Introduction
Football performance is underpinned by neuromus- cular qualities such as force and power, particularly during sprint acceleration and explosive movements [1, 2]. In recent years, force-velocity (F-V) profiling has emerged as a practical approach to characterise these mechanical capacities and to inform individualised training strategies in applied sport settings [3–5].
Field-based methodologies have facilitated the as- sessment of key F-V variables, including theoretical maximal force (F0), theoretical maximal velocity (V0), and maximal power (Pmax), using vertical jump and sprint tasks [6, 7]. While vertical F-V profiling derived from jumping tests is widely employed to evaluate lower-limb explosive performance, its relationship with sprint-specific mechanical outputs remains unclear. Sprint acceleration, in particular, appears to rely more heavily on the ability to generate and orient force in the horizontal direction [8, 9]. Consequently, horizontal F-V profiling has been proposed as a complementary approach to more effectively capture sprint-related me- chanical characteristics [10].
Recent studies and systematic reviews suggest that vertical and horizontal F-V profiles may reflect task- specific mechanical characteristics rather than fully interchangeable neuromuscular qualities, although these findings should be interpreted with caution [3, 11, 12]. Indeed, the relationships between vertical jump performance and sprint ability are generally reported to range from weak to moderate, indicating limited overlap between these movement capacities [12–14]. However, it remains unclear whether vertical and hor- izontal F-V profiles derived from field-based methods reflect overlapping or distinct mechanical qualities when assessed within the same population, particu- larly among semi-professional football players. More- over, limited attention has been devoted to the potential mismatch in profile orientation between vertical and horizontal assessments, as well as to the extent of inter- individual variability within relatively homogeneous groups of athletes. These aspects are particularly rel- evant in applied settings, where practitioners often rely on a single assessment modality to guide training decisions.
Beyond average relationships, growing attention has been paid to inter-individual variability in F-V profiles. Previous research has shown that athletes with similar performance levels may exhibit markedly different me- chanical strategies, such as force-oriented or velocity- oriented profiles [5, 15]. This variability may have im- portant implications for individualised training pre- scription, as athletes with similar performance levels may require different mechanical interventions. How- ever, this aspect remains insufficiently explored in studies simultaneously examining vertical and hori- zontal F-V characteristics [16]. In addition, potential mismatches in the profile orientation between verti- cal and horizontal tasks have received limited atten- tion, despite their relevance for understanding move- ment-specific mechanical demands. Notably, sprint mechanical effectiveness, defined as the ability to ori- ent ground reaction force in the direction of motion, has been identified as a key determinant of sprint per- formance [6, 9, 17]. Variables, such as RFmax and DRF, may therefore provide additional insight beyond ver- tical assessments alone.
Furthermore, most existing studies have been con- ducted in European or South American populations, with limited data available on African football players. This lack of representation may restrict the general- isability of current findings, highlighting the need for population-specific investigations.
Therefore, the primary aim of the present study was to examine the associations between vertical and horizontal force-velocity profiles among Moroccan U23 football players using validated field-based tools (MyJump2 and MySprint). A secondary aim was to ex- plore the extent of inter-individual variability and po- tential differences in profile orientation between move- ment tasks. It was hypothesised that vertical and horizontal F-V variables would show limited associa- tions, reflecting task-specific mechanical characteris- tics rather than interchangeable performance qualities. By addressing these aspects, this study sought to pro- vide a more nuanced understanding of neuromuscular profiling and to support individualised training ap- proaches in football.
Material and methods
Participants and study design
This study employed a cross-sectional observational design to examine vertical and horizontal F-V profiles among Moroccan U23 football players and to explore the associations between these mechanical charac- teristics. All testing procedures were conducted during the competitive season within a short data-collection period (≤ 7 days) to minimise potential variation related to fatigue, training load, or seasonal changes. Testing was performed under standardised outdoor conditions and, whenever possible, at the same time of day for all participants to reduce potential circadian variability.
