Effect of Different Nonstarter Compensatory Strategies on Training Load in Female Soccer Players: A Pilot Study

Abstract

Background:

In soccer, the day of the week with the highest external load is match day (MD), with starters (>60 minutes per match) showing higher levels of physical fitness and seasonal high-intensity loading. It is necessary, therefore, to determine training strategies to reduce the differences between starters and nonstarters. The aim of this study was to analyze and compare the external load of different training compensatory strategies with match external load in female nonstarters.

Hypothesis:

A strategy combining small-sided games (SSG) and running-based drills (RBD) would reproduce match demands, with RBD leading to greater high-intensity running and SSG leading to a greater number of accelerations and decelerations.

Study Design:

Descriptive and comparative study.

Level of Evidence:

Level 4.

Methods:

The training and match external load of 14 female players belonging to the same reserve squad of a Spanish First Division Club (Liga Reto Iberdrola) was registered. In the first session after the match (MD+1), nonstarters (<60 minutes in the match) performed 1 of 3 different compensatory strategies: RBD, SSG, or a mixed intervention combining the previous strategies (RBD+SSG). Starters carried out a recovery session.

Results:

A marked difference in load was observed between the compensatory training strategies and MD. In comparison with MD, RBD showed greater high-intensity and sprint distances and lower acceleration, SSG showed less high-intensity running and sprint distances, lower peak velocity, and greater acceleration, and RBD+SSG registered lower accelerations. In addition, nonstarters covered greater high-intensity running and sprint distances in RBD and achieved higher accelerations in SSG.

Conclusion:

RBD and SSG compensatory strategies could be recommended to nonstarter female soccer players in MD+1 to compensate for match external load deficits.

Clinical Relevance:

This study provides comprehensive information on the compensatory exercises of female soccer players, which can be useful for strength and conditioning coaches when developing recovery strategies during a microcycle.

Women’s soccer has increased in popularity at all levels, ³⁴ and is experiencing incredible growth in terms of media impact, competition, and players’ physical development. ³⁰ Proof of this is that, in the 2019 Women’s World Cup, distances covered at high intensity (HID, 19-23 km/h) increased by 15% and distance covered in sprints (SPRD, >25 km/h) increased by approximately 29%, ⁶ compared with the 2015 Women’s World Cup. To manage these higher match external loads, it is necessary to optimize training periodization through the adjustment of volume and intensity within the training microcycle, ⁴² and to apply training methods that prepare players for the match external load. This becomes even more important with nonstarters.^32,41

Due to growth in terms of competitiveness, physical development, and minutes played on match days (MDs), starters covered more total distance (22%), HID (47%), and SPRD (74%) than nonstarters. They also had a greater seasonal load and reported a higher perceived load (29%).^2,26 The main characteristic of starters versus nonstarters is greater participation (ie, minutes) in competition, allowing them to accumulate higher physical and physiological loads during the microcycle. Notably, it has been demonstrated that matchplay provides an important stimulus for adaptation. Jajtner et al ²⁴ found that female starters in the National Collegiate Female Division I had improved speed after an 8-week line drill test, with no changes in the nonstarters. Countermovement jump performance has also been shown to improve in starters but not in nonstarters. ³⁵ This points to a need for coaches and practitioners to manage player workloads to ensure both starters and nonstarters are adequately prepared for matchplay. ¹³ As such, it may be necessary to include compensatory training strategies for nonstarters to maintain or improve their training status. ³² To achieve this aim, several training strategies (eg, high-intensity interval, small-sided games (SSG), or plyometric training) have been applied in female soccer populations.^16,37,39 However, the impact of these strategies on training demands, differentiating between starters and nonstarters, has not been considered in previous studies.

Different compensatory training strategies may be used to minimize decrements in players’ physical performance. Ade et al ¹ observed that soccer players covered greater distances in high-intensity running and sprinting in running-based drills (RBD) compared with SSG drills, although more accelerations and decelerations were registered during SSG. When nonstarters were supplemented with SSG in the first session after a match (MD+1), greater total distance covered, higher average metabolic power, and accelerations and decelerations were recorded, but high-intensity and sprint qualities were not developed. ³²

The aim of this study was first to analyze and compare the external load of different training compensatory strategies (ie, RBD, SSG, and a mixed intervention combining RBD+SSG) on match external load in female nonstarters. Second, the microcycle load between nonstarters and starters was compared with consideration to the compensatory strategies applied. Based on previous studies,^1,32 we hypothesized that a strategy combining SSG and RBD would reproduce match demands, since RBD leads to greater HID and SSG leads to a greater number of accelerations and decelerations.

