Spatiotemporal and kinetic characteristics during maximal sprint running in fast running soccer players

Abstract

This study aimed to elucidate spatiotemporal and kinetic variables in fast-running soccer players in comparison with sprinters or slow-running soccer players. Sixty-seven male soccer players and 17 male sprinters (Sp) performed 60-m maximal effort sprint running. The soccer players were classified into three groups: high-speed (SOC_High), medium-speed, and low-speed (SOC_Low). The antero-posterior and vertical ground reaction forces were measured with a 50-m long force plates system at every step during the sprint. Step length and step frequency were also computed from the position of center of pressure, contact time, and flight time. During the initial acceleration phase, SOC_High exhibited similar running speeds to Sp. This was attributed to a higher step frequency in SOC_High compared to Sp, while net antero-posterior impulse was lower in the former than in the later. In the range of running speed from 7.5 m/s to 8.5 m/s, net antero-posterior impulse for SOC_High was similar to that for Sp. At 9.0 m/s, SOC_High exhibited a lower net antero-posterior impulse compared to Sp, primarily due to a reduced propulsive impulse. Additionally, vertical impulse during the braking phase was larger in SOC_High compared to Sp, due to a longer braking time, while vertical impulse during the propulsive phase was smaller, due to a tendency for a reduced propulsive time and vertical force during the corresponding phase. Compared to SOC_Low, SOC_High exhibited higher step frequency through sprint running and longer step lengths from the 2nd acceleration phase to maximal speed phase. Additionally, net antero-posterior impulse at the same running speed was greater in SOC_High compared to SOC_Low. Vertical impulse was lower during the braking phase but higher during the propulsive phase in SOC_High than in SOC_Low. Thus, the sprint mechanics of SOC_High is characterized by a similar ability of speed acquisition up to the 2^nd acceleration as sprinters. However, at 9.0 m/s or over, SOC_High exhibits a greater vertical impulse, leading to a lower step frequency.

Introduction

In soccer players, sprint running ability is one of the crucial factors related to competition performance [1, 2]. Approximately 10% of the total distance covered during a game (ranging from 10–12 km) involves near-straight running [3]. Moreover, high-speed straight running often precedes scoring opportunities for both offensive players and defenders [1,2,4]. The relative distribution of high-speed running (defined as speeds >6.67 m/s) is approximately 7% for distances of ~10m, 48% for distances of ~20m, and 45% for distances exceeding 20m [5]. This implies that soccer players are required to run over both short and long distances at high speeds during a game. In essence, both high acceleration ability and maximum sprint speed are crucial for high level soccer players. Therefore, elucidating the mechanics of maximal sprint running in fast-running soccer players during the acceleration and maximal speed phases will provide useful information for designing effective training programs in practical fields.

There are three primary phases during sprint running to achieve maximal speed: the 1^st acceleration, the 2^nd acceleration, and the maximal speed phases [6–9]. Although maximal running speed depends on speed gain in the preceding acceleration phase [10], the correlation coefficients between acceleration and maximal running speeds have been shown to range from 0.56–0.79 [11–13], indicating potential differences in sprint mechanics between the acceleration and maximal running speed phases [14]. Thus, examining sprint mechanics during both phases will provide useful information to elucidate the mechanics for achieving high sprint running performance.

Spatiotemporal and kinetics parameters of sprint running during acceleration and maximal speed phases have been observed to differ between competitive sprinters (hereafter referred to as sprinters) and athlete non-sprinters (non-sprinters), as well as among non-sprinters with varying running capabilities within the same event [15–17]. For example, it has been reported that the antero-posterior ground reaction forces (GRFs) varied between faster and slower sprinters throughout the acceleration phase [18]. Further, the propulsive and braking forces were lower in soccer players than in sprinters at equivalent speed (8 and 8.5 m/s) during the 2^nd acceleration phase [15]. During the maximal speed phase, non-sprinters exhibit longer contact times and shorter step lengths compared with sprinters [19]. As compared to non-sprinters, sprinters demonstrate larger vertical GRFs and exert greater vertical GRFs during the first half of ground contact in the maximal speed phase [16]. Furthermore, elite soccer players demonstrate a higher theoretical maximal horizontal force derived from horizontal force-velocity relationship compared with sub-elite soccer players [17], indicating that the formers have greater acceleration capability in the initial acceleration phase. Previous studies have shown that soccer players exhibit inferior horizontal force exertion at high running speeds and differ in the spatiotemporal variables during maximal speed phase compared with sprinters [15]. Additionally, high-level soccer players exhibit superior antero-posterior force production in the acceleration phase compared with low-level players [17]. However, current literature does not clarify whether fast-running soccer players can generate antero-posterior GRFs similar to that of sprinters in the initial acceleration phase. It remains unknown how the vertical components of GRFs differ among soccer players with different running speeds, as well as between sprinters and fast-running soccer players in the 2^nd acceleration and maximal speed phases.

Therefore, this study aimed to elucidate the differences in spatiotemporal and kinetics parameters between sprinters and fast-running soccer players, and between fast- and slow-running soccer players. We hypothesized that: 1) in the 1^st acceleration phase, antero-posterior force production and spatiotemporal variables for fast-running soccer players would be similar to that of sprinters and superior to that of slow-running soccer players; 2) in the 2^nd acceleration and the maximal speed phases, fast-running soccer players would produce lower vertical force and would have lower step frequency and step length compared with sprinters.

Methods

Participants

Data were collected from December 21, 2021 to September 17, 2023. Sixty-seven male soccer players (SOC) and 17 male sprinters (Sp; 20.6 ± 0.9 years, 1.73 ± 0.05 m, 67.7 ± 5.1 kg; mean ± SD) participated in this study. All participants underwent their specific training 5-6 times per week, with each session lasting 1.5-2 hours. None of the participants reported any illness and were not prescribed medications for cardiovascular, metabolic, or orthopedic disorders. According to the athlete taxonomy classification reported [18], the study participants categorized as Tier 3 or Tier 4. The participants had been competing in sprint events for more than 5 years. Based on the maximal running speed (as described in the data analysis section), soccer players were divided into three groups: high-speed running soccer players whose speed was above the mean + 0.5SD (SOC_High; n = 16, 20.1 ± 0.9 years, 1.72 ± 0.05 m, 68.1 ± 7.3 kg), medium-speed soccer players whose speed was within the mean ± 0.5SD (SOC_Med; n = 30, 20.3 ± 1.1 years, 1.73 ± 0.05 m, 67.5 ± 5.4 kg), and low-speed soccer players whose speed was below the mean - 0.5SD (SOC_Low; n = 21, 20.2 ± 1.9 years, 1.71 ± 0.05 m, 65.7 ± 5.3 kg). Physical characteristics of the participants did not significantly differ between groups. This study was approved by the ethics committee of the National Institute of Fitness and Sports in Kanoya (#5-60). All procedures were conducted in accordance with the Declaration of Helsinki. Prior to the experiments, all the participants were fully informed of the purpose and risks of the experiments and written consent was obtained.

