Assessment of Injury Risk in Professional Soccer Players: A Long-Term Study

Abstract

The aim of this study was to record and interpret external load parameters in professional soccer players in competitive microcycles with one or two matches per week, while investigating the interaction between training load and non-contact musculoskeletal injuries during training and matches. Musculoskeletal injuries in athletes are closely associated with workload fluctuations, particularly the acute:chronic workload ratio (ACWR) over preceding weeks. This study analyzed the physical workload of 40 high-level soccer players competing in the Greek championship across two seasons, encompassing 50 competitive microcycles, 60 official matches, and 300 training sessions. GPS-based assessments recorded total distance (TD), running speeds (15–20 km/h, 20–25 km/h, 25–30 km/h), accelerations (>2.5 m/s²), and decelerations (>2.5 m/s²). An independent sample t-test was conducted to compare injured and uninjured athletes, with statistical significance set at p < 0.05. Results showed that 20 injured athletes frequently exceeded the ACWR threshold (>1.3) compared to uninjured players. Analysis of the four weeks preceding the injury revealed that increased workload in high-intensity categories significantly contributed to non-contact injuries. Specifically, high running speeds of 15–20 km/h (p = 0.015), 20–25 km/h (p = 0.045) and >25 km/h (p = 0.008), as well as accelerations (p = 0.010), were linked to a higher risk of injury. The three-week ACWR model indicated statistically significant differences in the ACWR index for total distance (p = 0.033), runs at 15–20 km/h (p = 0.007), >25 km/h (p = 0.004), accelerations (p = 0.009), and decelerations (p = 0.013). In the two-week model, significant differences were found in runs at 15–20 km/h (p = 0.008) and >25 km/h (p = 0.012). In the final week, significant differences were observed in runs at 15–20 km/h (p = 0.015), >25 km/h (p = 0.016), and accelerations (p = 0.049). Additionally, running speeds of 25–30 km/h (p values between 0.004 and 0.016) played a key role in injury risk when limits were exceeded across all weekly blocks. These findings highlight the importance of monitoring ACWR to prevent injuries, particularly by managing high-intensity workload fluctuations in elite athletes.

1. Introduction

Soccer is the most widely played sport in the world [1]. In recent years, the physical demands of training and matches in soccer have increased significantly, leading to a higher risk of injury. Most of these injuries occur in the lower body, making injury prevention a critical focus for players and coaches [2]. These injuries cause immediate tissue damage that may be an indication of a predisposition to re-injury [3], while higher loads enhance the appearance of asymmetries in joints and muscles as they negatively affect stability, mobility, and neuromuscular control, also causing simultaneous load shifts, sometimes even to the opposite extremities [4]. The factors that can cause injury in soccer, according to the theoretical verification of Inklaar [5], are distinguished into endogenous and exogenous factors related to environmental conditions, such as direct physical contact with an opposing soccer player [2], training loads (wrong or overtraining) [6,7], protective equipment [3] and court surfaces [8]. Training loads, especially their variations, are an important part of training programming, aimed both at developing and improving performance and at preventing injury. In particular, the daily recording of training load in soccer is an integral part of the training planning of professional teams, as it contributes both to the development of progressivity in terms of training load and to the creation of conditions to avoid excessive stress that soccer causes in players who may be prone to injury. Total distance, high-intensity runs, sprints, accelerations, and decelerations are key indicators of external load.

In this context, we should focus on the most correct calculation of the weekly training load. By collecting data weekly, the technical staff can compare workloads with those of previous weeks to minimize injury risk, while optimizing performance—an essential factor for a team’s success. For effective injury prevention—the primary goal of our study—daily monitoring of a footballer’s workload is crucial. Comparing current loads with those of previous weeks helps to avoid excessive external load and reduces the risk of overtraining.

There are studies that agree that during a week the player must cover distances with high intensity but always in moderation so that there are no injuries [9]. On the other hand, there are studies that indicate that a soccer player must train frequently at high intensities to deal with difficult situations in a match, so that he or she can deal with the worst case scenarios he or she will face during matches; therefore, fitness coaches must calculate the weekly load in every detail [10], avoiding fatigue that could prove to be a negative factor in the match to follow [11].