Eighteen male Moroccan football players (n = 18) voluntarily participated in this study. In addition to these conditions, inclusion criteria required partici- pants to be active U23 male football players competing at a semi-professional level, regularly engaged in struc- tured team training ( 4 sessions per week) and com- petitive match play, with a minimum of 2 years of con- tinuous training experience. Players were excluded if they had sustained a lower-limb injury within the 3 months preceding testing, were undergoing reha- bilitation, or were unable to perform maximal sprint- ing or jumping efforts during testing. Their mean ± standard deviation (SD) age, height, and body mass were 20.39 ± 1.82 years, 176.94 ± 9.69 cm, and 65.39 ± 8.53 kg, respectively.
Given the correlational nature of the study and the difficulty of recruiting a homogeneous group of com- petitive football players, the present study should be considered exploratory in nature. Nevertheless, the sample size was consistent with previously applied studies investigating F-V profiling in team-sport ath- letes [18–20], and findings should therefore be inter- preted with appropriate caution.
Prior to participation, all players were informed about the study procedures, risks, and potential bene- fits, and written informed consent was obtained from each participant. All procedures were conducted in accordance with recognised ethical standards for hu- man research and ensured participant confidentiality, voluntary participation, and data protection.
Experimental procedures
All assessments were performed on the same out- door football facility under similar environmental con- ditions. Testing took place on a synthetic football pitch at an ambient temperature of approximately 21°C. Par- ticipants were instructed to avoid strenuous physical activity for at least 24 hours before each testing session and to maintain their usual hydration and nutritional habits.
To minimise fatigue and avoid potential interfer- ence between neuromuscular assessments, vertical and horizontal F-V profiling were performed on sep- arate days, with a 48-hour recovery period between sessions, in line with previously used sprint and jump mechanical assessment procedures [21, 22]. The order of the testing sessions (vertical and horizontal F-V pro- filing) was randomised across participants to control for potential order and learning effects. Each testing session consisted of a standardised warm-up followed by the specific testing protocol (vertical or horizontal assessment). Participants were familiarised with the testing procedures and exercises before data collection.
The standardised warm-up lasted approximately 12–15 min and was designed to progressively prepare the neuromuscular system for maximal sprint and jump efforts. It included 5 min of low-intensity run- ning, dynamic mobility exercises for the hip, knee, and ankle joints, dynamic stretching drills, and 3–5 pro- gressive accelerations over 20–30 m. The warm-up in- tensity was progressively increased until participants were considered ready to perform maximal efforts. This type of structured warm-up is commonly used in sprint and jump testing and has been shown to improve neuromuscular readiness and acute performance [23, 24].
Vertical force-velocity profiling
The vertical F-V profile was assessed using counter- movement jump (CMJ) tests analysed with the MyJump2 mobile application. MyJump2 has previously demon- strated good validity and reliability for measuring jump height under field conditions when compared with force platforms [25–27]. Participants performed three maxi- mal CMJs in each of four loading conditions corre- sponding to 0%, 20%, 30%, and 40% of their individual body mass. During the unloaded condition (0% body mass), participants kept their hands on their hips to minimise the influence of arm swing and ensure stand- ardised execution. During the loaded conditions (20%, 30%, and 40% body mass), a standard Olympic barbell was positioned across the shoulders and held securely throughout each jump. A recovery period of 2–3 min was provided between loading conditions to minimise fatigue. For each loading condition, the best jump height was retained for subsequent analysis to reflect maxi- mal neuromuscular capacity.
Vertical F-V variables were calculated using the field method proposed by Samozino et al. [28, 29], which estimates the mechanical outputs of jumping perfor- mance based on anthropometric data and jump out- comes. The following variables were derived: theoreti- cal maximal force (F0), theoretical maximal velocity (V0), maximal power output (Pmax), and the slope of the F-V relationship.
Profile orientation was interpreted relative to the theoretical optimal F-V profile for jumping performance according to the practical framework proposed by Sa mozino et al. [30] and later applied by Jiménez- Reyes et al. [15, 31]. Based on this comparison, players were classified as force-oriented, velocity-oriented, or near-optimal according to whether their individual profile deviated toward the force or velocity end of the optimal continuum.
Horizontal force-velocity profiling
The horizontal F-V profile was determined from 30-m sprint performance using the validated field- based mechanical model proposed by Morin et al. [6] and Samozino et al. [10]. Sprint times were recorded at split distances of 5, 10, 15, 20, 25, and 30 m and ana- lysed using the MySprint mobile application, which has shown acceptable validity and reliability when com- pared with radar-based systems [32]. Each participant performed two maximal sprints from a standing start position, with a recovery period of at least 3 min be- tween attempts. The best sprint trial was retained for analysis, as it is considered to best reflect maximal neuromuscular and mechanical capacity and is com- monly adopted in force-velocity profiling studies due to its acceptable reliability in field-based sprint as- sessments [11, 33].