Experimental protocol

Prior to the test session, all participants engaged in approximately 30 min of warm-up exercises, comprising of approximately 8 min of jogging to running, approximately 10 min of static and dynamic stretching, approximately 10 min of running technique exercises, 2–4 submaximal-to-maximal sprint running. Following at least 10 min of rest after the warm-up session, the participants performed maximal effort sprint running from standing split-stance position on an indoor track with 54 force plates (TF-90100, TF-3055, and TF-32120; Tec Gihan, Uji, Japan), as described in previous studies [7,19]. The soccer players wore their own running shoes, and the sprinters used their own spikes.

Data analysis

Antero-posterior and vertical force signals were used for analysis. The analog force signals from each force plate were captured in a control box (FP Control Unit, Tec Gihan, Japan), converted from analog to digital, and transmitted to a computer. The 50-m long force plates were treated as a single unit using dedicated software, and the antero-posterior and vertical forces and the location of the center of pressure on the track during ground contact, were computed at a sampling rate of 1000 Hz. Fig 1 presents representative data on antero-posterior and vertical ground reaction forces for SOC_High, SOC_Low, and Sp.

[Figure omitted. See PDF.]

(A) First step, (B) Running speed of approximately 8.0 m/s, (C) Second acceleration phase, (D) Running speed of approximately 9.0 m/s, and (E) Maximal speed phase. The solid black line represents a fast-running soccer player, the solid gray line represents a sprinter, the dotted black line represents a slow-running soccer player.

In accordance with the methodology described in our previous study [19], the force data underwent filtering using a fourth-order zero-lag low-pass Butterworth filter with a cutoff frequency of 50 Hz. Ground contact and takeoff times were determined based on thresholds set at 20 N for vertical force, allowing identification of the braking and propulsion phases based on negative and positive values of the antero-posterior force, respectively. Spatiotemporal parameters were calculated as described in previous studies [19,20]. Ground contact time was defined as the duration between ground contact and takeoff for each step. Flight time was the period between takeoff in one step and ground contact in the next step. Step frequency was calculated as the inverse of the sum of contact time and flight time, and step length was determined as the distance between the midpoint of ground contact and takeoff in one step and the corresponding midpoint in the next step. Running speed was obtained by multiplying step frequency and step length. Antero-posterior and vertical forces were averaged over the duration from ground contact to takeoff for each step, and net antero-posterior and vertical impulses were computed through time integration of these forces. To mitigate the influence of left-right imbalance on the measured variables, the step-by-step data were smoothed using a moving average over two steps. In accordance with a previously published study [21], the average value of each variable was calculated for each participant during the first four steps (A_1st), the middle four steps (specific to individuals) (A_2nd), and four steps including maximal speed (including two steps before and one step after maximal speed (MS). The middle four steps were defined relative to the middle step during the acceleration phase from the start to the step at maximal speed.

To examine group differences in each variable at the same running speed, we interpolated each variable as a function of running speed within the data range of each participant. From the interpolated data, we extracted the measured variables at 7.5 m/s, 8.0 m/s, 8.5 m/s, and 9.0 m/s. In SOC_Low, the running speed of two participants did not reach a maximum of 8.0 m/s. In SOC_Med, three participants did not reach 8.5 m/s, while in SOC_High, two participants did not reach 9.0 m/s. In Sp, one participant also did not reach 9.0 m/s. Therefore, the independent variables at the corresponding running speed were treated as missing values. To determine the theoretical maximal running speed (MS₀), the stance-averaged horizontal forces were plotted against running speed and fitted linearly with a first-order polynomial function. The y-intercept was computed as MS₀. The running speed in each phase was also expressed as the value relative to MS₀ (%MS₀).

Statistical analysis

Descriptive data are presented as means and standard deviations. Given the assumption of group differences in running speed, a generalized linear mixed effect model (GLMM) was employed to examine group differences in the independent variables. The model incorporated group (Sp, SOC_High, SOC_Med, and SOC_Low), phase (7.5 m/s, 8.0 m/s, 8.5 m/s, 9.0 m/s, A_1st, A_2nd, and V_max), and a group-by-phase interaction. The GLMM included a random intercept to accommodate random variation in running speed at the maximal speed phase between participants and a random slope for phase. This was done to address random variation changes in the independent variables across phases between the participants. Covariation between the random intercept and slope was assessed to determine if the change across phases was dependent on running speed in the maximal speed phase. When significant main effects of group and phase were detected, a Bonferroni test was conducted with all combinations for post hoc comparisons. Additionally, in the presence of a significant group-by-phase interaction, a GLMM with a Bonferroni test was applied within each group to assess the significance of phase-related differences in the independent variables. One-way analysis of variance (ANOVA) with factorial analysis was conducted to test group differences in mechanical parameters derived from the stance-averaged GRFs and running speed relationship. In this analysis, when the F value was significant, a Bonferroni test was conducted for post hoc comparisons. The level of significance was set at 5%, and the p-value was computed using statistical software (IBM SPSS Statistics 26, IBM, Japan).

Results

Fig 2 summarizes the main results of this study. To characterize the spatiotemporal and kinetic parameters of fast-running soccer players, we compared each variable between SOC_High and Sp and between SOC_High and SOC_Low.

[Figure omitted. See PDF.]

Data points are represented as sprinters (white-filled triangles), fast-running soccer players (black-filled circles), medium-running soccer players (gray-filled circles), and slow-running soccer players (white-filled circles).