Our research will provide specific guidelines for managing training loads each week, presenting data on the limits of an ACWR that should not be exceeded as well as on the most vulnerable indicators of those that will be analyzed. The data, after being compared with those of previous weeks, will be able to determine the limits of each athlete, so that they are not exceeded, which would lead to a non-contact injury. However, despite the abundance of data collected from GPS, no research has been carried out to date that has attempted to synthesize a comprehensive profile of causal factors that can lead to the occurrence of musculoskeletal injuries for the four weeks before a non-contact injury, or to calculate the time period over which they should ultimately be evaluated. Studies that have attempted to compare some of the four weeks present conflicting results, while the parameters of the external load that should be evaluated are not generally clear. Our study presents the basic parameters, such as total distance, running at 15–20 km/h, 20–25 km/h, 25–30 km/h, accelerations > 2.5 m/s² and decelerations > 2.5 m/s², providing data on which of these factors plays the most important role in an injury. It is innovative to analyze all four weeks (fourth, third, second, and first) before an injury in a two-year research process. The results will also show the importance (or not) of each week separately. The ratios that most frequently lead to non-contact injury, which range mainly at levels that often approach or exceed 1.3, will be notable.

In a study involving 369 players, a total of 206 players (55.8%) suffered 538 lower limb injuries in two seasons [12]. However, an injury can depend on many other factors, such as lower limb muscle strength and anaerobic power, which are neuromuscular variables that affect performance in many sports activities, including soccer. This means that, if an athlete is injured, he or she loses many of the adaptations he or she had acquired through the training process, which can lead to a re-injury. In a study involving 57 soccer players, thirty-six of them were identified as having sustained a previous serious lower extremity injury. Of these 36 players, 23 (64%) still had significant muscle imbalance [13].

Our research is mainly based on the acute to chronic workload ratio (ACWR) to explain injuries. Essentially, ACWR aims to record both internal and external loads, with a primary focus on investigating data for their quantification. This research is based on findings that athletes’ performance can be calculated as the difference between fitness and fatigue [14]. The acute chronic workload ratio (ACWR) is based on this research, with subsequent research data focusing on the potential relationship between ACWR and injury rather than performance [10,15,16,17,18,19,20]. Essentially, ACWR records the loads an athlete is subjected to in one week (acute workload) relative to the loads of the previous four weeks (chronic workload).

An important factor to consider is the different positions in which soccer players compete on the field. Comparisons should be made between players who operate in similar areas of the pitch, as each position requires a specific level of fitness [21], and GPS devices typically provide relatively similar data for players in comparable positions. Excluding previous studies [22] that compare the workload of players in the same position and age group, no other research has examined this aspect in depth. Essentially, comparisons should focus on players who play in the same position. The variety of positions on the field is crucial, as each position serves a distinct tactical role, which is further supported by a study [23] examining the correlation between match performance and the running patterns of players in specific positions. Statistically, midfielders experience the highest frequency of injuries (43.6%), followed by defenders (30.0%), and forwards (17.9%), according to a relevant study [24].

GPS technology plays an important role in the assessment of external load and has been shown to help prevent injuries when this load is found to exceed certain limits established in previous weeks, especially for soccer players [22]. This is directly related to the findings of Tiernan [25], who studied ACWR over one season in 15 high-level soccer players and suggested that, to reduce the risk of non-contact injuries, the training load should be increased gradually and an increase in acute injury could be avoided by limiting week-to-week load change (≥9%). However, research into the relationship between load and injury is in its early stages. In this study, specific instructions will be given to manage the loads that a soccer player receives, so that he or she does not become injured. The analysis looks at all four weeks separately, which has not been thoroughly explored by existing research, and data varying from week four to week one will be presented. Thus the comparison of the microcycle that will be employed by a technical staff will be able to be compared at any time with one of the previous microcycles, knowing the limits that should not be exceeded.

The aim of this study was to contribute to the understanding of muscle injury prevention by evaluating and comparing data to reduce the risk of injury. Specifically, the study sought to determine whether there is a correlation between injury occurrence and variations in training volume, as measured by the acute:chronic workload ratio (ACWR), during the four, three, two, and one week periods leading up to an injury. Existing research reveals a gap in the literature, particularly regarding the recording of data from the three weeks immediately prior to an injury. Additionally, there are conflicting findings concerning the external load parameters that contribute to non-contact injuries, as well as the optimal time frame for evaluation. This research aimed to address these gaps by exploring the impact of exogenous injury factors, particularly training load, on injury prevention. The goal was to enhance both the physical and mental resilience of athletes, maximize athletic performance, and reduce the risk of injury. The study also aimed to evaluate key parameters of external training load and physical performance in athletes, and to investigate the correlation between these factors and injuries through statistical tests and prospective controls. Another aim was to compare two groups of soccer players, matched by attributes such as playing position and age. Both groups were analyzed over the same time frames and using the same parameters.