Horizontal F-V variables included theoretical maxi- mal horizontal force (F0), theoretical maximal velocity (V0), maximal power output (Pmax), the peak ratio of horizontal force (RFmax), and the rate of decrease in the force ratio (DRF). These variables provide insight into the athlete’s capacity to produce and effectively orient force during sprint acceleration.
Profile orientation was interpreted based on the re- lationship between the individual sprint F-V profile and the theoretical optimal profile for acceleration per- formance, following the framework proposed by Samo- zino et al. [5] and its subsequent applications in sprint profiling [3, 33]. Athletes were classified according to the direction and magnitude of the difference between the individual and optimal F-V slope ( FV). Deviations exceeding the typical error of measurement were con- sidered meaningful, whereas values within this range were interpreted as near-optimal. Accordingly, players were categorised as force-oriented, velocity-oriented, or near-optimal, depending on their deviation from the theoretical optimal profile.
Statistical analysis
All statistical analyses were performed using IBM SPSS Statistics (version 28.0, IBM Corp., Armonk, NY, USA). Descriptive statistics are presented as mean ± SD, along with 95% confidence intervals (CI 95%) to indicate the precision of the estimated means. The co- efficient of variation (CV) was also calculated for each variable to express relative variability as a percentage of the mean, with lower CV values indicating more con- sistent measurements across participants.
The normality of data distribution was assessed using the Shapiro–Wilk test. Based on these results, parametric or non-parametric tests were applied ac- cordingly. Pearson’s correlation coefficients (r) were used to examine relationships between normally dis- tributed vertical and horizontal force-velocity varia- bles. Spearman’s rank correlation coefficient ( ) was used for the variables that did not meet the assump- tion of normality (i.e., RFmax). For all correlation analy- ses, 95% confidence intervals were calculated to pro- vide an estimation of the precision of the coefficients. The magnitude of correlations was interpreted accord- ing to the following thresholds: trivial (< 0.1), small (0.1–0.3), moderate (0.3–0.5), large (0.5–0.7), and very large (> 0.7) [34].
The McNemar-Bowker test was performed to ex- amine the association between categorical force-ve- locity profiles (force-oriented, velocity-oriented, and optimal) derived from vertical and horizontal assess- ments. This test enables the evaluation of symmetry in paired categorical data and is appropriate for detect- ing differences in profile distribution between condi- tions. Statistical significance was set at p < 0.05. Effect sizes were interpreted based on the magnitude of cor- relation coefficients.
Results
The results are presented in a structured manner, including descriptive statistics of the force-velocity pro- files, distribution of player profiles, association between vertical and horizontal classifications, and correlations between mechanical variables.
Table 1 presents the descriptive statistics of the ver- tical and horizontal F-V variables. For the vertical pro- file, mean values were 35.53 ± 5.50 N • kg–1 for F0, 3.17 ± 0.73 m • s–1 for V0, 28.11 ± 5.42 W • kg–1 for Pmax, and –12.10 ± 3.73 N • s • m–1 for FV_slope. For the horizontal profile, mean values were 9.06 ± 1.03 N • kg–1 for F0, 8.56 ± 0.34 m • s–1 for V0, 19.36 ± 2.39 W • kg–1 for Pmax, 0.56 ± 0.03 for RFmax, and –0.096 ± 0.010 for DRF.
Table 1
Descriptive statistics of vertical and horizontal force-velocity variables
[i] F0 – theoretical maximal force, V0 – theoretical maximal velocity, Pmax – maximal power output, FV_slope – slope of the force-velocity relationship, RFmax – peak ratio of horizontal force, DRF – rate of decrease in the ratio of force, vert – vertical profile, horiz – horizontal profile, CMJ – countermovement jump
The 95% confidence intervals indicated relatively narrow ranges for most horizontal variables, particu- larly V0_horiz, RFmax, and DRF. The CV ranged from 4.0% to 30.8%, with the lowest variability observed for V0_horiz and RFmax, and the highest variability ob- served for FV_slope.