Initial sprint acceleration phase (A_1st)

Table 1 presents running speed and spatiotemporal variables during sprint running for all groups. No significant difference was found in running speed between SOC_High and Sp, but the %MS₀ tended to be higher in SOC_High compared with Sp (p = 0.061). Running speed was significantly higher in SOC_High than in SOC_Low. SOC_High exhibited a shorter propulsive time, resulting in a shorter ground contact time and a higher step frequency compared to Sp and SOC_Low (Table 1). There were no significant group differences in the braking time. Flight time tended to be longer in SOC_High compared to SOC_Low (p = 0.057). There was no significant difference in step length between SOC_High and Sp or between SOC_High and SOC_Low.

[Figure omitted. See PDF.]

Table 2 presents antero-posterior and vertical force production during sprint running. SOC_High exhibited lower net antero-posterior impulse and propulsive impulse compared to Sp. No significant differences in stance-averaged antero-posterior force, braking or propulsive force were observed between SOC_High and Sp. Net antero-posterior impulse was greater in SOC_High than in SOC_Low, resulting from a greater propulsive impulse. SOC_High exhibited a greater propulsive force compared to SOC_Low, resulting in a greater stance-averaged antero-posterior force.

[Figure omitted. See PDF.]

The vertical impulse was lower in SOC_High than in Sp and SOC_Low (Table 2). The averaged vertical force during the propulsive phase tended to be greater in SOC_High compared with SOC_Low (p = 0.058).

The second acceleration phase (A_2nd)

Running speed was lower in SOC_High than that in Sp but higher than in SOC_Low (Table 1). There were no significant group differences in the %MS₀. No significant difference in ground contact time was found between SOC_High and Sp (Table 1). SOC_High exhibited a shorter ground contact time compared to SOC_Low, primarily due to a shorter propulsive time. There was no significant difference in flight time between SOC_High and Sp, nor between SOC_High and SOC_Low. Step length was shorter in SOC_High compared to Sp, while no significant difference in step frequency was observed between the two group. SOC_High demonstrated a higher step frequency and a longer step length compared to SOC_Low.

There were no significant differences between SOC_High and Sp and between SOC_High and SOC_Low in each of the net antero-posterior impulse, braking impulse and propulsive impulse (Table 2). SOC_High had a lower braking force compared with Sp. Propulsive force was greater in SOC_High compared to SOC_Low.

No significant differences were found between SOC_High and Sp in vertical impulse, vertical impulse during the braking or propulsive phases (Table 2). The averaged vertical force during the propulsive phase was lower in SOC_High than in Sp. SOC_High exhibited a lower vertical impulse and greater averaged vertical force and vertical force during the propulsive phase compared to SOC_Low.

Maximal speed phase (MS)

Maximal running speed was lower in SOC_High than in Sp but higher than in SOC_Low (Table 1). There was no significant group difference in the %MS₀. SOC_High exhibited a longer braking time compared to Sp, resulting in a longer ground contact time. Flight time was shorter in SOC_High than in Sp. SOC_High demonstrated a shorter step length compared to Sp, while no significant difference in step frequency was observed. Compared to SOC_Low, SOC_High exhibited a shorter ground contact time. Braking time tended to be shorter in SOC_High compared to SOC_Low (p = 0.058). No significant difference was observed in flight time between both groups. SOC_High exhibited a longer step length and a higher step frequency compared to SOC_Low.

There were no significant differences between SOC_High and Sp, nor between SOC_High and SOC_Low in net antero-posterior impulse, braking impulse or propulsive impulse. Both braking and propulsive forces were lower in SOC_High compared to Sp. There were no significant differences between SOC_High and SOC_Low in stance-averaged antero-posterior force, braking or propulsive forces.

There were no significant differences between SOC_High and Sp in vertical impulse, vertical impulses during the braking and propulsive phases. Stance-averaged vertical force was lower in SOC_High than in Sp, primarily due to lower averaged vertical forces during the propulsive phases. The averaged vertical force during the braking phase tended to be lower in SOC_High than in Sp (p = 0.060). Vertical impulse was lower in SOC_High than in SOC_Low. SOC_High exhibited a greater averaged vertical force compared to SOC_Low.

Comparison between fast-running soccer players and sprinters at the same running speed (Table 3)

Regardless of running speed, %MS₀ was higher in SOC_High compared to Sp. No significant differences in ground contact time and flight time were observed between the two groups at any running speed. At 9.0 m/s, braking time was longer but propulsive time tended to be shorter (p = 0.062) in SOC_High than in Sp. There were no significant differences in step frequency or step length between SOC_High and Sp at any running speed, except for 7.5 m/s.

[Figure omitted. See PDF.]

At 9.0 m/s, net antero-posterior impulse, stance-averaged antero-posterior and braking forces were lower in SOC_High compared to Sp, but no significant differences were found between the two groups in those at other running speeds. Propulsive impulses at 8.0 m/s and more were lower, and propulsive forces at 7.5 m/s and over were lower in SOC_High than in Sp. No significant differences were observed between the two groups in braking impulse at any running speed.

Compared to Sp, SOC_High exhibited larger vertical impulse during the braking phase but lower vertical impulse during the propulsive phase at 9.0 m/s, while the corresponding difference was found in vertical impulse on ground contact. There were no significant differences between the two groups in vertical impulse during the braking or propulsive phases at any running speed. At 9.0 m/s, the averaged vertical force during the propulsive phase was lower in SOC_High compared to Sp, while no significant differences were found between the two groups in the stance-averaged vertical and vertical forces during the braking phase.

Comparison between fast- and slow-running soccer players at the same running speed (Table 4)

Regardless of running speed, %MS₀ was lower in SOC_High compared to SOC_Low. No significant differences were observed between the two groups in ground contact time and flight time. Braking time was shorter in SOC_High compared to SOC_Low, while propulsive time was longer. SOC_High exhibited higher step frequency compared to SOC_Low. Step length at 8.0 m/s tended to be shorter in SOC_High than in SOC_Low, but the difference was not significant (p = 0.070).

[Figure omitted. See PDF.]

At 7.5 m/s and 8.0 m/s, net antero-posterior impulse was greater in SOC_High compared to SOC_Low, primarily due to lower braking impulse and larger propulsive impulse in the former. The stance-averaged antero-posterior forces at the two running speeds were higher in SOC_High than in SOC_Low. At 8.0 m/s, propulsive force was higher in SOC_High than in SOC_Low, and braking force tended to be lower in the former compared to the latter (p = 0.060).

