Relationships between Sprint, Acceleration, and Deceleration Metrics with Training Load in Division I Collegiate Women’s Soccer Players

Soccer is an aerobically taxing, physically demanding intermittent team sport involving physical contact and different dynamic tasks that are performed at varying intensities (Stølen et al., 2005). In the USA, the National Collegiate Athletic Association (NCAA) Fall competitive soccer season consists of 16 weeks of competition, with a maximum of 20 matches during the 12-week in-season conference period (NCAA, 2020). This results in about two games per week with 48 hours between games for players competing at this level (McLean et al., 2012). Within a collegiate soccer match, athletes could cover distances of approximately 8–10 kilometers (km) with 70–190 efforts (or 16–17%) performed at higher speeds (>15 km/h) (McFadden et al., 2020). Nearly 250 short-duration anaerobic actions are completed as a result of field dimensions (105.2–109.7 meters [m] x 64.0–68.6 m) and the nature of the game, with the average number of efforts depending on the playing position (i.e., forward, midfielder, or defender) (Bangsbo et al., 2006; Radzimiński et al., 2021; Stølen et al., 2005). In addition to low-intensity aerobic movements, high-intensity anaerobic actions in soccer include jumping, kicking, tackling, agility, and sprinting. These variables relate to match performance; however, it is also important to recognize how they affect training performance.

To infer external workload demands of soccer players, sport scientists commonly utilize global positioning system (GPS) technology. GPS devices are widely used portable monitoring tools for assessing demands of practices and competitive matches in team-based sports (Coutts and Duffield, 2010; Ravé et al., 2020). A commonly reported variable of relevance extrapolated by GPS technology is player load (Ravé et al., 2020; Vanrenterghem et al., 2017). Player load, or external load, is a numerical value quantified by the amount of physical work performed over time by an athlete in the form of movement (Wallace et al., 2013). As GPS technology provides a collection of data during practice sessions, it is important to decipher which variables are most relevant for sport coaches to focus on. This way, coaches can manage these variables to optimize performance and achieve desired training goals (Ravé et al., 2020; Vanrenterghem et al., 2017). Along with external player load, the GPS device captures specific data for running-related variables. This information has received attention in field sports because running variables can be directly manipulated by coaches. A review by Ravé et al. (2020) documened that relevant GPS variables in monitoring player load in elite soccer include high-speed running distance, sprint distance, specific maximal speed, the number of accelerations (≥3 m/s²), and the number of decelerations ( ≤ 3 m/s²). The rationalization is because these metrics reflect training intensity and can be used to construct specific training sessions. Practitioners often focus on total and high-speed running distance covered as they are some of the most demanding and important qualities for athletes (Cardinale and Varley, 2017). Thus, highlighting these characteristics could prove valuable to sport scientists and soccer coaches at the collegiate level.

Sprinting is high-intensity anaerobic activity involving repeated impacts and great eccentric stress to the posterior tissues to overcome body mass inertia (Vanrenterghem et al., 2017) and can distinguish between successful match performance in soccer (Vescovi, 2012). Total distance covered while sprinting has been reported as approximately 1.5–2.5 km in elite players (Datson et al., 2014; Mohr et al., 2008), or about 1–10% of total distance in team sports (Girard et al., 2011). Accelerations and decelerations increase metabolic demand due to rapid force application and heavy eccentric loading of directional change efforts (Spiteri et al., 2015). Change-of-direction (COD) describes the preplanned ability to decelerate and re-accelerate in a different direction (Jones et al., 2008). Within competitive soccer matches, a high volume of approximately 700 directional changes can occur at maximal intensity (Bloomfield et al., 2007).

Greater external loads indicate higher intensities and metabolic cost experienced by athletes (Gaudino et al., 2013; Vanrenterghem et al., 2017). Due to the significant amount of high-intensity running actions performed in competitive soccer matches (Datson et al., 2014; Vescovi, 2012), it is plausible that sprint-related metrics may heavily influence a player’s workload demands (Thorpe et al., 2017). This includes sprint count, distance, volume, and speed, as well as the number of accelerations and decelerations in different speed zones. However, the current body of literature describing women’s soccer is limited, specifically at the collegiate level (Gentles et al., 2018). Further analysis is required to determine which sprint metrics may most affect training loads experienced by players.

The purpose of this study was to determine which sprint metrics correlated with player load in Division I women soccer athletes using GPS technology. A further aim was to determine which sprint variables were the greatest predictor of player load. It was hypothesized that the correlation analysis would demonstrate significant relationships between all sprint metrics and player load. A further hypothesis was total sprint distance and the number of accelerations and decelerations completed by players would predict player load the most of all sprint-related variables.