2. Methods

2.1. Experimental Approach to the Problem

Previous research has correlated loads at two and four weeks before an injury [22]. Therefore, we found it interesting to analyze all four weeks (fourth, third, second and first) prior to the week the injury occurred in a research process spanning two seasons. Forty professional soccer players participated in this study. Using GPS technology with a frequency of 10 Hz (Vector S7, Catapult, Catapult Sports Ltd., Melbourne, Victoria, Australia) the external load from the athletes was collected and analyzed for the seasons 2021–2022 and 2023–2024, according to the study of Clavel [26], regarding its reliability. During the weeks when specific indicators were evaluated, there were also 24 microcycles, where in the middle of the week there was an additional match. Both training sessions and official matches took place on outdoor pitches with natural grass, in a study that included 300 training sessions in 50 microcycles and 60 official matches.

2.2. Subjects

The study involved professional soccer players from Greece’s Super League 1 and Super League 2 over the course of two seasons. Specifically, the data from injured soccer players were analyzed and compared with the loads of non-injured players over the same 5-week period, ensuring that both groups played in the same position. In total, 40 professional soccer players (20.6 ± 1.6 years, 179.7 ± 4.8 cm, 76.8 ± 4.3 kg, 8.6 ± 1.3% body fat) participated in the study. Prior to the study’s commencement, all procedures were clearly and thoroughly explained to the participants. The athletes then read and signed a consent form to participate. Furthermore, in line with ethical guidelines for sports and exercise research, the study received approval from the ethics committee of Aristotle University of Thessaloniki (96/2021).

2.3. Running Demand Analysis

Throughout all training sessions and official matches, the GPS device was positioned on the upper back of each soccer player. Each player used the same device during the research period. The reliability and validity of GPS systems have been reported [27,28] and they have been used by soccer players during official matches [1,25,29,30]. Total distance covered, high-speed running (14.4–19.8 km/h), very high-speed running (19.8–25.0 km/h), sprints (>25.0 km/h), and accelerations with decelerations (>2.5 m/s²) were all measured.

The data analyzed consisted of 50 microcycles per complete one-week training block, per player and position, as outlined by Owen [12]. Training days were coded as MD-6, MD-5, MD-4, MD-3, MD-2, and MD-1 within a weekly block, which concluded with the match day (MD). For a six-day microcycle, MD+1 represents recovery, MD+2 represents rest, while MD-4, MD-3, MD-2, and MD-1 are observed. In a seven-day microcycle, which includes an additional day before the official match, MD+1 is recovery, MD+2 is rest, and MD-5, MD-4, MD-3, MD-2, and MD-1 are followed. Longer blocks are also utilized. For instance, in an eight-day microcycle, MD+1 is recovery, MD+2 is rest, and MD-6, MD-5, MD-4, MD-3, MD-2, and MD-1 are incorporated, while a nine-day microcycle includes MD+1 for recovery, MD+2 for rest, and MD-7, MD-6, MD-5, MD-4, MD-3, MD-2, and MD-1. All athletes were thoroughly familiar with the experimental procedures. Data were only collected from soccer players who completed the full daily training regimen and participated in matches, while data from goalkeepers and players whose overall load was modified during the study were excluded to avoid potential fatigue or injury management issues, ensuring the study’s validity and reliability. To calculate the initial value prior to the microcycle in which the player was injured, the ACWR used data from the previous four weeks, three weeks, two weeks, and the week of the injury. Initially computed as chronic workload, the immediate workload was associated with the fourth, third, second, and first weeks, respectively. The ACWR was then calculated by dividing the immediate workload by the chronic workload (i.e., the load accumulated during the injury week divided by the average load from the previous four, three, two, or one week). Values below 0.80 or above 1.3 were considered high-risk zones for injury [17]. This implies that, if the compared loads fall between 0.80 and 1.30, the injury risk is theoretically low, but if the value is below 0.80 or exceeds 1.30, the risk of non-contact injury is elevated.