At 7.5 m/s and 8.0 m/s, vertical impulses during the braking phase were lower and those during the propulsive phase were higher in SOC_High than in SOC_Low. There was no significant difference in stance-averaged vertical force between the two groups. However, the averaged vertical force during the braking phase was smaller and that during propulsive phase was greater in SOC_High compared to SOC_Low.

Discussion

This study conducted group comparisons through the two approaches: (1) aligning sprint running phase based on definitions from a previous study [8] and (2) matching running speeds between SOC_High and Sp (7.5–9.0 m/s), as well as among soccer players (7.5 m/s and 9.0 m/s). The first approach revealed that running speed for SOC_High was similar to that for Sp during the A_1st but was lower than Sp during the A_2nd and the MS phases. The second approach indicated that at 9.0 m/s, SOC_High exhibited smaller net antero-posterior impulse and propulsive impulse compared to Sp. In SOC_High, vertical impulse during the braking phase increased due to a longer braking time, whereas vertical impulse during the propulsive phase decreased due to both a shorter propulsive time and a lower vertical force during the corresponding phase. In the running phase from A_1st to MS, running speeds were higher in SOC_High than in SOC_Low. This was due to a higher step frequency throughout all sprint phases and a longer step length from the A_2nd to the MS in SOC_High. Additionally, SOC_High exhibited a greater net antero-posterior impulse during the A_1st and lower vertical impulse throughout entire running phase compared to SOC_Low. These results were consistent regardless of the approaches. Thus, the current results indicate that as compared to sprinters, fast-running soccer players may exhibit different sprint mechanics for speed acquisition during the initial acceleration phase, although they can achieve similar running speeds and maintain comparable acceleration capability up to 8.5 m/s, but beyond the 2^nd acceleration phase, they generate lower propulsive forces and greater vertical force production during the braking phase due to a longer braking time compared to sprinters.

Initial sprint acceleration phase

In the A_1st, SOC_High exhibited higher running speed compared with SOC_Low. Running speed is theoretically the product of step length and step frequency. Step frequency was higher in SOC_High than in SOC_Low, whereas no significant difference was found in step length between the two groups. Therefore, group difference in running speed depends on step frequency. Consequently, a shorter ground contact time resulted in a higher step frequency for SOC_High compared with SOC_Low. Additionally, in this study, the antero-posterior impulse was higher in SOC_High than in SOC_Low. Step length is known to be associated with both antero-posterior and vertical impulses [22]. A lack of group difference in step length could be due to a lower vertical impulse in SOC_High in comparison with SOC_Low. Considering that stance-averaged antero-posterior force was greater in SOC_High than in SOC_Low, it is likely that SOC_High can generate greater antero-posterior force than SOC_Low even within a short duration of ground contact in the 1^st acceleration phase.

The SOC_High demonstrated a running speed comparable to that of Sp. The step frequency was higher in SOC_High compared with Sp, resulting from a shorter ground contact time. On the other hand, SOC_High had a lower net antero-posterior impulse compared to Sp, primarily due to a lower propulsive impulse. There were no significant differences in stance-averaged antero-posterior force, baking or propulsive forces between the two groups. Therefore, fast-running soccer players may be able to achieve similar speeds to sprinters by reducing ground contact time, particularly during the propulsive phase, and increasing step frequency.

The 2^nd acceleration phase

During the A_2nd, SOC_High exhibited a higher running speed compared to SOC_Low. The mechanism contributing to the difference in running speed between the two groups was associated with the magnitude of step frequency and step length. Compared to SOC_Low, SOC_High demonstrated higher step frequency and longer step length. SOC_High had shorter ground contact time, and higher stance-averaged vertical force as compared to SOC_Low. The stance-averaged vertical force was negatively related to ground contact time in the A_2nd (r = 0.726 for soccer players) (unpresented data). These indicate that as compared to SOC_Low, SOC_High can produce greater vertical force within a short ground contact time. Additionally, since there was no significant difference in the flight time between SOC_High and SOC_Low, a higher step frequency for SOC_High may be due to a shorter ground contact time than SOC_Low. The vertical and antero-posterior impulses could not explain the group differences in step length. Net antero-posterior impulse is associated with the subsequent change in running speed during the entire acceleration phase [23]. The lack of difference in antero-posterior impulse indicates that there was no difference in the change in running speed. Additionally, the stance-averaged antero-posterior force in the first step correlates with each 5-m split time from 5 m to 20 m for semi-professional soccer players [24]. Thus, the difference in the antero-posterior impulse before the 2^nd acceleration phase might contribute to the observed difference in running speed. In addition, antero-posterior impulse is a determinant of step length [22]. In this study, the net antero-posterior impulse in the initial acceleration phase was higher in SOC_High compared with SOC_Low. It implies that group difference in step length might be due to the antero-posterior impulse before the 2^nd acceleration phase. Therefore, a higher running speed observed in SOC_High during the 2^nd acceleration phase could be due to a longer step length, driven by a greater antero-posterior impulse before the A_2nd, as well as a higher step frequency, resulting from shorter ground contact time and greater vertical force production.

SOC_High exhibited a lower running speed compared to Sp. Step length was lower in SOC_High than in Sp, whereas no significant difference was found in step frequency between both groups. Therefore, the difference in the running speed between SOC_High and Sp was due to differences in the step length. In the 1^st acceleration phase, SOC_High had a higher step frequency compared to Sp, but there was no significant difference in step frequency between the two groups in the 2^nd acceleration phase. This could be due to the lack of group differences in ground contact time and flight time during the 2^nd acceleration phase. On the other hand, while no significant differences in antero-posterior and vertical impulses were observed, SOC_High had a lower step length compared to Sp. As for the antero-posterior components, a previous study demonstrated that soccer players exhibit lower net antero-posterior impulse than Sp, and the force vector is more inclined in the direction of propulsion in high-speed running (8 and 8.5 m/s) [15]. This discrepancy may stem from differences in analytical approaches: comparing at the same running speed vs. within the same sprint phase. To examine this assumption, we plotted the relationship between each measured variable and %MS₀ (S1 Fig). The results showed that both net antero-posterior impulse and stance-averaged antero-posterior force exhibited linear relationships, regardless of the group. This suggests that, despite differences in absolute running speed, the ability to accelerate within the same running phase may be comparable between sprinters and fast-running soccer players. As mentioned above, the group differences in step length may have been attributable to the antero-posterior impulse before the 2^nd acceleration phase. Therefore, the difference in running speed between SOC_High and Sp during the 2^nd acceleration phase could be due to a greater antero-posterior impulse before the A_2nd.