Methods

Subjects

A total of 14 female soccer players (age, 21.7 ± 1.7 years; height, 164.3 ± 5.1 cm; body mass, 55.8 ± 6.9 kg; and body mass index, 20.7 ± 1.6 kg/m²) participated in this study. Data were recorded during the 2020 to 2021 competitive midseason period, and all participants belonged to the same reserve team of a Spanish First Division Club (Liga Reto Iberdrola). Goalkeepers were excluded from the analysis due to their specific role. Players who had suffered an injury in the previous 2 months and who did not complete all of the intervention sessions were not included in the analysis. Before signing informed consent forms or beginning the study, participants were informed of the study’s objectives, risks, and benefits. The study was conducted according to the requirements of the Declaration of Helsinki and was approved by the ethics committee of the University of Leon (code: 005-2021).

Design

This descriptive and comparative pilot study compared the training load generated by 3 different interventions (RBD, SSG, and RBD + SSG) in nonstarters. During MD+1, the female soccer players were assigned to a Recovery Group (starters) or a Compensatory Group (nonstarters) according to the minutes played in the previous match (Recovery Group ≥60 minutes; Compensatory Group <60 minutes). ³² The intervention consisted of supplementing nonstarters with 3 different training strategies (RBD, SSG, or RBD+SSG), each performed independently on MD+1 for 3 consecutive weeks. Each week of the intervention period was composed of 4 training sessions (ie, MD+1, the first session after the previous match without a recovery day; MD-3; MD-2; and MD-1, with 3, 2, and 1 session before the next match, respectively), and an official MD session. The usual distribution of the week during the competitive season in the 3 previous months was as follows: recovery or compensatory, endurance, tactical and activation in MD+1, MD-3, and MD-1, respectivel. ⁸ During the MD+1 intervention period, the starters performed a recovery session (MD+1) consisting of a 15-min technical drill followed by a 4 versus 4 SSG on a surface of 10 × 15 m for 8 minutes, finishing with regeneration exercises (eg, foam roller, mobility). Measures of external load and the session rating of perceived exertion (sRPE) were collected in each session, as well as a wellness questionnaire before the MD-3 session. All training sessions were conducted on the same playing surface (third-generation artificial turf) at the same time (7:30 pm). Matches were played on 3 pitches with similar dimensions (100 × 64 m) and artificial surfaces.

Procedures

Compensatory Strategies

In addition to the normal training, nonstarters completed 1 of the following interventions each consecutive week: RBD, in which players performed a speed endurance drill consisting of 2 × 6 × 20 seconds all out sprints with 90 seconds of active recovery, had 5 minutes of recovery, and then performed a repeated sprint drill consisting of 2 × 5 × 25-m sprints followed by shooting at goal with 25 seconds of passive recovery. In SSG, players performed a 4 versus 4 SSG (25 × 20 m; individual interaction space, 62.5 m²) consisting of 3 bouts of 4 minutes separated by 90 seconds of passive recovery, and a 4 versus 4 with goalkeepers (20 × 15 m) consisting of 2 bouts of 8 minutes and 120 seconds passive recovery. In RBD+SSG (mixed intervention), players performed a combination of parts of both strategies: first, a repeated sprint drill consisting of 2 × 5 × 25 m with 25 seconds of recovery between repetitions and 5 minutes between sets, and second, after 5 minutes of recovery, the same small game as in the SSG strategy (4 vs 4 SSG [25 × 20 m; individual interaction space, 62.5 m²] consisting of 3 bouts of 4 minutes separated by 90 seconds of passive recovery].

