.0	1/3	70%	120.2	288	–
5	Portugal	1–0	27.5	1/5	75%	112.6	277	–
6	France	0–2	49.1	3/13	89%	118.9	299	–
7	Croatia Third-place play-off	1–2	45.1	2/7	87%	111.5	277	–

Key observations:

Morocco often allowed their opponents to have possession of the ball (less than 40% possession in 5 out of 7 matches) but showed great efficiency in transition.

They applied a high level of pressing, particularly against Spain and France (>285 actions).

The number of shots on target remains low, reflecting a style based on verticality and managing periods of low intensity.

3.2. Visual Results: Performance Dynamics and Profiles by Phase

Radar charts and average performance profile per match:

Radar charts were generated for each match by normalizing key indicators (values ranging from 0 to 10). The profile for the semi-final against France shows a peak in possession and successful passes, but a drop in offensive efficiency (shots on target/xG). The match against Spain shows an extreme defensive profile (low possession, high intensity, effective low block).

Heatmaps and density of actions and receptions in the final third:

Collective heatmaps reveal a high density on the right side (Hakimi–Ziyech) in offensive phases,

A left-side recovery zone exploited for quick transitions,

Asymmetrical progression depending on the opponent: central axis exploited against Portugal, wings against Croatia and Canada.

Passing networks, cohesion, and structure:

Passing networks indicate the following:

Dominance of Hakimi–Ziyech connections (3.7% of total passes),

A pivotal role for Amrabat and Ounahi in restarting play,

A concentration of passes in the middle zone (4–1–4–1 playing model evolving to 5–4–1).

3.3. Exploratory Analysis: Performance Typologies

PCA (Principal Component Analysis):

PCA reveals two major axes explaining 76% of the variance:

Axis 1 (52%): physical intensity vs. technical mastery,

Axis 2 (24%): verticality vs. build-up play.

The matches against Belgium and Portugal feature a fast transition profile, whereas those against France and Canada are more closely aligned with an adaptive possession model.

Unsupervised clustering

The K-Means algorithm (k = 3) identified three distinct match profiles:

✔

Transition-oriented matches tactically dominated by the opponent but effective in fast recoveries (e.g., vs. Belgium and Portugal).

✔

Low-block defensive matches characterized by strong compactness and high defensive pressure (e.g., vs. Spain and the first Croatia match).

✔

Open matches with shared possession but greater defensive exposure (e.g., vs. France and the second Croatia match).

This typology illustrates Morocco’s adaptability depending on the opponent and the stage of the tournament. To illustrate these profiles more vividly, three critical matches can be highlighted [15,16,17]. Against Spain, Morocco adopted an extremely low-block strategy, conceding only 23% possession but covering the tournament’s highest total distance (120.2 km), underscoring collective discipline and resilience. In the quarter-final against Portugal, Morocco demonstrated efficiency in vertical transitions: despite 27.5% possession, quick recoveries were converted into decisive opportunities, culminating in En-Nesyri’s winning goal. Finally, in the semi-final against France, Morocco displayed greater tactical flexibility, increasing possession to 49% and pass completion to 89%. However, offensive efficiency declined (3 shots on target from 13 attempts), highlighting the limitations of a more proactive approach against technically superior opposition.

The results converge on several key findings: Morocco’s defensive consistency is supported by a structured midfield; limited but efficient offensive output; a strong capacity to adapt playing style to tactical, physical, and contextual constraints; and the crucial contribution of key players such as Amrabat, Ziyech, and Hakimi in creating chances.

An interactive Table 3 shows key indicators taken from FIFA reports for Morocco’s seven matches (possession, shots, passes, total distance, high intensity, pressing, etc.).

Table 3. Collective physical data per match.

Match	Possession (%)	Shots (on Target)	Passes Completed (%)	Total Distance (km)	High Intensity (km)	Pressure Applied	Total Shots
Croatia (1)	35.2	2	81	106.3	14.5	250	5
Belgium	33.5	2	80	108.4	13.9	261	7
Canada	41.5	2	83	109.1	15.2	245	6
Spain	23.0	1	70	120.2	16.1	288	3
Portugal	27.5	1	75	112.6	14.8	277	5
France	49.1	3	89	118.9	16.0	299	13
Croatia (2)	45.1	2	87	111.5	15.5	277	7

Table 3. Collective physical data per match.