Maximal speed phase

In the MS, SOC_High demonstrated a higher running speed compared to SOC_Low, resulting from a longer step length and a higher step frequency. The higher step frequency observed in SOC_High can be attributed to a shorter ground contact time. It is known that footballers with high vertical stiffness during hopping tasks have higher maximal speed compared to footballers with low vertical stiffness, and the group differences in sprint performance can be observed at a distance of 30–60 m [25]. Combining the current results with the previous findings, it can be suggested that SOC_High might have higher vertical stiffness via stance-averaged vertical forces compared with SOC_Low. Additionally, since no significant group differences in antero-posterior and vertical impulses were found between the two groups, longer step length in SOC_High might result from the magnitude of running speed before maximal speed phase. Therefore, SOC_High can produce greater vertical force during a short ground contact compared with SOC_Low in the maximal speed phase.

The difference in running speed between SOC_High and Sp mirrored that in the A_2nd. Step length was shorter in SOC_High than in Sp, whereas no significant difference was found in step frequency between both groups. These findings support the previous findings that during the maximal speed phase, non-sprinters demonstrated shorter step lengths whereas step frequency of non-sprinters was comparable to that of sprinters [26]. The difference in running speed was due to the magnitude of step length between the two groups. Shorter step length for SOC_High in comparison with Sp might be due to the difference in running speed before the maximal speed phase. SOC_High exhibited vertical impulse comparable to Sp, but SOC_High had a longer ground contact time and a shorter flight time. Since step frequency is the inverse of the sum of ground contact time and flight time, the longer ground contact and shorter flight time offset each other, resulting in no difference between SOC_High and Sp for step frequency. A previous study found that trained team sport athletes exhibited lower vertical stiffness compared with sprinters [26]. A stiffer leg spring allows high vertical forces [27] and a short contact time [28]. In this study, the stance-averaged vertical force was lower in SOC_High compared to Sp, which is consistent with the previous findings [16,29]. The vertical force during the braking phase tended to be lower in SOC_High than in Sp. Clark and Weyand [16] demonstrated a positive correlation between maximal running speed and stance-averaged vertical force during the braking phase. Additionally, SOC_High exhibited a lower vertical force during the propulsive phase compared to Sp. The development of maximal speed during sprinting is accompanied by an increase in vertical stiffness [30]. Based on these findings, it is likely that fast-running soccer players cannot stiffen their lower limbs particularly during the maximal speed phase, and consequently generate lower vertical forces, and experience longer ground contact times compared to sprinters.

Factors contributing to differences in maximal running speed between fast-running soccer players and sprinters

In this study, we compared spatiotemporal variables and ground reaction forces between SOC_High and Sp with relation to the same running speed and the same running phase. In this section, we discuss the factors contributing to the lower maximal running speed of SOC_High compared to Sp, based on the findings of this study. As seen in Fig 1, SOC_High exhibited a higher step frequency than Sp up to 7.5 m/s; however, beyond this speed, no significant difference was observed. Additionally, in SOC_High, step frequency decreased from A_2nd to MS, whereas it remained constant during the corresponding phase in Sp. This decrease may be attributable to a longer ground contact time in SOC_High, likely due to lower vertical forces, particularly during the propulsive phase. As greater vertical force during ground contact was associated with a shorter ground contact time above mentioned, it is assumed that SOC_High may be unable to reduce their ground contact time due to lower vertical force during the propulsive phase, which may have prevented them from maintaining step frequency.

Another mechanism underlying the difference in maximal running speed between SOC_High and Sp may be the difference in net antero-posterior impulse at the same running speed. At 9.0 m/s, SOC_High exhibited a lower net antero-posterior impulse, primarily due to a lower propulsive impulse compared to Sp. Given these findings, the lower maximal running speed in SOC_High may be attributable to the inability to generate sufficient acceleration to achieve higher speeds beyond A_2nd. Taken together, fast-running soccer players exhibit lower acceleration capability beyond the 2^nd acceleration phase and reduced vertical force during the propulsive phase, which may contribute to their lower maximal running speed compared to sprinters.

Factors contributing to differences in maximal running speed between fast- and slow-running soccer players

In this section, we discuss the factors contributing to the lower maximal running speed of SOC_High compared to SOC_Low, based on the findings of this study. As shown in Fig 1, SOC_High exhibited a higher step frequency during entire sprint running. This may be due to a shorter flight time in SOC_High, primarily resulting from lower vertical impulse, as no significant differences were observed in ground contact time between the two groups. Additionally, beyond A_2nd, SOC_High exhibited higher vertical forces during ground contact, particularly in the propulsive phase, leading to a shorter ground contact time compared to SOC_Low. Therefore, the higher step frequency in SOC_High beyond the 2^nd acceleration phase can be attributed to greater vertical force.

In both groups, step length increased with running speed, and SOC_High was able to extend their step length even further as running speed increased. Both antero-posterior and vertical impulses are determinants of step length [22]. Since vertical impulse was lower in SOC_High compared to SOC_Low, the longer step length in SOC_High may be attributed to a greater net antero-posterior impulse. In particular, SOC_High exhibited less braking impulse due to brief braking time and greater propulsive impulse resulting from both longer propulsive time and propulsive force up to the A_2nd. Based on these findings, fast-running soccer players can produce greater propulsive force throughout the entire running phase and generate larger vertical forces at higher speeds compared to slow-running soccer players.

Limitations of this study

There are some limitations in this study. First, we did not have data on sprint running kinematics. A study by Garden [31] has shown that the kinematics of sprint running may differ between sprinters and soccer players, which could explain the differences in spatiotemporal and mechanical variables observed in this study. In particular, biomechanical analysis is needed to investigate the factors contributing to the increase in braking time and the reduced propulsive time from the 2^nd acceleration phase to the maximal speed phase. Second, the difference in ground reaction forces between sprinters and soccer players might be influenced by footwear (spiked shoes vs. running shoes). To the best of our knowledge, no study has explicitly examined this factor. In this study, participants were instructed to wear their regularly used shoes to ensure familiarity and natural sprint performance. Finally, this study examined regional collegiate soccer players, whereas the previous study included regional to national level players [17]. Further research is needed to clarify these points.