External Load Quantification

External load was recorded individually for each player using an 18-Hz Global Positioning System (GPS) with an integrated 100-Hz triaxial accelerometer (WIMU PRO, RealTrack Systems). This technology has been used previously in soccer research on activity-demand profiles,^21,36 and shows high levels of validity and reliability (relative technical error of measurement [%TEM], 1.47). ⁵ The GPS units (70 g; 81 × 45 × 16 mm) were activated 15 minutes before the start of each session in accordance with the manufacturer’s recommendations, and were harnessed in a tight-fitting vest worn by the female soccer players during the experimental study. To avoid interunit variability, each player wore their assigned unit in all training sessions and matches. ⁵ After each training session or match, GPS data were downloaded onto a personal computer using the specific software package (WIMU SPRO) and exported for further analysis. Absolute (meters, TD) and relative (meter per minute, RD) values for total distance, high-intensity distance (HID ≥19.0 km/h), sprint distance (SPR ≥23.0 km/h), high-intensity acceleration (ACC >3 m/s²), high-intensity deceleration (DCC >-3.0 m/s²), and peak velocity (PV) were recorded. These are similar ranges to those used in previous studies with female soccer players. ⁷ The average number of satellites registering data during the measurements was 10.1 ± 1.0 and horizontal dilution of precision was 0.96.

Internal Load and Wellness Quantification

A 0 to 10 category ratio scale was used to register players’ perceived effort 30 minutes after each training session. ¹⁸ Furthermore, each individual sRPE value was multiplied by the training session duration to quantify players’ training load. ²⁰ All participants were familiar with the category ratio scale as they use it regularly in their training sessions and matches. In addition, the players completed a wellness questionnaire in the morning on MD-3. The items of the questionnaire included sleep quality, stress, fatigue, and muscle soreness on a 7-point Likert scale.^20,40 Players rated on the scale how much they agree (1, strongly agree) or disagree (7, strongly disagree). The sum of the 4 ratings was used to calculate Hooper’s index.^20,40

Statistical Analysis

Results are presented as means and standard deviations. Normality was verified using Shapiro-Wilk’s test. A 1-way analysis of variance (ANOVA) was conducted to compare all studied variables among the training strategies (RBD, SSG, and RBD+SSG) and MD. Pairwise comparison was performed using Bonferroni’s post hoc test. An independent t test was used to analyze the external load differences among starters and nonstarters in each training microcycle and training session. The standardized difference or effect size (ES, 90% CI) in the selected variables was calculated using Cohen’s d with values of <0.2 (trivial), between ≥0.2 and 0.49 (small), ≥0.5 and <0.79 (medium), and ≥0.8 (large). ¹² Significance level was set at P < 0.05. Statistical analysis was conducted using SPSS Version 25.0.

Results

Table 1 presents the results of the external load of each intervention and MD. In the SSG and SSG+RBD interventions, players covered significantly (P < 0.05) less RD than MD (ES = 3.9 and 4.6, respectively). In RBD, players covered significantly (P < 0.05) more absolute and relative HID and SPRD than MD (ES = 3.9 and 2.0, respectively). Less (P < 0.05) HID and SPRD was covered than MD in the SSG (ES = 4.1 and 1.6, respectively). However, in RBD+SSG, similar HID and higher (P < 0.05) SPRD were covered than on MD (ES, 1.1). Only the SSG intervention reached a similar ACC to the MD (ES = 1.1). In RBD and RBD+SSG players covered significantly (P < 0.05) more HID and SPRD and reached a higher PV than in SSG.

Table 1.

Comparison of external load of each compensatory strategy and MD

	MD	RBD	SSG	RBD+SSG
TD, m	8257 ± 1229	5463 ± 149*	4385 ± 300*	3614 ± 277* †
RD, m/min	94.6 ± 8.0	90.9 ± 4	69.4 ± 5.0* †	62.9 ± 6.0* †
HID, m	281.3 ± 99.6	964.5 ± 233.3*	4.9 ± 5.7* †	202.7 ± 28.0 † ‡
HID, m/min	3.5 ± 1.5	16.1 ± 3.9*	0.1 ± 0.1* †	3.5 ± 0.5 † ‡
SPRD, m	45.8 ± 41.6	147.1 ± 59.4*	0.0 ± 0.0 † *	112.2 ± 57.2* ‡
SPRD, m/min	0.6 ± 0.5	2.5 ± 0.9*	0.0 ± 0.0 †	1.9 ± 0.9* ‡
ACC HI, No.	23.9 ± 5.6	12.0 ± 3.8*	17.0 ± 6.4	16.1 ± 3.7*
DCC HI, No.	44.9 ± 10.9	10.7 ± 4.2*	28.0 ± 8.9*	20.5 ± 6.2*
ACC HI, No./min	0.3 ± 0.1	0.2 ± 0.1	0.3 ± 0.1	0.3 ± 0.1
DCC HI, No./min	0.5 ± 0.1	0.2 ± 0.1*	0.4 ± 0.1 †	0.4 ± 0.1 †
PV, km/h	24.6 ± 1.9	25.9 ± 1.5	19.5 ± 1.5* †	24.4 ± 1.8 ‡
sRPE, AU	555.3 ± 175.3	495.0 ± 57.4	261.0 ± 31.8* †	330.8 ± 31.5*

