Next Article in Journal
Effects of Musical Classes on Motor Creativity According to Age, Sex, and Weight Status in Young Students: A Music-Oriented versus Conventional Education Plan
Next Article in Special Issue
Investigating the Impact of Food Rewards on Children’s Motivation to Participate in Sport
Previous Article in Journal
Home Immunization with Palivizumab-A Randomized Pilot Study Describing Safety Aspects and Parents’ Preferences
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Validation of a New Soccer Shooting Test Based on Speed Radar Measurement and Shooting Accuracy

Felix Engler
Andreas Hohmann
1 and
Maximilian Siener
BaySpo—Bayreuth Center of Sport Science, University of Bayreuth, Universitätsstraße 30, 95447 Bayreuth, Germany
Institute of Professional Sport Education and Sport Qualifications, German Sport University Cologne, Am Sportpark Müngersdorf 6, 50933 Cologne, Germany
Author to whom correspondence should be addressed.
These authors contributed equally to this work and share first authorship.
Children 2023, 10(2), 199;
Submission received: 11 December 2022 / Revised: 16 January 2023 / Accepted: 17 January 2023 / Published: 20 January 2023
(This article belongs to the Special Issue Current Issues and Future Perspectives in Youth Sports)


Due to poor results, the German talent identification and development of the German soccer association DFB no longer performs a shooting test since a few years. The aim of this study was to create and validate a new soccer shooting test that allows valid conclusions to be drawn from the shooting quality of youth soccer players about their overall soccer skills. The shooting test was performed with a total of 57 male club players (age: 15.24 ± 0.864 years) from four different teams from the first, second, fifth, and the seventh division of the respective age group (under 15-year-olds until under 17-year-olds). Each subject took one shot at maximum shot speed and eight target shots, measuring accuracy and the shot speed. A multivariable linear regression analysis with forward selection revealed significant values for the variables average shot speed nondominant leg (p < 0.001) and total score (p = 0.004; accuracy × speed of every target shot). Based on these two variables, the soccer skills could be derived from the shooting skills of the adolescents in 57.4% of the cases. The study shows the importance of a good technique with the nondominant leg and of the ability to shoot accurately as well as fast simultaneously.

1. Introduction

In order for a club or association to establish itself sustainably at national and international levels in the highest, most profitable competitions, it is essential to lay a solid foundation in youth teams. That was one of the reasons for setting up the youth development centers in Germany, by the German Soccer Association (DFB) [1]. Due to the decreasing average age in Germany’s top division [2], the development time for junior players is decreased. Thus, early talent identification and selection (in combination with youth performance development) based on objective criteria plays a decisive role at youth level [3]. Therefore, since 2004, there have been sports scientific performance tests at DFB bases in which all selected players must complete [4]. In order to obtain a comprehensive picture of the overall soccer performance of the base’s players, the following essential skills are tested during the tests: speed, agility, dribbling, ball control, and juggling [5]. The long-term evaluation of the talent development program showed that players with better test results have a relatively better chance of being selected for a further, or higher, development program in the future [6]. However, it is noticeable that an important element, the goal shot, is missing in this test battery. In the past, goal shot was tested at the DFB bases, but the test did not provide reliable results, which is why it was removed from the test battery later.
In order to explain why this happened, it is necessary to take a closer look at the set-up and procedure of the former goal shot test [7]: The subjects played the ball with one touch in a shooting area in front of them. This area was placed centrally in front of the goal at a distance of 14.06 m and was 2.44 m long. After that, the subjects shot the moving ball toward the goal out of motion (the ball had to be shot from a distance of 14.06 m–16.5 m from the goal). The shooting area was 4.88 m wide. The goal was divided into only three areas (left, right, and center); each was 2.44 m in width and height. The subjects were to attempt to shoot at the left and right target fields of the goal with two shots per leg—i.e., to fire a total of eight shots—and to hit the target with their maximum shot speed. Hits were scored as 1 and a miss as 0. Speed was rated only subjectively by the test administrator. A low velocity shot was scored as 1, whereas medium and high velocity shots were scored as 2 and 3, respectively [7]. The test could not fulfill the quality criteria of scientific tests under these specifications. Above all, objectivity is by no means given in the measurement of the shot speed due to the subjective classification of the test administrator into the categories low, medium, and high speed. In addition to this, Höner et al. [8,9] examined the entire test battery of the DFB Talent Development Program regarding its validity and reliability, whereby the goal shot test also turned out to be unreliable (means of Cronbach’s alpha = 0.41; means of test–retest reliability = 0.31).
However, the solution should not be the elimination of the goal scoring test from performance diagnostics because goal scoring is what ultimately decides games. This is illustrated by the relevance of efficiency of chances (i.e., the ratio of goals scored to chances created), which is a decisive factor for winning games [10,11,12]. This, in turn, is conditioned by accuracy and shooting speed, which is why a goal shot test should examine these two variables.
There are already studies that deal with the measurement of shooting accuracy and shooting speed of youth soccer players [13,14,15]. However, these studies focus on the relationship between accuracy and speed of shots and not on performance diagnostics. In addition, the test designs of the mentioned studies used camera systems and evaluation programs, which are not intended for large-scale mass testing. Furthermore, this sensitive data collection is not suitable to be integrated into testing (such as in a DFB base) by untrained test administrators.
The goal of this study is to develop a soccer shooting test that allows us to draw valid conclusions from the shooting quality of youth soccer players about their general soccer quality and performance level. Both the shooting accuracy and the shooting speed are to be considered. These results can later be used to evaluate the shooting skills of players and to identify talents.