Conclusion

Sprint mechanics of fast-running soccer players is characterized by similar ability of speed acquisition up to the 2^nd acceleration to sprinters but, at above 9.0 m/s and more, they exhibit an increase in vertical impulse during ground contact, leading to a lower step frequency. Additionally, fast-running soccer players produce greater acceleration capacity, lower vertical force during the braking phase, and higher vertical force during the propulsive phase compared to slow-running soccer players.

Supporting information

S1 Fig. Associations of spatiotemporal variables and ground reaction forces with the percentage of running speed relative to theoretical maximal running speed.

Data points are represented as sprinters (white-filled triangles), fast-running soccer players (black-filled circles), medium-running soccer players (gray-filled circles), and slow-running soccer players (white-filled circles).

https://doi.org/10.1371/journal.pone.0322216.s001

(DOCX)

S1 Dataset. SupplInfo_PlosOne DataSet.

https://doi.org/10.1371/journal.pone.0322216.s002

(XLSX)

Acknowledgments

This study was supported by a NIFS project for the assessment of physical fitness for athletes.

References

1. 1. Faude O, Koch T, Meyer T. Straight sprinting is the most frequent action in goal situations in professional football. J Sports Sci. 2012;30(7):625–31. pmid:22394328

* View Article

* PubMed/NCBI

* Google Scholar

2. 2. Kai T, Horio K, Aoki T, Takai Y. High-intensity running is one of the determinants for achieving score-box possession during soccer matches. Football Science. 2018;15:61–9.

* View Article

* Google Scholar

3. 3. Kai T, Hirai S, Anbe Y, Takai Y. A new approach to quantify angles and time of changes-of-direction during soccer matches. PLoS One. 2021;16(5):e0251292. pmid:33999931

* View Article

* PubMed/NCBI

* Google Scholar

4. 4. Schulze E, Julian R, Meyer T. Exploring factors related to goal scoring opportunities in professional football. Sci Med Footb. 2022;6(2):181–8. pmid:35475738

* View Article

* PubMed/NCBI

* Google Scholar

5. 5. Andrzejewski M, Chmura J, Pluta B, Strzelczyk R, Kasprzak A. Analysis of sprinting activities of professional soccer players. J Strength Cond Res. 2013;27(8):2134–40. pmid:23168374

* View Article

* PubMed/NCBI

* Google Scholar

6. 6. Ettema G, McGhie D, Danielsen J, Sandbakk Ø, Haugen T. On the existence of step-to-step breakpoint transitions in accelerated sprinting. PLoS One. 2016;11(7):e0159701. pmid:27467387

* View Article

* PubMed/NCBI

* Google Scholar

7. 7. Nagahara R, Kanehisa H, Fukunaga T. Ground reaction force across the transition during sprint acceleration. Scand J Med Sci Sports. 2020;30(3):450–61. pmid:31705835

* View Article

* PubMed/NCBI

* Google Scholar

8. 8. Nagahara R, Matsubayashi T, Matsuo A, Zushi K. Kinematics of transition during human accelerated sprinting. Biol Open. 2014;3(8):689–99. pmid:24996923

* View Article

* PubMed/NCBI

* Google Scholar

9. 9. von Lieres Und Wilkau HC, Irwin G, Bezodis NE, Simpson S, Bezodis IN. Phase analysis in maximal sprinting: an investigation of step-to-step technical changes between the initial acceleration, transition and maximal velocity phases. Sports Biomech. 2020;19(2):141–56. pmid:29972337

* View Article

* PubMed/NCBI

* Google Scholar

10. 10. van Ingen Schenau GJ, de Koning JJ, de Groot G. Optimisation of sprinting performance in running, cycling and speed skating. Sports Med. 1994;17(4):259–75. pmid:8009139

* View Article

* PubMed/NCBI

* Google Scholar

11. 11. Harris NK, Cronin JB, Hopkins WG, Hansen KT. Relationship between sprint times and the strength/power outputs of a machine squat jump. J Strength Cond Res. 2008;22(3):691–8. pmid:18438252

* View Article

* PubMed/NCBI

* Google Scholar

12. 12. Little T, Williams AG. Specificity of acceleration, maximum speed, and agility in professional soccer players. J Strength Cond Res. 2005;19(1):76–8. pmid:15705049

* View Article

* PubMed/NCBI

* Google Scholar

13. 13. Mendez-Villanueva A, Buchheit M, Kuitunen S, Douglas A, Peltola E, Bourdon P. Age-related differences in acceleration, maximum running speed, and repeated-sprint performance in young soccer players. J Sports Sci. 2011;29(5):477–84. pmid:21225488

* View Article

* PubMed/NCBI

* Google Scholar

14. 14. Haugen TA, Tønnessen E, Seiler S. Anaerobic performance testing of professional soccer players 1995-2010. Int J Sports Physiol Perform. 2013;8(2):148–56. pmid:22868347

* View Article

* PubMed/NCBI

* Google Scholar

15. 15. Colyer SL, Nagahara R, Takai Y, Salo AIT. How sprinters accelerate beyond the velocity plateau of soccer players: waveform analysis of ground reaction forces. Scand J Med Sci Sports. 2018;28(12):2527–35. pmid:30230037

* View Article

* PubMed/NCBI

* Google Scholar

16. 16. Clark KP, Weyand PG. Are running speeds maximized with simple-spring stance mechanics?. J Appl Physiol (1985). 2014;117(6):604–15. pmid:25080925

* View Article

* PubMed/NCBI

* Google Scholar

17. 17. Devismes M, Aeles J, Philips J, Vanwanseele B. Sprint force-velocity profiles in soccer players: impact of sex and playing level. Sports Biomech. 2021;20(8):947–57. pmid:31223073

* View Article

* PubMed/NCBI

* Google Scholar

18. 18. McKay AKA, Stellingwerff T, Smith ES, Martin DT, Mujika I, Goosey-Tolfrey VL, et al. Defining training and performance caliber: a participant classification framework. Int J Sports Physiol Perform. 2022;17(2):317–31. pmid:34965513