2.4. Statistical Analysis

Statistical analysis was conducted using Excel 2019 and SPSS 26. Continuous variables were presented as mean ± standard deviation, while categorical variables were reported as frequency and percentage. It is important to note that the load experienced by one athlete was compared to that of a second athlete who shared common characteristics, including the position played on the field and age. The loads compared were collected during the same weeks for all participants. The normality of the variables was evaluated using the Kolmogorov–Smirnov test. To assess differences between injured and non-injured athletes, the t-test was used. Statistical significance for the tests and the differences observed in the analysis was set at p < 0.05. The determination of the sample size was carried out through statistical power analysis with the G*Power 3.1.9.7 software. The independent samples t-test was chosen for the analysis, as the research concerned the comparison of mean values between two independent groups (injured and non-injured athletes). The analysis was carried out with the following parameters: effect size (Cohen’s d) = 0.8, type I error (α) 0.05, power (1 − β) 0.80 and sample distribution ratio 1:1. Based on these parameters, G*Power estimated that the minimum required sample size to draw reliable conclusions was 42 individuals (21 per group), with an estimated statistical power of 0.816. In the present study, data were ultimately collected from 40 professional soccer players, of whom 20 athletes sustained an injury during the two-year follow-up period, while 20 athletes remained healthy. Although the final sample was slightly smaller than initially suggested by the power analysis, it was considered sufficient to conduct statistical processing and draw conclusions, given the homogeneity of the participants and the nature of the study in elite professional soccer players.

3. Results

In total and during the competitive microcircuits, twenty non-contact soccer players were injured. In the playing season and over fifty microcycles, the rate of specific injuries was 4.5/1000 h. Another important element of the research process concerns the position played by the athletes who were injured; there were six central midfielders, five forwards, five fullbacks, two central defenders and two wingers (

Table 1. Positions and details of injured soccer players.

		Mean	SD	N	N%
Age		20.6	1.6
Position	Central defender			2	10.0%
	Fullback			5	25.0%
	Central midfielder			6	30.0%
	Winger			2	10.0%
	Forward			5	25.0%

).

3.1. ACWR Four Weeks Prior to Injury

According to the results shown in

Table 2. ACWR at four weeks.

	Injury						p-Value
	No		Yes		Total
	Mean	SD	Mean	SD	Mean	SD
ACWR—Total Distance (km)	1.07	0.14	1.18	0.21	1.13	0.18	0.063
ACWR—Distance Speed Range (15–20 km/h)	1.04	0.28	1.40	0.58	1.22	0.49	0.015 *
ACWR—Distance Speed Range (20–25 km/h)	1.05	0.26	1.25	0.36	1.15	0.33	0.045 *
ACWR—Distance Speed Range (>25 km/h)	1.03	0.32	1.49	0.65	1.26	0.56	0.008 *
ACWR—# of Accelerations (>2.5 m/s2)	1.07	0.20	1.30	0.31	1.19	0.28	0.010 *
ACWR—# of Decelerations (>2.5 m/s2)	1.09	0.21	1.29	0.33	1.19	0.29	0.025

(*) indicates statistical significance at p < 0.05. (#) indicates number.

(four-week moving average ACWR), a significant statistical difference was identified for the parameters Distance Speed Range (15–20 km/h), Distance Speed Range (20–25 km/h), Distance Speed Range (>25 km/h), Accelerations (>2.5 m/s²), and Decelerations (>2.5 m/s²) based on the ACWR index. In every instance, the mean value for injured players was higher than that of non-injured players.

3.2. ACWR Three Weeks Prior to Injury

For the three-week moving average ACWR model, a statistically significant difference was found in the parameters of Total Distance (km), Distance Speed Range (15–20 km/h), Distance Speed Range (>25 km/h), Accelerations (>2.5 m/s²), and Decelerations (>2.5 m/s²), as presented in

Table 3. ACWR at three weeks.