Descriptive data for all variables is displayed Table 1. Significant relationships were found between player load and sprint count (large), maximum top speed (moderate), sprint distance (large), sprint volume (large), the number of decelerations at –1, –2, and –3 m/s²(all moderate), and the number of accelerations at 1, 2, 3, and 5 m/s²(all moderate) (Table 2). There were no significant relationships found between player load and average top speed, the number of decelerations at –4 and –5 m/s², or the number of accelerations at 4 m/s². The regression analyses indicated that sprint distance (p < 0.001, r²= 0.741) predicted player load with 72.5% explained variance. When the number of decelerations at –1 m/s²was added to the equation, explained variance increased to 85.1% (p = 0.001, r²= 0.867).

This study investigated the relationship between sprint-related variables with player load recorded during training sessions in collegiate women field soccer players. The data indicated relationships between player load and sprint count, maximum top speed, sprint distance, sprint volume, the number of decelerations at –1, –2, and –3 m/s², and the number of accelerations at 1, 2, 3, and 5 m/s². This supports the research hypotheses as these sprint metrics demonstrated significant and positive relationships to player load. The regression analysis revealed that sprint distance predicted player load the most, along with the number of decelerations at –1 m/s². These data have important implications for coaches and sport scientists working with collegiate women soccer players.

There are very few studies that have examined maximal sprint speeds at the collegiate or professional level for women’s soccer. High-speed running in soccer allows players to remain involved in play through their ability to keep up with game transitions. Assessing sprint variables in soccer players is meaningful as the ability to achieve faster running speeds can lead to significant opportunities during match play, such as creating space and scoring goals (Hewitt et al., 2014). In this study, the average maximal speed of all training sessions for the sample was approximately 6.74 m/s. The results indicated maximal sprint top speed had a large correlation to player load, with a positive relationship between the two variables. These data suggest that during training, athletes who achieved faster sprinting speeds experienced higher external loads. Interestingly, average top speed did not have a significant relationship with player load, suggesting that the peak speed attained by players had a greater impact on the external load experienced by players. Indeed, maximal sprinting speed requires high ground reaction forces (Čoh et al., 2018), so it could be expected that faster speeds place greater stress on a soccer athlete. Accordingly, as demonstrated by the results of the current study, when planning drills for practice sessions, sport coaches must consider that incorporating drills which require running at faster speeds will lead to increases in player load.

It is a challenge to compare sprint distance means of previous investigations to the present study due to the use of different GPS technologies, which each defines a sprint as something different. As noted previously, the Titan GPS sensors used in the current study defined a sprint as speeds at or above 4.76 m/s. Additionally, previous investigations have predominantly assessed professional soccer athletes during competition, rather than during training. An investigation of Danish national women soccer teams by Panduro et al. (2021), for instance, calculated an average of 676 m of very high-speed running (specified as speeds faster than 5 m/s), and approximately 245 m covered while sprinting (specified as speeds faster than 6.96 m/s) in field players during competitive games. Another study of Australian national women soccer players conducted by Hewitt et al. (2014) measured an average of 338 m of sprint distance (classified as speeds faster than 5.28 m/s), and approximately 2407 m of high-intensity running (classified as 3.33–5.27 m/s) during matches. These findings from Panduro et al. (2021) and Hewitt et al. (2014), as well as the results from this study, demonstrate why it is important for sport scientists and soccer coaches to recognize the effects of the amount, volume, and distance of sprints as these variables will influence external load demands.

Limited studies have examined sprint count, distance, and volume, particularly during practice sessions and in collegiate women soccer players. In the present study, the analysis of the collective team indicated very large correlations between player load and sprint count, sprint distance, and sprint volume. Sprint distance also predicted player load, with 72.5% explained variance. The average sprint count was approximately 14 sprints per session, average sprint distance was about 331 m per session, and average sprint volume was approximately 344 total sprints. Based on the findings of this investigation, a greater number, distance, and volume of sprints led to higher player load values, with sprint distance being particulalrly impactful. As noted previously, maximal sprinting requires the generation of high external forces (Čoh et al., 2018). In line with the results for maximal sprint speed, a greater number of sprints and distance covered while sprinting will elevate the load experience by the soccer athlete. Thus, training sessions that involve more sprint efforts, and sprints over longer distances, increased the level of difficulty of the session. The distance covered sprinting maximally within training drills could be adjusted up or down to manipulate the intensity of training sessions. Future research could track how training drills manipulated according to this process could affect fitness adaptations experienced by collegiate women soccer players.