* View Article

* PubMed/NCBI

* Google Scholar

19. 19. Sugisaki N, Kobayashi K, Yoshimoto T, Mitsukawa N, Tsuchie H, Takai Y, et al. Influence of horizontal resistance loads on spatiotemporal and ground reaction force variables during maximal sprint acceleration. PLoS One. 2023;18(12):e0295758. pmid:38085716

* View Article

* PubMed/NCBI

* Google Scholar

20. 20. Colyer SL, Nagahara R, Salo AIT. Kinetic demands of sprinting shift across the acceleration phase: Novel analysis of entire force waveforms. Scand J Med Sci Sports. 2018;28(7):1784–92. pmid:29630747

* View Article

* PubMed/NCBI

* Google Scholar

21. 21. Nagahara R, Takai Y, Haramura M, Mizutani M, Matsuo A, Kanehisa H, et al. Age-related differences in spatiotemporal variables and ground reaction forces during sprinting in boys. Pediatr Exerc Sci. 2018;30(3):335–44. pmid:29478372

* View Article

* PubMed/NCBI

* Google Scholar

22. 22. Hay GJ. The biomechanics of sports techniques. Edition t, editor. London: Prentice Hall International; 1994.

23. 23. Murata M, Takai Y, Kanehisa H, Fukunaga T, Nagahara R. Spatiotemporal and kinetic determinants of sprint acceleration performance in soccer players. Sports (Basel). 2018;6(4):169. pmid:30544869

* View Article

* PubMed/NCBI

* Google Scholar

24. 24. Wdowski MM, Gittoes MJR. First-stance phase force contributions to acceleration sprint performance in semi-professional soccer players. Eur J Sport Sci. 2020;20(3):366–74. pmid:31167614

* View Article

* PubMed/NCBI

* Google Scholar

25. 25. Kalkhoven JT, Watsford ML. The relationship between mechanical stiffness and athletic performance markers in sub-elite footballers. J Sports Sci. 2018;36(9):1022–9. pmid:28697691

* View Article

* PubMed/NCBI

* Google Scholar

26. 26. Douglas J, Pearson S, Ross A, McGuigan M. Reactive and eccentric strength contribute to stiffness regulation during maximum velocity sprinting in team sport athletes and highly trained sprinters. J Sports Sci. 2020;38(1):29–37. pmid:31631783

* View Article

* PubMed/NCBI

* Google Scholar

27. 27. Moir GL, Brimmer SM, Snyder BW, Connaboy C, Lamont HS. Mechanical limitations to sprinting and biomechanical solutions: a constraints-led framework for the incorporation of resistance training to develop sprinting speed. Strength Cond J. 2018;40(1):47–67.

* View Article

* Google Scholar

28. 28. Kuitunen S, Komi PV, Kyröläinen H. Knee and ankle joint stiffness in sprint running. Med Sci Sports Exerc. 2002;34(1):166–73. pmid:11782663

* View Article

* PubMed/NCBI

* Google Scholar

29. 29. Weyand PG, Sternlight DB, Bellizzi MJ, Wright S. Faster top running speeds are achieved with greater ground forces not more rapid leg movements. J Appl Physiol (1985). 2000;89(5):1991–9. pmid:11053354

* View Article

* PubMed/NCBI

* Google Scholar

30. 30. Nagahara R, Zushi K. Development of maximal speed sprinting performance with changes in vertical, leg and joint stiffness. J Sports Med Phys Fitness. 2017;57(12):1572–8. pmid:27406013

* View Article

* PubMed/NCBI

* Google Scholar

31. 31. Garden CA, Tawiah-Dodoo J, Blenkinsop GM, editors. A comparison of sprint technique between professional football players and elite sprinters. 40th International Society of Biomechanics in Sports Conference; 2022; Liverpool, UK.

Citation: Takai Y, Miyazaki T, Sugisaki N, Yoshimoto T, Mitsukawa N, Kobayashi K, et al. (2025) Spatiotemporal and kinetic characteristics during maximal sprint running in fast running soccer players. PLoS One 20(5): e0322216. https://doi.org/10.1371/journal.pone.0322216

About the Authors:

Yohei Takai

Roles: Conceptualization, Methodology, Writing – original draft

E-mail: [email protected]

Affiliation: National Institute of Fitness and Sports in Kanoya,

ORICD: https://orcid.org/0000-0002-0142-0993

Terumitsu Miyazaki

Roles: Formal analysis, Methodology

Affiliation: National Institute of Fitness and Sports in Kanoya,

Norihide Sugisaki

Contributed equally to this work with: Norihide Sugisaki, Hiroaki Kanehisa

Roles: Conceptualization, Methodology, Writing – review & editing

Affiliation: Center for Liberal Arts, Meiji Gakuin University,

Takaya Yoshimoto

Roles: Data curation, Investigation

Affiliation: Faculty of Welfare Society, The International University of Kagoshima,

Naotoshi Mitsukawa

Roles: Data curation, Investigation, Methodology, Writing – review & editing

Affiliation: Faculty of Human Sciences, Toyo Gakuen University,

Kai Kobayashi

Roles: Data curation, Investigation

Affiliation: Faculty of Information Sciences and Arts, Toyo University,

Hiroyasu Tsuchie

Roles: Data curation, Formal analysis, Investigation

Affiliation: Faculty of Law, Toyo University

Hiroaki Kanehisa

Contributed equally to this work with: Norihide Sugisaki, Hiroaki Kanehisa

Roles: Conceptualization, Writing – review & editing

Affiliation: National Institute of Fitness and Sports in Kanoya,

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

References

1. Faude O, Koch T, Meyer T. Straight sprinting is the most frequent action in goal situations in professional football. J Sports Sci. 2012;30(7):625–31. pmid:22394328

2. Kai T, Horio K, Aoki T, Takai Y. High-intensity running is one of the determinants for achieving score-box possession during soccer matches. Football Science. 2018;15:61–9.