2. Materials and Methods

2.1. General Study Design

In this study, both the shot power based on speed radar measurements and the shot accuracy will be measured. The latter was achieved by dividing a regular men’s soccer goal (7.32 m × 2.44 m, [16]) into five areas (Figure 1). The four areas in the corners of the goal mark the targets at which the subjects should aim. The target net shown in Figure 1, which was made specifically for this study, is attached to the post and crossbar of the goal in an uncomplicated manner using the Velcro tabs shown (Figure 1) and pulled tight. Shot speed was measured using a radar sensor with associated speed indicator from [17]. The radar sensor is placed centrally at behind the goal so that the balls shot at the goal reflect the signal emitted by the sensor back to it, and thus the speed of the balls can be measured.
To be able to make valid statements, teams from different performance classes had to be tested. These included one team from each of the following leagues: 1st division U15 (under 15 years old), 2nd division U17, 5th division U17, and 7th division U15. The measured values were later adjusted for age differences and physical maturity stage during the evaluation to ensure comparable results.

2.2. Participants

The soccer shooting test was carried out engaging four different male junior teams of approximately the same age, whereby the performance level of the teams was to vary. Thus, for organizational reasons, the study concentrated on southern Germany, and four teams from different divisions—from the highest to the lowest league—were recruited for the study.
A total of n = 57 participants (age: 15.24 ± 0.86 years; height: 174.4 ± 6.9 cm; weight: 59.9 ± 7.4 kg) took part in the study. From the 7th division, 12 subjects (age: 14.01 ± 0.43; height: 170.5 ± 7.2 cm; weight: 55.0 ± 7.1 kg) were tested; from the 5th division, 18 subjects (age: 15.55 ± 0.76 years; height: 175.3 ± 6.2 cm; weight: 62.8 ± 6.0 kg); from the 2nd division, 17 subjects (age: 15.91 ± 0.29 years; height: 178.1 ± 6.2cm; weight: 62.3 ± 7.8 kg); and from the 1st division, 10 subjects (age: 14.99 ± 0.21 years; height: 170.9 ± 5.5 cm; weight: 56.9 ± 6.1 kg) participated in the test.
All participants gave verbal consent for the test to be conducted. The data are available in anonymous form. Ethics confirmation by the university was not applicable.

2.3. Test Set-Up

Figure 2 shows the set-up of the goal shot test. The radar sensor (1) is set up in a central position behind the goal at a distance of 4 m from the goal line and connected to the speed indicator of the radar sensor (2). The target net (4) is attached to the crossbar and post of the goal with the Velcro tabs provided (Figure 1) and pulled as tight as possible.
In order to improve the very simplified goal segmentation of the original DFB shot test and thus improve the measurement accuracy of the shot test, the described goal target net (Figure 1) was manufactured especially for this test. The dimensions of the target areas in the corners shown in Figure 1 are given below using the example of the upper left field: The upper side is 3 m long; the left side measures 1.14 m. The lower side, which is the boundary to the target area on the lower left, is 1.5 m long, and the diagonal side on the right is 1.86 m long. These dimensions now result in a target area size of 2.57 m2—i.e., less than half the size of the target fields from the DFB’s goal shot test, which are 5.95 m2 in size.
The BallSpeedoMeter radar sensor (1) [17] is aligned on a vertical plane to the center of the goal using the height-adjustable tripod on which it is installed. A cable connects the sensor to the associated speed display board (2), which is powered by a rechargeable battery and also supplies power to the sensor. According to the manufacturer, when tested with a set-up such as the one used in this study, the device measures the speed of the ball with a maximum deviation of 1.021%.
Since radar measuring devices cannot be directly standardized, comparative tests with other measuring devices were already carried out in the run-up to the study. Thereby, the speed of a moving object was measured using GPS (Google maps), a laser measurement device (LDM300C by ‘Jenoptic’, Jena, Germany), and a light barrier (Speedtrap 2 by ‘Brower Timing Systems’, Draper, UT, USA) and the BallSpeedoMeter. A correlation analysis proved in our own studies that the results of the BallSpeedoMeter correlated with other measuring devices with a significance of p < 0.001 and Pearson’s r > 0.983, which shows that the used BallSpeedoMeter has a very high measuring accuracy, making it very suitable for the planned shooting test. Cronbach’s alpha was α = 0.80 for the test–retest reliability of the present sample. Applying the Bland and Altman method [18,19], it can be seen that the second attempt was, on average, 2.81 ± 7.14 km/h (minimum = −15.3, maximum = 20.6) faster than the first attempt.
The test administrator records the results (3) and stands next to the goal so that a good view of both the speed display and the goal is ensured in order to record the results correctly. A cone 5.5 m from the goal line in a central position (5) serves as a marker for the maximum shot speed test. The shooting area from which the subjects have to take the target shots (6) is marked by four cones. It is 4.88 m wide and is located centrally in front of the goal at a distance of 14.06 m. The length is 2.44 m, which then places the cones far from the goal at a distance of 16.5 m from the goal line. Behind the shooting area, there is the starting position of the subjects (7). There, the normal game balls are ready—ideally eight of them, so that the subjects can perform their eight target shots without interruption.