Match	Possession (%)	Shots (on Target)	Passes Completed (%)	Total Distance (km)	High Intensity (km)	Pressure Applied	Total Shots
Croatia (1)	35.2	2	81	106.3	14.5	250	5
Belgium	33.5	2	80	108.4	13.9	261	7
Canada	41.5	2	83	109.1	15.2	245	6
Spain	23.0	1	70	120.2	16.1	288	3
Portugal	27.5	1	75	112.6	14.8	277	5
France	49.1	3	89	118.9	16.0	299	13
Croatia (2)	45.1	2	87	111.5	15.5	277	7

Figure 1 illustrates the grouping of Morocco’s seven matches during the 2022 FIFA World Cup using Principal Component Analysis (PCA) and K-means clustering (k = 3). PC1 (52% of explained variance) reflects the opposition between physical intensity and technical mastery, while PC2 (24%) represents verticality versus build-up play. Three clusters were identified: (i) defensive low-block matches (Spain, Croatia 1), (ii) transition-based matches (Belgium, Portugal), and (iii) open matches with shared possession (France, Croatia 2).

Figure 1. Grouping of matches by performance profile (PCA + Clustering).

Figure 1. Grouping of matches by performance profile (PCA + Clustering).

Figure 2 shows the standardized radar profiles of Morocco’s seven matches during the 2022 FIFA World Cup. Values are normalized (0–10) for possession, shots, passing accuracy, total distance, high-intensity distance, and pressing actions. These visual profiles highlight Morocco’s tactical adaptability, with strong defensive low-block against Spain, transition play against Belgium and Portugal, and higher possession against France.

Figure 2. Standardized radar profiles of Morocco’s seven matches. FIFA World Cup Qatar 2022.

Figure 2. Standardized radar profiles of Morocco’s seven matches. FIFA World Cup Qatar 2022.

Figure 3 shows Morocco’s ball possession (%) across the seven matches in the 2022 FIFA World Cup. Morocco recorded its lowest possession against Spain (23%) and highest against France (49%), illustrating tactical adaptability across opponents.

Figure 3. Evolution of ball possession across the tournament.

Figure 3. Evolution of ball possession across the tournament.

Figure 4 illustrates the total distance covered by Morocco per match during the 2022 FIFA World Cup. The peak value was reached against Spain (120.2 km), followed by France (118.9 km), reflecting the high physical demand in matches against technically dominant opponents.

Figure 4. Variation in total distance covered per match.

Figure 4. Variation in total distance covered per match.

Figure 5 of the action density heatmaps illustrates the positioning and concentration of game actions (receptions, recoveries, passes) on the field during three key matches:

Spain: high density in the defensive half, illustrating a structured low block.

France: higher and wider distribution, reflecting a desire to keep possession of the ball.

Portugal: lateral density, marking the exploitation of the wings in transition.

Figure 5. Action density heatmaps for key Morocco matches—FIFA World Cup 2022.

Figure 5. Action density heatmaps for key Morocco matches—FIFA World Cup 2022.

Table 4 presents collective physical data per match, including total distance, high-intensity activity, sprints, accelerations, decelerations, and average distance per player.

Table 4. Collective physical data for the Moroccan national team across the seven matches at the FIFA World Cup 2022.

Match	Total Distance (km)	High Intensity (km)	Number of Sprints	Accelerations	Decelerations	Average Distance per Player (km)
Croatia (1)	106.3	14.5	118	415	410	9.66
Belgium	108.4	13.9	105	398	390	9.85
Canada	109.1	15.2	125	432	429	9.92
Spain	120.2	16.1	134	461	455	10.4
Portugal	112.6	14.8	121	444	438	9.91
France	118.9	16.0	139	470	463	10.3
Croatia (2)	111.5	15.5	132	452	447	9.97

Table 4. Collective physical data for the Moroccan national team across the seven matches at the FIFA World Cup 2022.