Table 2 presents the distribution of force-velocity profile orientations obtained from the vertical and hor- izontal assessments. In the vertical profile, the majority of players were classified as velocity-oriented (n = 12, 66.7%), whereas smaller proportions were classified as force-oriented (n = 4, 22.2%) or optimal (n = 2, 11.1%).
Table 2
Distribution of vertical and horizontal force-velocity profile orientations
| Profile | Vertical (n, %) | Horizontal (n, %) |
|---|---|---|
| Velocity-oriented | 12 (66.7%) | 3 (16.7%) |
| Optimal | 2 (11.1%) | 8 (44.4%) |
| Force-oriented | 4 (22.2%) | 7 (38.9%) |
In contrast, the horizontal profile showed a different distribution, with a lower proportion of velocity-ori- ented players (n = 3, 16.7%) and higher proportions of optimal (n = 8, 44.4%) and force-oriented profiles (n = 7, 38.9%). Overall, the distribution of profile ori- entations differed between the vertical and horizon- tal assessments.
Table 3 presents the results of the McNemar-Bowk- er test employed to examine whether profile orientation differed between the vertical and horizontal force- velocity assessments. The analysis revealed a statis- tically significant difference in profile distribution be- tween the two conditions (χ 2 = 8.667, df = 3, p = 0.034).
Table 3
Results of the McNemar-Bowker test comparing vertical and horizontal profile orientation
| Test | χ2 | df | p-value |
|---|---|---|---|
| Nemar–Bowker | 8.667 | 3 | 0.034 |
Table 4 presents the results of the Shapiro–Wilk normality test for the mechanical variables. Most vari- ables were normally distributed (p > 0.05). RFmax was the only variable that did not meet the assumption of normality (W = 0.868, p = 0.017), and was therefore analysed using Spearman’s rank correlation coefficient.
Table 4
Results of the Shapiro–Wilk normality test for vertical and horizontal force-velocity variables
[i] F0 – theoretical maximal force, V0 – theoretical maximal velocity, Pmax – maximal power output, FV_slope – slope of the forcevelocity relationship, DRF – rate of decrease in the ratio of force, RFmax – peak ratio of horizontal to resultant force, vert – vertical profile, horiz – horizontal profile
Table 5 presents the correlations between vertical and horizontal F-V variables. Overall, the relationships between the variables were weak to moderate in mag- nitude and did not reach statistical significance (p > 0.05). A moderate negative correlation was observed between F0_vert and F0_horiz (r = –0.412, p = 0.089), although the associated confidence interval was wide and included zero, indicating uncertainty in the esti- mate. The remaining correlations were trivial to small in magnitude, including V0_vert × V0_horiz (r = –0.087, p = 0.732), Pmax_vert × Pmax_horiz (r = –0.016, p = 0.950), and FV_slope × DRF (r = 0.283, p = 0.256). Overall, the relatively wide confidence intervals suggest a degree of imprecision in the estimated relationships.
Table 5
Correlations between vertical and horizontal force-velocity variables (r, 95% CI, and p-values)
Table 6 presents the Spearman correlation involv- ing RFmax. A small positive, non-significant correla- tion was observed between FV_slope and RFmax ( = 0.305, p = 0.219). The associated 95% confidence in- terval was relatively wide and included zero, indicat- ing uncertainty in the estimated relationship.
Discussion
The main finding of the present study is that ver- tical and horizontal F-V profiles showed limited asso- ciations among Moroccan U23 football players, suggest- ing that these assessments may reflect task-specific mechanical characteristics rather than interchange- able neuromuscular qualities, although this observa- tion should be approached with caution. This interpre- tation is supported not only by the weak and mostly non-significant correlations observed, but also by the differences in profile distribution and the substantial inter-individual variability identified across players. Taken together, these findings suggest that mechanical profiling in football should be regarded as movement- specific rather than as a single global representation of “explosive performance”.