2.4. Measurements

The test was conducted following the specified procedure: The subjects warm up collectively as part of their normal practice routine. After warming up, the group begins their normal practice, and in parallel, subjects are taken out of practice one by one to perform the goal shot test.
First, the anthropometric data of the subject were noted. This included date of birth, height, weight, and the subject’s dominant and nondominant shooting leg. Subsequently, the subject shot a stationary ball from the marking of the maximum shot speed test (Figure 2, No. 5) on the goal with maximum shot speed. For this single shot, the speed indicator of the radar sensor (Figure 2, No. 2) was pointed at the subject so that he could see it, thus providing him with additional motivation. For the following target shots, the speed indicator was turned away such that only the test administrator (Figure 2, No. 3) could see it. This was to ensure that the subject could concentrate on the accuracy of the shots and was not distracted by the shooting speed.
For the target shots, the subject played the ball from the starting position (Figure 2, No. 7) with one touch into the shooting area and then shot the moving ball out of the shooting area (Figure 2, No. 6) into the goal. The ball was to be played in motion in order to match the finishing situation as closely as possible to a game situation and not to play stationary balls, as in the case of a penalty or free kick. It was then started with four shots with the dominant leg (DL) in each case, with the targets in the corners to be aimed at in the following order: upper right, lower right, upper left, and lower left. Then four shots were shot with the nondominant leg (NDL) in the same target sequence. The order for this was fixed and identical for every subject, so that in case of accidental hits into wrong targets, the subjects could not claim that they had just aimed there. As a support, the target to be aimed at was briefly called out to the subjects before each shot, in order to avoid misunderstandings. After all nine shots (one maximum speed shot, eight target shots) had been made and noted by the test administrator, the subject was released back into the running practice, and the next subject came to the test set-up.
The accuracy of the tested shots was recorded using accuracy factors. For this purpose, hits were assigned a score of 3, boundary hits (i.e., the ball hit the band or post or crossbar of the desired target) were assigned a score of 2, and misses (i.e., the ball missed the goal or went into another target) were assigned a score of 1. The speed of the shot in km/h was then multiplied with the accuracy factor of the associated shot. Thus, scores were available for each individual shot, which could be added up as desired (depending on the goal of the analysis). For example, the variable score NDL describes only the score of the shots with the NDL, whereas the variable total score includes all scores of all target shots.

2.5. Coach Ranking

A ranking of the tested players on their team was created by each coach after the test. However, this ranking was not influenced by the results of the test, as the results were not yet available to the coaches. In this process, the coaches were asked to rank their players from good to poor based on overall shooting quality. Subsequently, the rankings of the coaches were combined into an overall ranking to serve as a reference value for the analyses later on: The 1st division player ranked best by his coach was assigned first place. The best rated player of the 2nd division was assigned the eleventh place, since 10 players from the 1st division were tested, and consequently, he was the next best player. Thus, the last place in the overall ranking was occupied by the player who was rated the worst by his coach from the worst team (i.e., 7th division). Hence, there was a continuous overall ranking, which was based on the assumption that the worst player from a team still had better shooting qualities than the best shooter from the team of a division below. It should be noted that although n = 57 players were tested, only 54 players appeared in the overall ranking, because on the day of the testing, three players from the 2nd division team were present for a tryout, so the coach did not have sufficient knowledge about their abilities. These players participated in the test, but the coach was not able to make a statement about the shooting qualities of these players, which is why they could not be included in the ranking.