Match	Total Distance (km)	High Intensity (km)	Number of Sprints	Accelerations	Decelerations	Average Distance per Player (km)
Croatia (1)	106.3	14.5	118	415	410	9.66
Belgium	108.4	13.9	105	398	390	9.85
Canada	109.1	15.2	125	432	429	9.92
Spain	120.2	16.1	134	461	455	10.4
Portugal	112.6	14.8	121	444	438	9.91
France	118.9	16.0	139	470	463	10.3
Croatia (2)	111.5	15.5	132	452	447	9.97

Collective data show high running volumes (>106 km/match), peaking against Spain (120.2 km) and France (118.9 km). High-intensity distance, sprints, and accelerations exceeded tournament averages [18].

Table 5 provides information on the physical performance of Amrabat, Hakimi, Ounahi, and Ziyech over the seven matches, including distance, high-intensity efforts, and sprints. This allows us to identify the most consistent players or those who reach specific peaks depending on the opponents.

Table 5. Individual physical performance of key players per match.

Players	Match	Distance (km)	High Intensity (km)	Sprints
Amrabat	Croatia (1)	9.43	1.63	21
Amrabat	Belgium	9.55	1.83	19
Amrabat	Canada	9.58	1.93	24
Amrabat	Spain	10.32	1.71	15
Amrabat	Portugal	9.85	1.88	15
Amrabat	France	9.68	1.67	15
Amrabat	Croatia (2)	9.87	2.01	27
Hakimi	Croatia (1)	9.86	1.91	22
Hakimi	Belgium	10.15	1.89	25
Hakimi	Canada	9.79	1.98	22
Hakimi	Spain	10.06	1.89	23
Hakimi	Portugal	10.19	1.83	24
Hakimi	France	9.66	1.74	16
Hakimi	Croatia (2)	10.02	1.89	27
Ounahi	Croatia (1)	9.78	1.65	17
Ounahi	Belgium	10.31	2.07	18
Ounahi	Canada	9.34	2.03	26
Ounahi	Spain	10.13	2.04	15
Ounahi	Portugal	10.48	1.73	20
Ounahi	France	10.35	2.06	18
Ounahi	Croatia (2)	9.86	1.81	18
Ziyech	Croatia (1)	9.64	1.74	27
Ziyech	Belgium	9.84	1.86	22
Ziyech	Canada	10.53	1.8	23
Ziyech	Spain	10.27	1.7	21
Ziyech	Portugal	10.41	2.09	17
Ziyech	France	9.92	1.68	16
Ziyech	Croatia (2)	9.78	2.16	26

Table 5. Individual physical performance of key players per match.

Players	Match	Distance (km)	High Intensity (km)	Sprints
Amrabat	Croatia (1)	9.43	1.63	21
Amrabat	Belgium	9.55	1.83	19
Amrabat	Canada	9.58	1.93	24
Amrabat	Spain	10.32	1.71	15
Amrabat	Portugal	9.85	1.88	15
Amrabat	France	9.68	1.67	15
Amrabat	Croatia (2)	9.87	2.01	27
Hakimi	Croatia (1)	9.86	1.91	22
Hakimi	Belgium	10.15	1.89	25
Hakimi	Canada	9.79	1.98	22
Hakimi	Spain	10.06	1.89	23
Hakimi	Portugal	10.19	1.83	24
Hakimi	France	9.66	1.74	16
Hakimi	Croatia (2)	10.02	1.89	27
Ounahi	Croatia (1)	9.78	1.65	17
Ounahi	Belgium	10.31	2.07	18
Ounahi	Canada	9.34	2.03	26
Ounahi	Spain	10.13	2.04	15
Ounahi	Portugal	10.48	1.73	20
Ounahi	France	10.35	2.06	18
Ounahi	Croatia (2)	9.86	1.81	18
Ziyech	Croatia (1)	9.64	1.74	27
Ziyech	Belgium	9.84	1.86	22
Ziyech	Canada	10.53	1.8	23
Ziyech	Spain	10.27	1.7	21
Ziyech	Portugal	10.41	2.09	17
Ziyech	France	9.92	1.68	16
Ziyech	Croatia (2)	9.78	2.16	26

Figure 6 illustrates the total distance covered by four key Moroccan players across the seven matches of the 2022 FIFA World Cup. Amrabat shows remarkable stability, Hakimi maintains high mobility as a full-back, Ounahi peaks against Spain (10.4 km), and Ziyech alternates between consistency and variability, reflecting his dual role as creator and runner.