ACC HI, high-intensity accelerations (>3 m/s²); AU, arbitrary units; DCC HI, high-intensity decelerations (>-3.0 m/s²); HID, high-intensity distance (>19.0 km/h); MD, match day; PV, peak velocity; RBD, running-based drills; RD, relative distance; SPRD, sprint distance (>23.0 km/h); sRPE, session rating of perceived exertion; SSG, small-sided game; TD, total distance.

^*Difference from MD; ^†difference from RBD; ^‡difference from SSG; P < 0.05.

When RBD was performed, nonstarters covered significantly more HID and SPRD and less (P < 0.05) total ACC and DCC than starters (Table 2). Nonstarters covered a lesser (P < 0.05) TD and ACC than starters when SSG was performed. Similarly, less (P < 0.05) TD was covered by nonstarters in the RBD+SSG.

Table 2.

Comparison of accumulated external loads during microcycle between starter and nonstarter female soccer players

	RBD			SSG			RBD+SSG
	S	NS	ES	S	NS	ES	S	NS	ES
TD, m	25,603 ± 2995	22,513 ± 1752	1.5	25,308 ± 1988	20,689 ± 1620*	2.6	25,986 ± 1275	21,312 ± 2861*	2.2
RD, m/min	65.7 ± 5.3	69.6 ± 4.4	0.8	66.6 ± 5.1	66.7 ± 6.4	0.0	69.3 ± 4.9	66.8 ± 8.3	0.4
HID, m	730.5 ± 281.2	1165.3 ± 521.5*	1.1	545.3 ± 181.5	365.1 ± 209.4	0.9	705.1 ± 272.4	624.5 ± 125.0	0.4
HID, m/min	1.9 ± 0.7	3.6 ± 1.6*	1.4	1.5 ± 0.5	1.2 ± 0.7	0.5	2.1 ± 0.7	1.9 ± 0.3	0.4
SPRD, m	86.8 ± 55.3	170.9 ± 54.7*	1.6	89.4 ± 73.9	55.8 ± 50.5	0.5	137.7 ± 92.5	179.4 ± 96.4	0.5
SPRD, m/min	0.35 ± 0.2	0.7 ± 0.4	1.1	0.2 ± 0.2	0.2 ± 0.2	0.0	0.3 ± 0.2	0.6 ± 0.3	1.0
ACC HI, No.	91.3 ± 24.3	58.6 ± 8.7*	1.9	104.0 ± 22.3	80.2 ± 8.9*	1.4	97.5 ± 22.4	83.6 ± 31.5	0.5
DCC HI, No.	0.2 ± 0.1	0.2 ± 0.0	0.0	0.3 ± 0.1	0.3 ± 0.0	0.0	0.3 ± 0.1	0.3 ± 0.1	0.0
ACC HI, No./min	136.0 ± 21.2	94.8 ± 25.8*	1.8	146.8 ± 19.9	123.4 ± 43.0	0.7	157.1 ± 21.8	116.0 ± 46.2	1.2
DCC HI, No./min	0.3 ± 0.1	0.3 ± 0.1	0	0.4 ± 0.1	0.4 ± 0.2	0.0	0.4 ± 0.1	0.4 ± 0.1	0.0
PV, km/ h-1\\	21.9 ± 1.0	22.9 ± 0.8	1.1	22.4 ± 0.8	23.0 ± 1.9	0.4	22.6 ± 1.5	22.5 ± 1.2	0.1
sRPE, AU	1671.1 ± 388.4	1587.4 ± 193.1	0.3	1594.5 ± 369.1	1471.4 ± 254.1	0.4	1708.4 ± 429.0	1518.5 ± 195.1	0.6