2.6. Statistical Analyses

All statistical analyses were performed using SPSS (version 26.0; SPSS Inc., Chicago, IL, USA) with a significance level of p ≤ 0.05.
To ensure comparability of the data despite the age differences between the teams, the measured values were adjusted for age. For this purpose, linear regression analyses were calculated, each of which placed the corresponding test variable (dependent variable) in relation to age in months (independent variable). From this, z-standardized residuals were then generated. Thus, age-adjusted values for each variable were available for further calculations. All values were approximately normally distributed, as shown by the Shapiro–Wilk test (p > 0.05) and a visual inspection of the histograms.
To obtain an overview of the explanatory power of the different test variables, a one-factor analysis of variance (ANOVA) was performed to analyze the differences between the teams or leagues regarding their test results. Afterward, a post hoc test (Bonferroni/Dunnet-T3) was used to test the pairwise differences between the mean scores of the teams.
In addition, a bivariate correlation analysis by Spearman was to show which test variables correlates with the coach ranking.
A regression analysis was necessary to finalize the variables with the best explanatory power. For this purpose, multiple linear regression analyses with forward selection were calculated with the test variables as independent variables and the coach ranking as the dependent variable (missing values: pairwise). The independent test variables included in the calculations were the uncorrected raw scores in a first analysis and the age-corrected values in a second analysis. This was supposed to provide information on whether the age differences of the adolescents have an influence on the results.

3. Results

Table 1 shows the descriptive statistics with the test results of the shooting test. We notice that score lower targets achieves significantly better values than score upper targets. In addition, despite the age differences, clear differences can be seen between the performance levels regarding total score. For max. shot speed, the two older teams achieved better values. Furthermore, the subjects from the highest performance level were able to shoot the hardest in relation to their maximum shot speed in the target shots, which is shown by the variable avg. shot speed/max. shot speed.

3.1. Comparison of Performance Levels

The ANOVA performed showed significant group differences between the four teams for almost all test variables. Only for the test value shots at post could no statistically significant difference between the teams be shown. The results of the ANOVA are shown in Table 2. The results of the post hoc test was marked as significant differences between the performance levels in Table 1. In contrast to the results of the ANOVA, the post hoc test showed that there were also no significant differences in some other variables. Thus, in addition to shots at post, no significant differences between the individual groups could be detected for shots missed, score DL, and score upper targets. Nevertheless, it can be seen that especially the group from the 1st division stands out from the two lower performance levels. There are no significant differences between the 1st division and the 2nd division in any of the variables. The 7th division performed the worst. It had significantly worse scores on most variables (8 of 13) than at least one other team. The teams from the highest and lowest levels differed most frequently (also eight times).

3.2. Correlation Analysis

The correlation analysis revealed that, except for shots at post, all variables correlated significantly with coach ranking at a significance level of 0.01. The Pearson R-values of the bivariate correlation analysis are shown in Table 3. The fact that shots at post is the only variable that does not correlate significantly with coach ranking and reaches only barely significant values overall is consistent with the poor results for this variable in the ANOVA. The remaining values correlate strongly with the coach ranking (p < 0.01).

3.3. Predictability of Soccer Quality through the Shooting Test

The multivariable linear regression with forward selection showed that only the two raw variables avg. shot speed NDL (p < 0.001, standardized coefficient = −0.498) and total score (p = 0.004, standardized coefficient = −0.394) had a significant influence on the model (all other variables were excluded due to nonsignificant influence). Accordingly, based on these two variables, it is possible to draw conclusions from the shooting quality to the soccer quality of the subjects. The analysis reached a corrected R2 of 0.574 and p < 0.001. The regression equation is as follows (Equation (1)):
R a n k i n g   p o s i t i o n = 108.103 0.858 × a v g . s p e e d   N D L 0.025 × t o t a l   s c o r e
If the same analysis is calculated with the age-adjusted z-scores, the same results appear. Again, the two test values from avg. shot speed NDL (p < 0.027) and total score (p = 0.014) are the only significant variables. However, the corrected R2 value decreases to 0.364.
This shows that the two included independent variables are negatively related to the dependent variable, the coach ranking. The negative correlation can be explained by the order of the coach ranking, which indicates the best rank with the lowest number.
Figure 3 and Figure 4 show the ranges of the variables avg. shot speed NDL and total score in each case in relation to the performance level. The differences between the teams are recognizable and allow areas to be defined on which basis a subject can be assigned to a performance level based on his shot quality. The overlaps show that there are smooth transitions between players of different levels, e.g., it is possible that some players from a lower league achieve better results in the shot test than a player from a higher league and vice versa. The presented mean lines show a clear loss of value for both variables the lower the performance level of the team is. Only the differences in avg. shot speed NDL between the 1st division and the 2nd division are not very large. However, the difference in performance level between these two groups is, in fact, the smallest.