	Injury						p-Value
	No		Yes		Total
	Mean	SD	Mean	SD	Mean	SD
ACWR—Total Distance (km)	1.07	0.10	1.18	0.20	1.12	0.17	0.033 *
ACWR—Distance Speed Range (15–20 km/h)	1.02	0.16	1.32	0.45	1.17	0.36	0.007 *
ACWR—Distance Speed Range (20–25 km/h)	1.05	0.25	1.20	0.29	1.12	0.28	0.081
ACWR—Distance Speed Range (>25 km/h)	1.04	0.32	1.46	0.53	1.25	0.48	0.004 *
ACWR—# of Accelerations (>2.5 m/s2)	1.07	0.17	1.26	0.25	1.16	0.23	0.009 *
ACWR—# of Decelerations (>2.5 m/s2)	1.08	0.16	1.26	0.25	1.17	0.23	0.013 *

(*) indicates statistical significance at p < 0.05. (#) indicates number.

. In all cases, the mean values for injured players were higher than those for non-injured players.

3.3. ACWR Two Weeks Prior to Injury

According to the results presented in

Table 4. ACWR at two weeks.

	Injury						p-Value
	No		Yes		Total
	Mean	SD	Mean	SD	Mean	SD
ACWR—Total Distance (km)	1.08	0.15	1.14	0.14	1.11	0.15	0.208
ACWR—Distance Speed Range (15–20 km/h)	1.04	0.17	1.25	0.30	1.14	0.26	0.008 *
ACWR—Distance Speed Range (20–25 km/h)	1.04	0.26	1.16	0.24	1.10	0.26	0.134
ACWR—Distance Speed Range (>25 km/h)	1.08	0.34	1.40	0.44	1.24	0.42	0.012 *
ACWR—# of Accelerations (>2.5 m/s2)	1.08	0.20	1.19	0.15	1.13	0.18	0.065
ACWR—# of Decelerations (>2.5 m/s2)	1.09	0.19	1.18	0.15	1.14	0.18	0.086

(*) indicates statistical significance at p < 0.05. (#) indicates number.

for the two-week moving average ACWR model, a statistically significant difference was observed in the Distance Speed Range (15–20 km/h) and Distance Speed Range (>25 km/h) based on the ACWR index. In every case, the mean value for injured players was higher than that for non-injured players.

3.4. ACWR One Week Prior to Injury

Based on the results presented in

Table 5. ACWR at one week.

	Injury						p-Value
	No		Yes		Total
	Mean	SD	Mean	SD	Mean	SD
ACWR—Total Distance (km)	1.04	0.11	1.08	0.12	1.06	0.11	0.214
ACWR—Distance Speed Range (15–20 km/h)	1.04	0.28	1.40	0.58	1.22	0.49	0.015 *
ACWR—Distance Speed Range (20–25 km/h)	0.98	0.16	1.08	0.22	1.03	0.19	0.126
ACWR—Distance Speed Range (>25 km/h)	1.06	0.27	1.30	0.33	1.18	0.32	0.016 *
ACWR—# of Accelerations (>2.5 m/s2)	1.02	0.14	1.11	0.13	1.07	0.14	0.049 *
ACWR—# of Decelerations (>2.5 m/s2)	1.03	0.13	1.10	0.12	1.06	0.13	0.110

(*) indicates statistical significance at p < 0.05. (#) indicates number.

for the one-week moving average ACWR model, a statistically significant difference was observed in the Distance Speed Range (15–20 km/h), Distance Speed Range (>25 km/h), and Accelerations (>2.5 m/s²) according to the ACWR index. In each case, the mean value for injured players was higher than that for non-injured players.

4. Discussion

Starting with the ACWR model analyzing the four weeks prior to injury, a statistically significant difference in the ACWR index was observed in the Distance Speed Range (15–20 km/h), Distance Speed Range (20–25 km/h), Distance Speed Range (>25 km/h), Accelerations (>2.5 m/s²), and Decelerations (>2.5 m/s²) (Table 2) (Figure 1). For the ACWR model covering the three weeks before injury, significant differences were found in Total Distance (km), Distance Speed Range (15–20 km/h), Distance Speed Range (>25 km/h), Accelerations (>2.5 m/s²), and Decelerations (>2.5 m/s²) (Table 3) (Figure 2). In the two-week ACWR model, significant differences appeared in the Distance Speed Range (15–20 km/h) and Distance Speed Range (>25 km/h) (Table 4) (Figure 3).

Across all models, the mean values for injured players were consistently higher than those for non-injured players.