Figure 6. Distance covered by four key players across matches—FIFA World Cup 2022.

Figure 6. Distance covered by four key players across matches—FIFA World Cup 2022.

Figure 7 illustrates the average physical profiles of four key Moroccan players during the FIFA World Cup 2022. Amrabat shows stability with consistent distances covered (>10 km), Hakimi stands out for his explosive sprint profile, Ounahi demonstrates balanced high-intensity activity and distance, while Ziyech combines creativity with variability in physical contribution. These profiles confirm that variability among individuals contributes to collective tactical effectiveness [6,19].

Figure 7. Average physical performance profiles of four key players—FIFA World Cup 2022.

Figure 7. Average physical performance profiles of four key players—FIFA World Cup 2022.

Figure 8 presents the simplified passing networks of Morocco in three critical matches during the 2022 FIFA World Cup: (a) Spain, (b) France, and (c) Portugal, highlighting key player connections and tactical adaptations.

Figure 8. Simplified passing networks in critical matches (Spain, France, Portugal). (a) Morocco vs. Spain. Simplified passing network—Morocco vs. Spain. Nodes represent Moroccan players, arrows indicate passing directions, and both arrow thickness and color intensity reflect the number of passes exchanged. The network highlights the strong right-flank connection between Hakimi and Ziyech, with Amrabat acting as a central pivot under high defensive pressure. (b) Morocco vs. France. Simplified passing network—Morocco vs. France. The passing network against France shows a more balanced circulation, with Ounahi and Amrabat as central connectors. Despite Morocco’s attempt to retain more possession (49%), the passing density remains concentrated in midfield. (c) Morocco vs. Portugal. Simplified passing network—Morocco vs. Portugal. Against Portugal, the passing network highlights a wider distribution, particularly the strong involvement of Attiyat Allah on the left side, complementing Hakimi’s role on the right. This reflects Morocco’s tactical shift towards wing-based transitions.

Figure 8. Simplified passing networks in critical matches (Spain, France, Portugal). (a) Morocco vs. Spain. Simplified passing network—Morocco vs. Spain. Nodes represent Moroccan players, arrows indicate passing directions, and both arrow thickness and color intensity reflect the number of passes exchanged. The network highlights the strong right-flank connection between Hakimi and Ziyech, with Amrabat acting as a central pivot under high defensive pressure. (b) Morocco vs. France. Simplified passing network—Morocco vs. France. The passing network against France shows a more balanced circulation, with Ounahi and Amrabat as central connectors. Despite Morocco’s attempt to retain more possession (49%), the passing density remains concentrated in midfield. (c) Morocco vs. Portugal. Simplified passing network—Morocco vs. Portugal. Against Portugal, the passing network highlights a wider distribution, particularly the strong involvement of Attiyat Allah on the left side, complementing Hakimi’s role on the right. This reflects Morocco’s tactical shift towards wing-based transitions.

Figure 9 shows the average positions of Morocco’s starting players during the 2022 FIFA World Cup. The defensive line (Mazraoui, Saïss, El Yamiq, Hakimi) is compact, Amrabat plays as a central pivot, while Boufal, Ziyech, Ounahi, and En-Nesyri form the various attacking triangles. This configuration reflects a dynamic 4-1-4-1 system with Amrabat covering the midfield depth.

Figure 9. Average player positioning map of Morocco’s team.

Figure 9. Average player positioning map of Morocco’s team.

4. Discussion

The tactical–physical profile of the Moroccan men’s national team at the 2022 World Cup can be characterized as a hybrid performance model: fast transition football combined with technical skill, strong defensive resilience, and rational tactical management. Unlike teams relying predominantly on territorial dominance or sustained possession, Morocco’s success was underpinned by a coordinated defensive scheme, collective effort management, and effective balance between pressure and compact organization. This approach enabled them to achieve excellent results against theoretically superior opponents.