4. Discussion

The goal of the study was to create a goal shot test that is suitable for making valid statements about the general soccer skills of adolescent soccer players based on the shooting quality by measuring shooting accuracy and shooting speed. Based on the test results of the subjects in interaction with the available classifications of the coaches about their players, the variables avg. shot speed NDL and total score turned out to be an accurate way to predict performance. The explanation for this is that the total score includes both the accuracy and the speed of all shots and thus represents the best overall picture of a player’s qualities. When trying to shoot accurately, shooting power decreases, and, in turn, accuracy suffers when trying to shoot very hard [14,15]. Hence, the difference between the different performance levels of players is not the maximum shot speed but rather whether players can still shoot relatively hard when they are supposed to shoot accurately, or whether they can shoot hard and accurately at the same time. This ability is comprehensively represented by the total score variable. The results of this study show that players from the higher performance levels seem to manage to shoot hard and accurately more easily than those from the lower divisions.
In addition, of great significance is the avg. shot speed NDL, since at youth age players from lower performance levels train less and can therefore focus less on training their weak foot than in higher performance levels, where practice takes place more frequently and is more intense. Ultimately, this is reflected in the avg. shot speed NDL, as poorly trained shooting technique with the NDL results in being able to shoot less strongly with it. The better results on this variable (i.e., better technique of the NDL) of the higher performance levels in this study coincides with a review article by Stoeckel and Carey [20]. They conclude that professional soccer players are not technically worse with the NDL than with the DL due to a lot of training. Transferred to the present study, this means that players of higher quality are technically better with the NDL, which is reflected in avg. shot speed NDL, as mentioned before.
Studies by Ali et al. [21] and Keller et al. [22] showed that elite soccer players on average shoot harder than nonelite soccer players. Professional athletes reached an average speed of about 80 km/h in their studies. The average shooting speed of the highest performance group in our sample was 72 km/h. This difference can probably be attributed to the age of the participants—they analyzed adult soccer players in their studies, whereas junior soccer players were tested in our study.
It is important to consider age and developmental effects, especially in the case of unevenly aged subjects, as in this study, because they influence talent identification [23,24,25]. Remarkably, the effects of age differences among teams do not appear to be evident in the results of this study, which contrasts with results of previous studies [26,27,28]. These studies investigated the development of shooting power over the developmental progression of adolescents and showed that shooting power increases the older the player becomes. Therefore, it would have been expected that in this study the adjusted values of the variables would produce better results. However, this was not the case. One explanation for this could be that the age differences were not severe, as in the previous studies. In addition, it is possible that these effects eliminated themselves because the two younger groups were those with the highest and those with the lowest performance levels. For the same reasons, it is possible that this study was unable to show age and developmental differences in terms of shooting accuracy. This also contradicts findings of other studies [13,29,30]. Given that there are no detectable effects, no conclusions can be drawn about the differences between the age and development effects that Towlson et al. [29] confirm.
Studies evaluating penalty kicks show that the probability of scoring is significantly greater when the ball is shot into the top corners of the goal compared to when it is placed at the low [31,32]. However, these studies also show that significantly more shots are taken at the bottom, as the risk of missing is significantly greater when shooting high. Thus, it is significantly more difficult to hit the upper corners, which is consistent with the results of the present study: every team tested scored better on score lower targets than on score upper targets (Table 1). Yet, the question exists how relevant the placement in the upper or bottom corners of the goal is in open game situations. Pertsukhov and Shalenko [31] found that in the 2019/2020 UEFA Champions League season, 68.1% of all goals were scored by shooting to the lower part of the goal. Consequently, the shot quality of low shots seems to be quite relevant for open play. Related to the results of this study, we can see that the better teams score significantly better on score lower targets than the worse teams. In contrast, there are no significant differences for score upper targets. Thus, it appears that the probability of scoring is higher for shots at the upper part of the goal, but so is the risk of missing the target entirely. Low shots are more likely to be blocked, but more goals are scored with low shots, which is consistent with the results of this study. Therefore, low shots seem to be more relevant than high shots due to the higher number of shots taken and the more goals that are scored.
It needs to be mentioned, however, that despite good results from the correlation analysis, the results of this study may have been affected by inaccuracies in the coach rankings because they relied solely on the subjective assessment of the coaches. One solution to this problem could have been to have each coach evaluate each player from each team. Yet this was not feasible. Therefore, due to lack of alternatives, it had to be assumed that the worst player of the best team would be ranked even better than the best player of the second worst team. This solution is not ideal, because in practice it is very likely that there is qualitative intermixing or overlapping between the teams, especially when focusing on only one skill (shooting), as can be seen in Figure 3 and Figure 4. Another solution could be a retrospective study after a few years, testing the same subjects in adulthood and then comparing them with the current values and their performance level at the men’s level. The results of the ANOVA performed, which assumes independence of the variables, should be interpreted cautiously. Due to the test design, there may be intercorrelations between variables.
For the future, it is necessary to perform the test with a larger number of adolescents in order to consolidate the results of this study. In doing so, attention should be paid to eliminate or minimize the limitations mentioned above. Talent diagnostics is a complex subject of research, because especially in sports such as soccer, it depends on many different factors that can contribute to a player reaching the professional level in later life [33]. Ultimately, it cannot be the goal to predict the overall soccer skills of a player only based on a specific shooting test. However, especially from this point of view, the results of the study are positive, showing that 57.4% of the soccer performance of youth players given by the coach ranking can be derived from the total score (total score) and the average shot speed with the NDL (avg. shot speed NDL). Furthermore, players playing in the highest performance level of their age group achieved a total score of 1219.45 points on average in the shooting test and averaged 70.223 km/h shooting with their weak foot. These values can serve as first reference values for further testing among subjects in this age group.