Regarding the ACWR model for the final week before injury, statistically significant differences were noted in the Distance Speed Range (15–20 km/h), Distance Speed Range (>25 km/h), and Accelerations (>2.5 m/s²), again with injured players showing higher mean values (Table 5) (Figure 4). Based on these findings, several conclusions can be drawn that offer important guidance for training management and injury prevention. Notably, ratios most associated with non-contact injuries often approached or exceeded a value of 1.3. Additionally, it was observed that in at least two of the six analyzed categories (Total Distance, 15–20 km/h, 20–25 km/h, 25–30 km/h, accelerations > 2.5 m/s², and decelerations > 2.5 m/s²), injured soccer players showed higher loads during the week of injury compared to the preceding weeks, whether analyzed across four-week (ACWR4), three-week (ACWR3), two-week (ACWR2), or one-week (ACWR1) blocks.

The speed parameter of 25–30 km/h appeared to play an important role when the limits were exceeded in all four-week blocks analyzed, adversely affecting an injured soccer player, which seemed to occur in conjunction with runs at speeds of 15–20 km/h, accelerations (>2.5 m/s²), as demonstrated by the research of Fousekis [22]. In the comparative analysis, a notable discrepancy was identified in the Distance Speed Range (>25 km/h) category, with values of 1.58 over four weeks and 1.47 over two weeks for soccer players who did not experience a contact injury. In the p-value category, it is still noteworthy that, even if the four weeks recorded before an injury gave significant indications of the thresholds crossed and negatively affecting a football player, in the third and fourth weeks, this phenomenon was more pronounced for five of the six parameters, in one week for three of the five parameters, and in two weeks for two of the six parameters.

Cumulatively and for all parameters, the athletes who were not injured were at 1.06 while those who were injured were at 1.18. More specifically, in the four weeks before the incident, those football players who were injured were cumulatively at 1.32, while those who were not injured were at 1.06. In the three weeks before the incident, the injured were at 1.17 and the uninjured were at 1.06. At two weeks, the distance between the two groups decreased to 1.14 for the injured and 1.07 for the uninjured, while at one week, it was 1.10 for the injured and 1.03 for the uninjured In the Distance Speed Range (>25 km/h) according to the analysis of all four weeks, there is a wide variation; the value starts in the fourth week at 1.49, with this trend decreasing slightly in the analysis of one week, reaching close to 1.18. Moreover, as some elite soccer players also compete with their national teams, and given the use of different tracking systems across teams, conflicting data often arise. According to Buchheit [31], this inconsistency makes the prediction of non-contact injuries significantly more challenging, as the data may vary considerably, leading to gaps in load monitoring and affecting the reliability of subsequent weekly load comparisons.

In general, the increase in load led to non-contact injuries in athletes, which is consistent with our findings. This aligns with the research by White [32], where the investigation of injury probability using the ACWR over the previous four weeks—based on total distance, high-intensity speed, acceleration, and deceleration at high intensity—demonstrated an increased risk of injury when the threshold was exceeded (>1.30), compared to the moderate ACWR zone (~0.91 to ~1.20).

A very important element is the technique and tactics in soccer, which greatly affects the physical requirements of each position on the field, thus differentiating the needs that each soccer player will have to cope with. Our research was aimed at this element for a fair comparison between the same positions played by injured and uninjured soccer players, something that had a natural consequence and the corresponding tensions that accompany each special position on the field. Depending on the different positions of the soccer players in a match, the performances vary [33] which necessitates the personalized training process, necessary during a microcycle.

In all cases, the mean value of injured players was higher than those who were not injured, which was in agreement with the findings of Ehrmann [24], who concluded that injuries occurred when there were significantly higher steps per minute in the weeks preceding an injury compared to their usual averages. In this study [34], in 19 professional soccer players playing in the Australian Hyundai A-League and followed for one full season, it was concluded that injuries occurred when there were significantly higher steps per minute in the weeks before an injury compared to their usual averages. In contrast, Suarez-Arrones [35], in a study involving 15 elite soccer players competing in a high-level team, showed that results from the ACWR were not related to subsequent incident injury in professional soccer players. Differences in participation in official competitions caused significant imbalances in chronic external loads between athletes on a team, which should be minimized in training, to avoid substantial changes in workload for those who do not usually play.