The results of this study indicated that the number of decelerations and accelerations at slower speeds seemed to have a greater effect on player load in collegiate women’s soccer athletes. Specifically, the number of decelerations at –1, –2, and –3 m/s²and the number of accelerations at 1, 2, 3, and 5 m/s²had large correlations with player load. The number of decelerations at –1 m/s²also predicted player load; when added to the regression equation with sprint distance, the explained variance was 85.1%. These results were anticipated as pace and directional changes are physically demanding, even at submaximal speeds (Douchet et al., 2021). Accelerations require an increase in momentum to propel the body in a new direction (Spiteri et al., 2015). Decelerations involve braking mechanisms that induce stressful loading to the posterior tissues (Papla et al., 2020). In this investigation, athletes executed over 500 accelerations and decelerations in the different speed zones per training session. Due to the high number of acceleration and deceleration efforts in soccer, athletes require exposure to these actions to prepare them for the demands of competition. Considering the significant number of pace changes completed by soccer athletes (Bloomfield et al., 2007), it is not surprising that these actions, even at relatively lower intensities, had a large impact on player load.

There was a greater total number of decelerations and accelerations at or below ±2 m/s²in comparison to decelerations and accelerations at or above ±3 m/s². These submaximal acceleration and deceleration efforts led to increases in player load likely due to the high volume of repetitions during training (approximately 246 decelerations below –2 m/s²and 253 accelerations below 2 m/s²). Douchet et al. (2021) found that an increased number of accelerations and decelerations in training led to an increase in perceptual exercise intensity and fatigue in elite professional women soccer players. The authors attributed this finding to produced muscle damage and soreness as a result of these metabolically demanding tasks. Considering the results from this study and those from Douchet et al. (2021), it would seem that drills involving a greater number of pace and COD actions could increase the intensity of training sessions for collegiate women’s soccer players. As an example, if a coach wished to increase the stress of a training session, designing training drills that required more direction changes should achieve this. This could include 2 x 3 or 4 x 4 grids, various cone drills that incorporate dribbling and other on-ball skills, and different variations of small-sided games with reduced playing field dimensions and fewer players. Using training drills with reduced COD demands should decrease the stress experienced. Any training adaptations experienced by collegiate women soccer players according to training drill manipulation relative to player load could be investigated in future studies.

The number of decelerations at –4 and –5 m/s²and accelerations at 4 m/s²did not demonstrate significant correlations to player load. These results could have been influenced by some of the inherent limitations to GPS devices (Coutts and Duffield, 2010). Accelerations and decelerations may be harder to record via GPS equipment if they are faster and shorter in duration (Windt et al., 2020). Nonetheless, the deceleration and acceleration data do reflect the demands of these actions relative to player load in collegiate women soccer players.

There are certain limitations of the study that should be considered. The sample size was relatively small (N = 18), and only one NCAA Division I collegiate women’s soccer team was assessed for this investigation. The number (N = 19) and type of practice sessions analyzed may have presented different results compared to a greater number of sessions or competitive match data. Workload demands of players vary depending on the playing position, thus, specific drills may affect certain positions more than others. Relationships between player load and sprint metrics specific to positions in collegiate women’s soccer players could be studied in future research. GPS technology may have shortcomings in the assessment of short, high-intensity actions such as accelerations and decelerations (Coutts and Duffield, 2010; Ravé et al., 2020). Nonetheless, GPS devices with higher sampling rates such as those used in this investigation have demonstrated greater reliability and validity (Rampinini et al., 2015). Furthermore, the use of GPS data in high-level soccer matches is commonplace, and thus has application in collegiate women’s soccer.

In conclusion, the results showed relationships between player load and sprint count, maximum top speed, sprint distance, sprint volume, the number of decelerations at –1, –2 and –3 m/s², and the number of accelerations at 1, 2, 3, and 5 m/s²in collegiate women soccer players. Sprint distance and the number of decelerations at –1 m/s²predicted player load. Understanding the impact of these variables on athlete workloads could allow coaches to design training drills to manipulate the difficulty of the session. This could also lead to improved application of drills and training modalities to better enhance performance and fitness of players. For example, training games could be altered via the number of players (i.e., 4 vs. 4 and 6 vs. 6), adapted rules (i.e., one touch, two touches, the target number of passes, etc.), and modified pitch dimensions to either increase or decrease the sprint distance and COD actions required. This could allow coaches to adjust player load as needed for collegiate women’s soccer players over the course of a season.