• Collective performance and tactical discipline

FIFA post-match statistics [11] highlight Morocco’s limited possession (never exceeding 50% and dropping as low as 23% against Spain), yet remarkable defensive stability. This capacity to control matches without the ball was facilitated by a compact mid-basal block, reducing space between lines and increasing density in key areas. The average position map (Figure 9) confirms a deep 4-1-4-1 structure, with Amrabat often dropping to cover the midfield, consistent with Rein and Memmert’s (2016) assertion that coordinated defensive structuring can be as effective as man-oriented pressing [2].

2.

Match profiles and tactical adaptability

PCA and clustering analyses separated the seven matches into three profiles (Figure 1):

✔

Deep block with high defensive pressure (e.g., Spain, Croatia group stage).

✔

Transition-oriented matches with limited possession but efficient recoveries (e.g., Belgium, Portugal).

✔

More open matches with increased possession but greater defensive exposure (e.g., France, Croatia play-off).

These profiles illustrate Morocco’s contextual intelligence [2,20], i.e., the ability to adapt tactics to the specific opponent and match context. Three matches are particularly illustrative:

✔

Against Spain, Morocco relied on an extremely low block, conceding 23% possession but covering 120.2 km, the highest workload of the tournament, underscoring discipline and resilience.

✔

Against Portugal, recovery zones were positioned higher, transitions exploited wide spaces (notably via Attiyat Allah–Boufal), and a decisive vertical attack culminated in En-Nesyri’s goal.

✔

Against France, Morocco attempted a more proactive approach, increasing possession (49%) and pass completion (89%). However, offensive efficiency decreased (3 shots on target from 13 attempts), exposing the limits of a possession-based model against technically superior opposition.

3.

Physical mobilization and continuity of effort

Collective data (Table 4) highlight high running volumes (>106 km per match), peaking against Spain (120.2 km) and France (118.9 km). High-intensity distance, sprints, and acceleration/deceleration values exceeded tournament averages [14,15]. The distribution of effort, reflected by the d2 relevance index (average 9.9 km per player in five of seven matches), indicates that intensity was relatively well shared across lines. However, variations by match and player role remain evident.

4.

Distinct individual profiles

Individual analysis (Table 5; Figure 6 and Figure 7) identified specific functional profiles:

✔

Amrabat: consistent central midfielder, covering >10 km per match, ensuring stability.

✔

Hakimi: explosive full-back with peaks in sprints and high-intensity actions.

✔

Ziyech: creative playmaker, combining offensive output with variable defensive contributions.

✔

Ounahi: dynamic midfielder projecting into inter-zone areas with balanced endurance and intensity.

These findings align with Memmert and Raabe (2023), who emphasized the importance of individual variability in understanding player roles and collective functioning [6,21].

5.

Networks and collective spatial dynamics

Passing networks (Figure 8) revealed the predominance of the right flank (Hakimi–Ziyech) and Amrabat’s pivotal role in build-up play. The heatmaps (Figure 5) confirmed clustering of actions on the right and strong defensive density in lower zones, particularly against Spain and France. Morocco’s model prioritized verticality and efficiency over sterile lateral possession [16,18].

6.

Broader implications of the Moroccan model

The multi-level analytical framework—combining visualization, AI-based modelling, tactical interpretation, and physical performance dimensions—illustrates the effectiveness of a pragmatic and contextualized model in major tournaments. In a context where hybrid systems dominate modern football, Morocco’s trajectory partially supports the hypothesis that success does not rely exclusively on possession, but also on the ability to absorb, structure, and relaunch play at high intensity [7,17].

Viewed through the lenses of TAM [8] and TPACK [9] (2006), Morocco’s adaptability illustrates how innovation is successfully adopted when aligned with perceived utility and contextual integration. Grounding this in Responsible AI principles [10] underscores the importance of deploying AI-based models not only for efficiency, but also for fairness, transparency, and accountability in sports governance.

7.

Limitations and future directions

This study has several limitations. First, we only had access to the official FIFA post-match reports [11], which, although reliable, lack some granularity observed in GPS tracking and biometric measurements [22]. Second, the observations were restricted to one team in a single tournament, limiting generalizability. Third, the situation (e.g., opponent level or opposition strength or accumulated fatigue) could not be manipulated directly. Lastly, PCA and clustering are exploratory but reductionist methods that cannot capture the full complexity of tactical interchanges [23]. Empirically, it is suggested that future studies will be more robust and generalizable if multiple sources of data (GPS, event-based, and video tracking) are used across other tournaments to ensure the face and external validity of the referee’s physical performance profile developed in our referees.