5. Conclusions

The study successfully demonstrated that the shooting quality is an indicator for the overall soccer performance of a young player. The slightly modified test, based on the former DFB-shooting test [7] achieved valid results. It was remarkable that the age of the youth soccer players did not seem to have any influence on the results of this study. The test showed that it was possible to draw conclusions about the overall soccer skills based on the shooting quality in far more than half of the cases on the basis of two parameters (avg. shot speed NDL and total score). Given that soccer is about more than just shooting skills and that there are many other qualities that make a good soccer player, these results should be viewed positively.

Author Contributions

Conceptualization, F.E. and M.S.; methodology, M.S.; software, F.E. and M.S.; validation, F.E. and M.S.; formal analysis, F.E.; investigation, F.E.; resources, F.E. and M.S.; data curation, M.S.; writing—original draft preparation, F.E.; writing—review and editing, M.S. and A.H.; visualization, F.E.; supervision, M.S.; project administration, F.E. All authors have read and agreed to the published version of the manuscript.


Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—491183248. Funded by the Open Access Publishing Fund of the University of Bayreuth.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Verbal consent was obtained from all subjects involved in the study.

Data Availability Statement

The data of the study are accessible on reasonable request.


We would like to thank all coaches and teams who participated in this study. We would also like to thank Michael Bösl, who prepared the way for this study and made it possible for us to carry out the investigations using the BallSpeedoMeter.

Conflicts of Interest

The authors declare no conflict of interest.