One of the most relevant observations of the present study concerns the apparently greater relevance of hor- izontal F-V variables for sprint acceleration perfor- mance. In particular, theoretical maximal horizontal force (F0) and mechanical effectiveness variables, such as RFmax, appeared to be more informative for under- standing acceleration-related capacities than vertical F-V outputs. This is consistent with the sprint biome- chanics literature, which has repeatedly shown that acceleration performance depends heavily on the ath- lete’s ability to produce and orient force horizontally during the initial steps of sprinting [6, 10, 17]. From a mechanical standpoint, early acceleration requires overcoming inertia through effective forward-directed force application, and variables such as RFmax and DRF have been proposed as meaningful indicators of this capability [9, 33]. Therefore, the present results support the notion that horizontal profiling may pro- vide movement-specific information of particular rele- vance to football sprint performance.
By contrast, the weaker relationships observed be- tween vertical F-V variables and horizontal sprint-re- lated variables suggest that vertical assessments such as countermovement jump profiling should be inter- preted with caution when used to infer sprint-specific mechanical capacities. Although vertical F-V profiling remains a valuable approach for assessing lower-limb explosive performance, particularly in relation to force and power expression during jumping tasks, its as- sociation with sprint acceleration appears to be limited and inconsistent. This finding should nonetheless be interpreted with caution given the relatively small sam- ple size and the variability in testing conditions reported across populations in the literature. However, it aligns with previous research showing only low-to-moderate relationships between vertical jump performance and sprint-related outcomes in athletes [3, 35–37]. It also supports the broader principle of biomechanical speci- ficity, whereby force production is highly dependent on task constraints, direction of force application, and movement mechanics [38]. Consequently, improve- ments in vertical mechanical output may not necessarily be accompanied by equivalent changes in sprint ac- celeration performance.
A particularly noteworthy observation was the ten- dency toward an inverse association between vertical and horizontal F0, even though this relationship did not reach statistical significance. Although this tendency should be interpreted with caution, it may suggest that the capacity to express force in vertically-oriented tasks does not necessarily translate into similar force expres- sion in horizontally-oriented sprint actions. This may reflect differences in mechanical orientation, inter- muscular coordination, and task-specific force appli- cation demands between jumping and sprinting tasks. In line with this interpretation, previous research has shown that sprint acceleration performance is not de- termined solely by the magnitude of force production, but rather by the athlete’s ability to effectively orient this force in the direction of motion [6, 9, 39].
Another meaningful observation was the near ab- sence of association between vertical and horizontal maximal power output (Pmax). Although maximal power is often considered a global marker of explosive capability, the present findings suggest that the power output derived from jumping and sprinting tasks may not reflect the same underlying mechanical or func- tional quality. This finding may be explained by the fact that power expression is constrained by task-spe- cific movement mechanics, force orientation, and ve- locity conditions. Accordingly, a player who appears highly powerful in a vertical jump context may not nec- essarily exhibit comparable power characteristics dur- ing sprint acceleration. This interpretation is consist- ent with the view that mechanical outputs should be interpreted within the specific context in which they are produced, rather than as universally transferable indicators of performance [4, 13, 38, 40].
A particularly meaningful aspect of the present study lies in the mismatch observed between vertical and horizontal profile orientations. While the vertical assessment showed a predominance of velocity-oriented profiles, the horizontal assessment revealed a more balanced distribution, with a greater proportion of force- oriented and optimal profiles [5]. This pattern may partly reflect the mechanical characteristics of the jump task itself, in which force is expressed against gravity, and movement execution is strongly influenced by ver- tical propulsion strategy. It may also reflect the neuro- muscular characteristics of the present sample, or the specific training background of these players, who are regularly exposed to sprinting, reactive actions, and stretch-shortening cycle demands in football [16]. Al- though this interpretation remains speculative, it sug- gests that profile orientation may be influenced not only by the athlete’s physical qualities, but also by the mechanical nature of the task used for the assessment [13]. These findings support the concept that force-ve- locity characteristics are task-specific and should there- fore be interpreted within the context of the assessment performed.
In practical terms, an athlete classified as velocity- oriented in a jump-based task may not necessarily present the same orientation in a sprint-based task. This challenges the assumption that a single mechanical profile can adequately represent an athlete’s neuro- muscular characteristics and highlights the potential for misclassification when relying on a single profiling modality. From this perspective, the concept of “pro- file mismatch” may provide a more informative frame- work for interpreting mechanical characteristics across different movement tasks as such mismatches may lead to incomplete or suboptimal training decisions. These findings highlight the importance of integrat- ing both vertical and horizontal force-velocity assess- ments to support more comprehensive and individu- alised training prescription. Consequently, relying on a single test may result in training prescriptions that do not adequately target the athlete’s actual mechani- cal needs. This may reinforce the practical importance of combining complementary profiling approaches when aiming to individualise training and optimise transfer to football-specific performance [3, 4].