3. Kai T, Hirai S, Anbe Y, Takai Y. A new approach to quantify angles and time of changes-of-direction during soccer matches. PLoS One. 2021;16(5):e0251292. pmid:33999931

4. Schulze E, Julian R, Meyer T. Exploring factors related to goal scoring opportunities in professional football. Sci Med Footb. 2022;6(2):181–8. pmid:35475738

5. Andrzejewski M, Chmura J, Pluta B, Strzelczyk R, Kasprzak A. Analysis of sprinting activities of professional soccer players. J Strength Cond Res. 2013;27(8):2134–40. pmid:23168374

6. Ettema G, McGhie D, Danielsen J, Sandbakk Ø, Haugen T. On the existence of step-to-step breakpoint transitions in accelerated sprinting. PLoS One. 2016;11(7):e0159701. pmid:27467387

7. Nagahara R, Kanehisa H, Fukunaga T. Ground reaction force across the transition during sprint acceleration. Scand J Med Sci Sports. 2020;30(3):450–61. pmid:31705835

8. Nagahara R, Matsubayashi T, Matsuo A, Zushi K. Kinematics of transition during human accelerated sprinting. Biol Open. 2014;3(8):689–99. pmid:24996923

9. von Lieres Und Wilkau HC, Irwin G, Bezodis NE, Simpson S, Bezodis IN. Phase analysis in maximal sprinting: an investigation of step-to-step technical changes between the initial acceleration, transition and maximal velocity phases. Sports Biomech. 2020;19(2):141–56. pmid:29972337

10. van Ingen Schenau GJ, de Koning JJ, de Groot G. Optimisation of sprinting performance in running, cycling and speed skating. Sports Med. 1994;17(4):259–75. pmid:8009139

11. Harris NK, Cronin JB, Hopkins WG, Hansen KT. Relationship between sprint times and the strength/power outputs of a machine squat jump. J Strength Cond Res. 2008;22(3):691–8. pmid:18438252

12. Little T, Williams AG. Specificity of acceleration, maximum speed, and agility in professional soccer players. J Strength Cond Res. 2005;19(1):76–8. pmid:15705049

13. Mendez-Villanueva A, Buchheit M, Kuitunen S, Douglas A, Peltola E, Bourdon P. Age-related differences in acceleration, maximum running speed, and repeated-sprint performance in young soccer players. J Sports Sci. 2011;29(5):477–84. pmid:21225488

14. Haugen TA, Tønnessen E, Seiler S. Anaerobic performance testing of professional soccer players 1995-2010. Int J Sports Physiol Perform. 2013;8(2):148–56. pmid:22868347

15. Colyer SL, Nagahara R, Takai Y, Salo AIT. How sprinters accelerate beyond the velocity plateau of soccer players: waveform analysis of ground reaction forces. Scand J Med Sci Sports. 2018;28(12):2527–35. pmid:30230037

16. Clark KP, Weyand PG. Are running speeds maximized with simple-spring stance mechanics?. J Appl Physiol (1985). 2014;117(6):604–15. pmid:25080925

17. Devismes M, Aeles J, Philips J, Vanwanseele B. Sprint force-velocity profiles in soccer players: impact of sex and playing level. Sports Biomech. 2021;20(8):947–57. pmid:31223073

18. McKay AKA, Stellingwerff T, Smith ES, Martin DT, Mujika I, Goosey-Tolfrey VL, et al. Defining training and performance caliber: a participant classification framework. Int J Sports Physiol Perform. 2022;17(2):317–31. pmid:34965513

19. Sugisaki N, Kobayashi K, Yoshimoto T, Mitsukawa N, Tsuchie H, Takai Y, et al. Influence of horizontal resistance loads on spatiotemporal and ground reaction force variables during maximal sprint acceleration. PLoS One. 2023;18(12):e0295758. pmid:38085716

20. Colyer SL, Nagahara R, Salo AIT. Kinetic demands of sprinting shift across the acceleration phase: Novel analysis of entire force waveforms. Scand J Med Sci Sports. 2018;28(7):1784–92. pmid:29630747

21. Nagahara R, Takai Y, Haramura M, Mizutani M, Matsuo A, Kanehisa H, et al. Age-related differences in spatiotemporal variables and ground reaction forces during sprinting in boys. Pediatr Exerc Sci. 2018;30(3):335–44. pmid:29478372

22. Hay GJ. The biomechanics of sports techniques. Edition t, editor. London: Prentice Hall International; 1994.

23. Murata M, Takai Y, Kanehisa H, Fukunaga T, Nagahara R. Spatiotemporal and kinetic determinants of sprint acceleration performance in soccer players. Sports (Basel). 2018;6(4):169. pmid:30544869

24. Wdowski MM, Gittoes MJR. First-stance phase force contributions to acceleration sprint performance in semi-professional soccer players. Eur J Sport Sci. 2020;20(3):366–74. pmid:31167614

25. Kalkhoven JT, Watsford ML. The relationship between mechanical stiffness and athletic performance markers in sub-elite footballers. J Sports Sci. 2018;36(9):1022–9. pmid:28697691

26. Douglas J, Pearson S, Ross A, McGuigan M. Reactive and eccentric strength contribute to stiffness regulation during maximum velocity sprinting in team sport athletes and highly trained sprinters. J Sports Sci. 2020;38(1):29–37. pmid:31631783

27. Moir GL, Brimmer SM, Snyder BW, Connaboy C, Lamont HS. Mechanical limitations to sprinting and biomechanical solutions: a constraints-led framework for the incorporation of resistance training to develop sprinting speed. Strength Cond J. 2018;40(1):47–67.

28. Kuitunen S, Komi PV, Kyröläinen H. Knee and ankle joint stiffness in sprint running. Med Sci Sports Exerc. 2002;34(1):166–73. pmid:11782663

29. Weyand PG, Sternlight DB, Bellizzi MJ, Wright S. Faster top running speeds are achieved with greater ground forces not more rapid leg movements. J Appl Physiol (1985). 2000;89(5):1991–9. pmid:11053354

30. Nagahara R, Zushi K. Development of maximal speed sprinting performance with changes in vertical, leg and joint stiffness. J Sports Med Phys Fitness. 2017;57(12):1572–8. pmid:27406013

31. Garden CA, Tawiah-Dodoo J, Blenkinsop GM, editors. A comparison of sprint technique between professional football players and elite sprinters. 40th International Society of Biomechanics in Sports Conference; 2022; Liverpool, UK.