ACC HI, high-intensity accelerations (>3 m/s²); AU, arbitrary units DCC HI, high-intensity decelerations (>-3.0 m/s²); ES, effect size Cohen’s d; HID, high-intensity distance (>19.0 km/h); NS, nonstarters; PV, peak velocity; RBD, running-based drills; RD, relative distance; S, starters; SPRD, sprint distance (>23.0 km/h); sRPE, session rating of perceived exertion; SSG, small-sided games; TD, total distance.

^*Difference from starters; P < 0.05.

Nonstarters who performed MD+1 training compensatory strategies showed a higher (P < 0.05) TD (ES = 9.6, 6.8, and 4.2 for RBD, SSGs, and RBD+SSG, respectively) and PV (ES = 4.7, 1.6, and 1.9 for RBD, SSGs, and RBD+SSG, respectively) relative to match load than starters in all training strategies (Figure 1). In addition, nonstarters performed greater (P < 0.05) HID and SPRD in RBD (ES = 1.5 and 5.3, respectively), DCC in SSG (ES = 1.7), and HID, SPRD, and ACC in RBD+SSG (ES = 1.9, 2.8, and 2.1 respectively). No significant differences were found between starters and nonstarters in the perception of wellness in any of the 3 interventions (~14 arbitrary units [AU]).

Figure 1.

Physical demands relative to match in RBD (a), SSG (b), and RBD+SSG (c) strategies in MD+1 session. ACC HI, high-intensity accelerations (>3 m/s2); DCC HI, high-intensity decelerations (>-3.0 m/s2); HID, high-intensity distance (>19.0 km/h); MD, match day; PV, peak velocity; RBD, running-based drills; SPRD, sprint distance (>23.0 km/h); SSG, small-sided games; TD, total distance. *Differences between starters and nonstarters; P < 0.05.

Discussion

Our results showed that players were exposed to higher total distance, decelerations, and sRPE in matches than in training sessions. However, RBD allowed players to reach greater high-intensity and sprint distances, boosting the weekly accumulation of these variables. Similarly, SSG involved a higher weekly accumulation of accelerations.

Matchplay represents the highest stimulus for professional soccer players in terms of external and internal load, ³⁵ with starters covering more high-intensity and sprint distance than nonstarters,^2,14 which may affect their physical fitness (ie, adaptations in skeletal muscle). ¹⁷ Therefore, compensatory strategies should be applied to nonstarters to maintain or increase their physical fitness level. ³² Our results showed that female players covered significantly more HID and SPRD during RBD compared with MD and the other training strategies (SSG and RBD+SSG). In speed endurance production (1 vs 1, 8 bouts of 30 seconds with 120 seconds of recovery) and maintenance (2 vs 2, 8 bouts of 60 seconds with 60 seconds of recovery) running drills, players covered more distance in high-intensity running parameters (ie, distance at 19.9-25.2 km/h) compared with the respective SSG. ¹ In addition, the RBD+SSG strategy meant that female soccer players covered significantly more HID than in SSG alone. This may have led to sRPE in RBD being significantly higher than in the SSG strategy. The greater high-intensity and sprint distances covered make RBD a useful tool for high-intensity and sprint training. Similarly, Lupo et al ²⁷ and Arslan et al ⁴ reported that a running-based training program could be more effective in improving young soccer players’ sprint performance (ie, 20 m) and speed-based conditioning than soccer-specific drills. These results are in agreement with a recent meta-analysis that concluded that running-based high-intensity interval training was superior to SSG-based interventions for sprint performance in soccer players. ¹⁰

Sprint ability is required by female soccer players to gain an advantage in attacking and defensive situations. ¹⁵ In addition, exposure to maximal velocity running reduces the risk of injury to players. ²⁹ Consequently, players require regular exposure to periods of sprinting during training environments. ¹⁹ RBD and RBD+SSG resulted in similar PV to that of MD (~25 km/h), as also reported for youth women soccer players (23-26 km/h). ⁴³ Therefore, these interventions stimulate match peak speed. In addition, RBD presented significant positive effects for linear sprinting and change-of-direction performance compared with SSG ¹⁰ ; this training intervention could be used on MD+1 as a speed exposure for nonstarters and may reduce the differences between starters and nonstarters, aiding the maintenance of squad physical fitness.