  1. DFL. Anhang V—Richtlinien für die Errichtung und Unterhaltung von Leistungszentren. In Lizenzierungsordnung; Deutsche Fußball Liga, Ed.; DFL: Frankfurt/Main, Germany, 2022. [Google Scholar]
  2. DFL. Durchschnittsalter in der Bundesliga Gesunken DFL Deutsche Fußball Liga. 2017. Available online: (accessed on 19 August 2022).
  3. Knoebel, S. Wie der Reifungsstatus die fußballerische Entwicklung beeinflusst. Z. Sportpsychol. 2021, 28, 38. [Google Scholar] [CrossRef]
  4. Höner, O. Erläuterungen zu den Individuellen Spielerauswertungen im Rahmen der Technisch-Motorischen Leistungsdiagnostik an den DFB-Stützpunkten, Tübingen. 2015. Available online: (accessed on 24 August 2022).
  5. Deutscher Fußball-Bund. Leistungsdiagnostik im DFB—Talentförderprogramm: Eine Anleitung zur Durchführung Objektiver und Subjektiver Testverfahren im Aufbaubereich. Testmanual. 2022. Available online: (accessed on 30 October 2022).
  6. Höner, O. Talentprognose: Ergebnisse zur Technomotorischen Diagnostik im DFB Talentförderprogramm; Philippka-Sportverlag: Münster, Germany, 2013. [Google Scholar]
  7. Deutscher Fußball-Bund. Testmanual für die Sportmotorische Leistungsdiagnostik—Anleitung zur Testdurchführung; Testmanual; DFB: Frankfurt am Main, Germany, 2022. [Google Scholar]
  8. Höner, O.; Votteler, A.; Schmid, M.; Schultz, F.; Roth, K. Psychometric properties of the motor diagnostics in the German football talent identification and development programme. J. Sport. Sci. 2015, 33, 145–159. [Google Scholar] [CrossRef] [PubMed]
  9. Höner, O.; Leyhr, D.; Kelava, A. The influence of speed abilities and technical skills in early adolescence on adult success in soccer: A long-term prospective analysis using ANOVA and SEM approaches. PLoS ONE 2017, 12, e0182211. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  10. Lago-Ballesteros, J.; Lago-Peñas, C. Performance in Team Sports: Identifying the Keys to Success in Soccer. J. Hum. Kinet. 2010, 25, 85–91. [Google Scholar] [CrossRef]
  11. Lepschy, H.; Wäsche, H.; Woll, A. Success factors in football: An analysis of the German Bundesliga. Int. J. Perform. Anal. Sport 2020, 20, 150–164. [Google Scholar] [CrossRef]
  12. Yue, Z.; Broich, H.; Mester, J. Statistical Analysis for the Soccer Matches of the First Bundesliga. Int. J. Sport. Sci. Coach. 2014, 9, 553–560. [Google Scholar] [CrossRef]
  13. Vieira, L.H.P.; Cunha, S.A.; Moraes, R.; Barbieri, F.A.; Aquino, R.; Oliveira, L.D.P.; Navarro, M.; Bedo, B.L.S.; Santiago, P.R.P. Kicking Performance in Young U9 to U20 Soccer Players: Assessment of Velocity and Accuracy Simultaneously. Res. Q. Exerc. Sport 2018, 89, 210–220. [Google Scholar] [CrossRef]
  14. Rakojević, B.; Mrdakovic, V.; Pažin, N.; Vulović, R.; Leontiejvic, B.; Ilic, D. Speed-accuracy tradeoff of instep kick in young players. Facta Univ. Ser. Phys. Educ. Sport 2019, 16, 543–555. [Google Scholar] [CrossRef]
  15. Andersen, T.B.; Dörge, H.C. The influence of speed of approach and accuracy constraint on the maximal speed of the ball in soccer kicking: Speed of approach. Scand. J. Med. Sci. Sport. 2011, 21, 79–84. [Google Scholar] [CrossRef]
  16. Deutscher Fußball-Bund. Fußballregeln 22/23. Regelwerk, Frankfurt am Main. 2022. Available online: (accessed on 30 October 2022).
  17. BallSpeedoMeter. The BallSpeedoMeter: Whats Your Speed? Available online: (accessed on 30 October 2022).
  18. Altman, D.G.; Bland, J.M. Measurement in Medicine: The Analysis of Method Comparison Studies. Statistician 1983, 32, 307. [Google Scholar] [CrossRef]
  19. Giavarina, D. Understanding Bland Altman analysis. Biochem. Med. 2015, 25, 141–151. [Google Scholar] [CrossRef] [Green Version]
  20. Stöckel, T.; Carey, D.P. Laterality Effects on Performance in Team Sports. In Laterality in Sports; Elsevier: Amsterdam, The Netherlands, 2016; pp. 309–328. ISBN 978-0-12-801426-4. [Google Scholar]
  21. Ali, A.; Williams, C.; Hulse, M.; Strudwick, A.; Reddin, J.; Howarth, L.; Eldred, J.; Hirst, M.; McGregor, S. Reliability and validity of two tests of soccer skill. J. Sport. Sci. 2007, 25, 1461–1470. [Google Scholar] [CrossRef]
  22. Keller, B.S.; Raynor, A.J.; Bruce, L.; Iredale, F. Technical attributes of Australian youth soccer players: Implications for talent identification. Int. J. Sport. Sci. Coach. 2016, 11, 819–824. [Google Scholar] [CrossRef]
  23. Towlson, C.; Cobley, S.; Midgley, A.; Garrett, A.; Parkin, G.; Lovell, R. Relative Age, Maturation and Physical Biases on Position Allocation in Elite-Youth Soccer. Int. J. Sport. Med. 2017, 38, 201–209. [Google Scholar] [CrossRef] [Green Version]
  24. Lovell, R.; Towlson, C.; Parkin, G.; Portas, M.; Vaeyens, R.; Cobley, S. Soccer Player Characteristics in English Lower-League Development Programmes: The Relationships between Relative Age, Maturation, Anthropometry and Physical Fitness. PLoS ONE 2015, 10, e0137238. [Google Scholar] [CrossRef]