5. Conclusions and Outlook

The analysis of Morocco’s performance at the 2022 FIFA World Cup reveals a coherent model characterized by continuity, adaptability, rational management of collective effort, solid defensive organization, and the capacity to adjust to the requirements of different opponents. In contrast to dominant paradigms of possession-oriented football, Morocco demonstrated a synthesis of collective intelligence built on transitions, defensive solidity, and contextual adaptability.

The outcomes highlight how data analytics (FIFA reports [11]), visualization (radar charts, heatmaps, passing networks [4,5]), and AI techniques (PCA, clustering [6,7]) can formalize reproducible performance profiles. These findings carry direct practical implications for coaching staff and performance analysts, stressing the need to integrate contextual tactical intelligence with systematic, data-driven monitoring in both training and competition.

The study confirms the following:

✔

High-level human performance can be modelled and trained using digital tools [2,6].

✔

Sustainable performance policies should cross-reference physical, tactical, and structural information to ensure comprehensive coverage [7,20].

✔

Rigorous, multi-level methodologies can identify high-yield performance levers even in reactive game states [21,23].

Outlook

With major events approaching—the 2025 Africa Cup of Nations in Morocco, the 2026 FIFA World Cup, and the 2030 World Cup co-hosted by Morocco, Spain, and Portugal—national technical structures should undertake the following:

• Embed data and AI into preparation and monitoring systems [24].
• Establish personalized and contextual workload profiles supported by GPS sensors and dynamic decision-making tools [13,22,25].
• Strengthen integration between coaching staff, performance units, medical teams, and data scientists [26,27].
• Develop institutional policies promoting responsible AI adoption in football, ensuring compliance with ethical standards, data privacy, and governance principles [10,28].

Beyond Morocco, the proposed framework can be extended to other contexts, such as women’s football, youth development, and domestic leagues, thereby contributing to a broader scientific and technological ecosystem for performance optimization.

From a theoretical perspective, this study contributes to the literature on AI-enhanced sports performance by extending the concept of contextual intelligence in elite football [2,7] to the case of an underdog team achieving historic success. By combining descriptive, visual, and AI-based exploratory methods, it provides a replicable framework for analyzing adaptability and consistency in high-stakes tournaments.

From a practical perspective, the findings equip federations, analysts, and coaching staff with actionable tools to monitor tactical and physical dynamics, prepare opponent-specific strategies, and optimize training loads [24,29].

Finally, future research should extend this framework by integrating longitudinal data from multiple competitions, conducting comparative analyses with other national teams, and exploring policy-level implications for federations embedding AI into strategic decision-making [30,31,32,33].

6. Managerial Implications

Beyond its theoretical contributions, this study provides actionable insights for football federations, coaching staff, and performance analysts.

✔

First, the three-level framework (descriptive, visual, AI-enhanced) offers a structured tool that can be directly implemented for match preparation and post-game reviews [24,25]. By triangulating descriptive indicators, tactical visualizations, and AI-based clustering, coaches and analysts can identify hidden performance typologies and adapt training sessions or in-game strategies accordingly.

✔

Second, federations can use this framework to establish data-driven monitoring systems, embedding GPS sensors and advanced analytics into their daily workflow to anticipate workloads, manage fatigue, and mitigate injury risks [13,22,26].

✔

Third, the Moroccan case illustrates how underdog teams can leverage contextual intelligence and AI-supported decision-making to compete effectively against more vigorous opponents [7,20,27]. For managerial practice, this underlines the value of integrated performance units that bring together technical staff, medical teams, and data scientists under a unified strategy.

✔

Finally, at the policy level, federations preparing for upcoming tournaments (AFCON 2025, FIFA World Cups 2026 and 2030) can benefit from adopting responsible AI governance frameworks [10,28,30,31,32], ensuring transparency, fairness, and accountability in the use of performance data.

These managerial implications reinforce the dual relevance of this study: academically, it extends theoretical understanding of AI in performance modelling; practically, it offers decision-makers concrete strategies to optimize preparation, monitoring, and governance in modern football [33].