Our results also showed that the SSG compensatory strategy did not expose nonstarters to high-intensity actions (HID or SPRD). Köklü et al ²⁵ reported that, when substituting 60 seconds of SSG with running drills (15 + 15 seconds), players covered significantly greater distances in high-intensity speed zones (>14.4 km/h), regardless of the number of players (3 vs 3 and 4 vs 4). This may be explained by the characteristics of SSG, which is played in individual interaction space of 62.5 m². Literature has shown that reducing the playing space also reduces the ability to reach high speeds. ⁹ Another possibility is that high-intensity actions during SSG are limited by a ceiling effect: players with high physical fitness experience a lower external load, which in turn could influence the distance covered at high-intensity during SSG. ²³ The PV reached in the SSG intervention was significantly lower than that reached in RBD, RBD+SSG, and MD. Implementing SSG with larger spaces might allow players to reach a higher speed and cover greater HID.^11,22

Players’ ability to accelerate may help optimize onfield performance and prevent injury. ³¹ The number of accelerations performed by female players in this study was lower than that reported by previous studies.^31,33 This might be due to the different levels of the players (elite vs reserve team) or the acceleration threshold considered (>2.26 m/s² or >2 m/s² vs >3 m/s²). Only the SSG intervention reproduced the number of accelerations that the players experience on MD. These results are in agreement with those of Ade et al, ¹ who reported that a greater number of accelerations are performed during sprint endurance production or maintenance training via SSG than in the respective running drills. Accelerating is more energetically demanding than constant-velocity movement. ³⁸ Therefore, despite the greater distance covered at high intensity and sprinting in RBD and RBD+SSG, coaches and physical trainers should include acceleration in the prescription and distribution of training tasks.

Previous studies reported that microcycle external load is conditioned by the number of matches or moment of the season,^3,28 but, to our knowledge, this is the first study to analyze the microcycle load between starters and nonstarters according to a compensatory strategy applied to female soccer players. Anderson et al ² did not find differences in the season-long external load between starters and nonstarters, but starters displayed lower external load than nonstarters in training sessions and covered more distance in high-intensity zones. Our results show that the compensatory strategy used with nonstarters may influence weekly training load. The HID and SPRD in the RBD compensatory session was greater than match demands and weekly training sessions by the starters, causing a higher external load (ie, HID and SPRD) in the nonstarters’ microcycle. Since training load predicts inseason injury and illness risk in female youth soccer players, ⁴⁴ coaches and practitioners need to take the strategy used into account.

This study has many limitations that should be considered by practitioners. Starters could experience a variability of up to 30% in the demands of the match but, in compensatory training, the load of the tasks could be better controlled, and therefore the intersession variability reduced. In addition, the study was carried out on a sample with specific characteristics (elite reserve team female soccer players), so care should be taken when applying it to players with other characteristics (ie, age and level) or genders (ie, male soccer players). Furthermore, although the intervention length was acceptable, a larger number of intervention sessions may be necessary to confirm the present results. Finally, no randomization in compensatory training strategies was established, as the application of each strategy depended on whether the female soccer player participated as a starter or nonstarter in the previous match.

Conclusion

As the match constitutes the main external load of the microcycle, nonstarters and starters show a different total microcycle load. Therefore, it is necessary to implement strategies to ensure nonstarters are adequately prepared for competition. The reduction of the differences in nonstarter external loads depend on the compensatory strategy employed. The RBD+SSG intervention exposed the players to matchlike demands. Compensatory strategies used in the MD+1 session in nonstarters may influence the accumulated load during the microcycle. Without strategies for load compensation, nonstarters may present lower fitness levels and have a greater risk of injury when they compete. Future studies could analyze different strategies, use SSG with a different format (ie, spatial, temporal, or different number of players), or implement SSG with larger spaces, which might allow players to reach a higher speed and achieve greater HID.