Another important finding was the substantial in- ter-individual variability observed across the sample. Despite belonging to the same competitive category and training environment, players displayed heterogeneous mechanical profiles and profile orientations. This ob- servation is consistent with previous works showing that athletes with comparable performance levels may nevertheless present markedly different force-velocity characteristics and mechanical deficits [16, 30]. Such variability likely reflects the combined influence of anthropometry, training background, neuromuscular adaptations, and individual movement strategies. From a practical perspective, this finding reinforces the rele- vance of individualised profiling in football, as group- level averages may obscure meaningful differences in player-specific mechanical needs and consequently reduce the precision of training prescription [13, 14].
From an applied perspective, the present study re- sults may support the integration of both vertical and horizontal F-V profiling within performance assess- ment and training design. If the primary goal is to improve acceleration performance, the present find- ings suggest that practitioners may benefit from placing greater emphasis on horizontal mechanical outputs and sprint-specific force application. In this context, interventions such as resisted sprinting, particularly with moderate-to-heavy sled loads, may be relevant when targeting horizontal force production [12]. Like- wise, horizontally-oriented plyometric exercises may offer more specific transfer to sprint acceleration than vertically-dominant exercises in some football contexts [42]. However, this should not lead to the exclusion of vertical profiling or vertical power-oriented training, as these remain relevant for other performance de- mands such as jumping, landing, and aerial duels. Rather, the practical implication is that performance assessment and training prescription should reflect the multidimensional and task-specific nature of foot- ball performance.
The present study also provides novel data on Mo- roccan football players, a population that remains un- derrepresented in literature addressing the force-veloc- ity profiling. Most available studies in this area have been conducted in European or South American cohorts, which may limit the ecological transferability of existing evidence to other football environments. By examining both vertical and horizontal F-V char- acteristics among the Moroccan U23 players, the pre- sent study contributes to expanding the available evi- dence base and may provide useful reference points for practitioners working in similar developmental and competitive contexts.
Several limitations should nevertheless be acknowl- edged. First, the relatively small sample size limits the statistical precision of the observed relationships and requires cautious interpretation, particularly when con- sidering weak or moderate associations. This limitation is further reflected in the relatively wide confidence intervals observed for several correlation coefficients, which often included zero. This indicates a degree of uncertainty regarding both the magnitude and direc- tion of the relationships, and suggests that the present findings should be interpreted as indicative of tenden- cies rather than definitive evidence of stable mechani- cal associations. Second, the cross-sectional design pre- cludes any causal inference regarding the relationship between F-V characteristics and performance out- comes. Third, although MyJump2 and MySprint offer practical and ecologically valid field-based assessment options, some degree of measurement error compared with laboratory-based systems cannot be excluded [41]. Finally, potentially relevant factors such as playing position, training history, and accumulated fatigue were not fully controlled and may have contributed to the observed variability.
Future research should examine these relationships in larger and more diverse football samples, while also exploring the longitudinal effects of individualised training interventions based on combined vertical and horizontal F-V profiling. In particular, future studies may benefit from investigating whether profile mis- match between movement tasks represents a stable mechanical characteristic or a trainable feature that can be modified through targeted intervention.
Conclusions
The present findings indicate that vertical and hor- izontal F-V profiles among Moroccan male U23 foot- ball players show only limited association, supporting the view that these assessments may reflect distinct movement-specific mechanical tendencies rather than a single global expression of explosiveness. The ob- served profile mismatch and the substantial inter-in- dividual variability further emphasise the complexity of neuromuscular profiling in football and the limi- tations of relying on a single testing modality. Practi- cally, these results support the integration of both jump- and sprint-based mechanical profiling to better characterise individual player needs and to guide more precise training prescription. By providing novel evi- dence from an underrepresented football population, this study contributes to the evidence base support- ing multidimensional and individualised approaches to performance assessment in football.