In our study, a clear correlation was found between ACWR scores and the occurrence of injuries over two full seasons among professional athletes. During the week in which the injury occurred, the 20 injured soccer players significantly exceeded the loads recorded during the preceding four weeks (ACWR4), three weeks (ACWR3), two weeks (ACWR2), and one week (ACWR1). Essentially, the findings highlight that a sudden increase in load within a microcycle, compared to the previous periods, does not help maintain optimal playing condition and does not protect the athlete from sustaining an injury during training sessions or matches at the end of the microcycle.

In each microcycle, sometimes separate data were presented, due to their diversity, since training days could be less or more, depending on the competition program and always according to Owen’s theory, that was followed [12].

When running volumes at higher intensities (15–20 km/h, 25–30 km/h) increased, the present study found a higher incidence of non-contact injuries. A possible association was also observed, particularly with accelerations and to a lesser extent with decelerations, in relation to the occurrence of non-contact injuries. This finding aligns with that of Schache [36], who noted that concentric and eccentric movements are largely responsible for many specific injuries. Furthermore, it is important to highlight that, during the two weeks and one week prior to an injury, the ACWR threshold should be set lower. Although it decreases, it remains a critical factor in injury prevention efforts, as indicated by Fousekis [22]. In addition, the analysis showed that, when comparing two-week and less-than-four-week periods, the differences in most variables were more pronounced. In contrast, no significant findings regarding the induction of non-contact injury were reported by Fanchini [37], with the change in load from week to week, but also cumulatively showing poor prediction for the two weeks prior to an injury.

There were several specific limitations in our research process. One limitation is that only male soccer players participated, and another is that only certain performance indicators were evaluated.

5. Practical Applications

The comparison of all four models (ACWR4, ACWR3, ACWR2, ACWR1) in this study provided evidence suggesting a strong relationship with the occurrence of non-contact injuries in professional soccer players. Among them, the ACWR4 and ACWR3 models appeared to be the most critical indicators, as greater deviations across most variables were observed, while the ACWR1 and ACWR2 models showed a clear trend toward injury risk. Specifically, in the four-week analysis (ACWR4) prior to injury, it was found that, aside from total distance, increased load across all high-intensity categories contributed to the occurrence of non-contact injuries. For the three-week analysis (ACWR3), a statistically significant difference in the ACWR index was detected for Total Distance (km), Distance in the 15–20 km/h range, Distance above 25 km/h, Accelerations (>2.5 m/s²), and Decelerations (>2.5 m/s²).

Similarly, in the two-week analysis (ACWR2), significant differences were found for Distance in the 15–20 km/h and >25 km/h speed ranges. Regarding the one-week model (ACWR1), statistically significant differences were observed in Distance in the 15–20 km/h range, Distance above 25 km/h, and Accelerations (>2.5 m/s²). The ratios most commonly associated with non-contact injuries often approached or exceeded 1.3. Furthermore, the 25–30 km/h speed range emerged as particularly important, especially when thresholds were exceeded during four-week monitoring periods. Finally, it was noted that across at least two of the six categories analyzed (Total Distance, running speeds of 15–20 km/h, 20–25 km/h, 25–30 km/h, accelerations > 2.5 m/s², and decelerations > 2.5 m/s²), injured soccer players consistently registered high loads during the week in which their injury occurred. Also, the comparison and analysis of data between soccer players from the two research groups who had the same characteristics (weeks, playing position, tactics and age), was crucial, as clearer conclusions were drawn about the real causes that led a soccer player to injury, compared to a soccer player who did not have an injury, evaluating a situation with many similarities. Regarding the ACWR limits that positively or negatively affect an injury, data were presented where football players who were not injured for all parameters cumulatively were at 1.06 and those who were injured were at 1.18. More specifically, in the four weeks before the event, the football players who were injured were cumulatively at 1.32, while those who were not injured were at 1.06. In the three weeks before the event, the injured were at 1.17 and the non-injured were at 1.06. In the two weeks before an injury, the distance between the two groups decreased to 1.14 for the injured and 1.07 for the healthy athletes, while at one week it was at 1.10 for the injured and 1.03 for the non-injured. According to the above data, it is understood that the analysis should be performed each week separately because the limits for avoiding an injury are different for each of them. From the above presented findings, it appears that personnel dealing with fitness and injury prevention among professional soccer players should take into account the indicators of all four weeks before an injury. In addition, because the workload of soccer players does not consist only of the training conducted on the field, future research should also include the soccer player’s workload as well as their training in the gym.