  25. Hill, M.; Scott, S.; McGee, D.; Cumming, S.P. Are relative age and biological ages associated with coaches’ evaluations of match performance in male academy soccer players? Int. J. Sport. Sci. Coach. 2021, 16, 227–235. [Google Scholar] [CrossRef]
  26. Matta, M.D.O.; Figueiredo, A.J.B.; Garcia, E.S.; Werneck, F.Z.; Seabra, A. Morphological and maturational predictors of technical performance in young soccer players. Mot. Rev. Educ. Física 2014, 20, 280–285. [Google Scholar] [CrossRef]
  27. Hunter, A.H.; Smith, N.M.A.; Camata, T.V.; Crowther, M.S.; Mather, A.; Souza, N.M.; Ramos-Silva, L.F.; Pazetto, N.F.; Moura, F.A.; Wilson, R.S. Age- and size-corrected kicking speed and accuracy in elite junior soccer players. Sci. Med. Footb. 2022, 6, 29–39. [Google Scholar] [CrossRef]
  28. Rodríguez-Lorenzo, L.; Fernández-Del Olmo, M.; Sánchez-Molina, J.A.; Martín-Acero, R. Kicking ability and kicking deficit in young elite soccer players. Kinesiology 2018, 50, 194–203. [Google Scholar] [CrossRef]
  29. Towlson, C.; MacMaster, C.; Parr, J.; Cumming, S. One of these things is not like the other: Time to differentiate between relative age and biological maturity selection biases in soccer? Sci. Med. Footb. 2022, 6, 273–276. [Google Scholar] [CrossRef]
  30. Lipecki, K. Age-Related Differences in Fitness Performance and Technical Skills of Young Soccer Players. Pol. J. Sport Tour. 2018, 25, 8–14. [Google Scholar] [CrossRef] [Green Version]
  31. Pertsukhov, A.; Shalenko, V. Features if effective goal shots by football players in gamew of high level teams. Slobozhanskyi Her. Sci. Sport 2020, 8, 17–28. [Google Scholar]
  32. Bar-Eli, M.; Azar, O.H. Penalty kicks in soccer: An empirical analysis of shooting strategies and goalkeepers’ preferences. Soccer Soc. 2009, 10, 183–191. [Google Scholar] [CrossRef]
  33. Hohmann, A.; Siener, M. Talent Identification in Youth Soccer: Prognosis of U17 Soccer Performance on the Basis of General Athleticism and Talent Promotion Interventions in Second-Grade Children. Front. Sport. Act. Living 2021, 3, 625645. [Google Scholar] [CrossRef]
Figure 1. Target net which is attached to the posts and the crossbar of the goal.
Figure 1. Target net which is attached to the posts and the crossbar of the goal.
Children 10 00199 g001
Figure 2. Test set-up for the shooting test. (1) Radar sensor, (2) Speed indicator of the radar sensor, (3) Test administrator with evaluation sheet, (4) Target net, (5) Marker of the maximum shot speed test, (6) Shooting area for target shots, (7) Starting position of the subjects.
Figure 2. Test set-up for the shooting test. (1) Radar sensor, (2) Speed indicator of the radar sensor, (3) Test administrator with evaluation sheet, (4) Target net, (5) Marker of the maximum shot speed test, (6) Shooting area for target shots, (7) Starting position of the subjects.
Children 10 00199 g002
Figure 3. Boxplots of total score by performance level.
Figure 3. Boxplots of total score by performance level.
Children 10 00199 g003
Figure 4. Boxplots of avg. shot speed NDL by performance level.
Figure 4. Boxplots of avg. shot speed NDL by performance level.
Children 10 00199 g004
Table 1. Descriptive statistics with noted significant differences between performance levels shown through the post hoc tests.
Table 1. Descriptive statistics with noted significant differences between performance levels shown through the post hoc tests.
VariablesPerformance Level
(Sign. Difference)
Shots on target1st division (c)103.900.9940.3143.194.6125
2nd division (c)173.591.1760.2852.984.1915
5th division (a,b)182.501.2490.2941.883.1215
7th division122.671.0730.3101.983.3515
Shots at post1st division101.001.1550.3650.171.8303
2nd division170.880.9280.2250.411.3602
5th division181.000.8400.1980.581.4202
7th division120.830.8350.2410.301.3602
Shots missed1st division103.101.1010.3482.313.8914
2nd division173.531.2810.3112.874.1915
5th division184.501.4650.3453.775.2327
7th division124.501.3820.3993.625.3837
Max. shot speed (km/h)1st division (d)1093.106.6312.09788.35697.84481.2103.8
2nd division (d)1797.678.0001.94093.557101.78483.7109.0
5th division (d)1898.126.8941.62594.700101.55684.8108.1
7th division (a,b,c)1282.456.6331.91578.23586.66568.190.7
Avg. shot speed (km/h)1st division (d)1072.762.5840.81770.913074.609569.0876.78
2nd division (d)1770.665.6011.35967.785473.545560.1579.38
5th division (d)1870.005.4461.28467.293872.710360.4079.14
7th division (a,b,c),1258.754.5601.31655.856961.651448.5165.73
Avg. shot speed DL (km/h)1st division (d)1075.302.9430.93173.194477.405670.4878.85
2nd division (d)1773.106.1731.49769.929176.276862.5882.75
5th division (d)1875.006.2481.47371.892878.107265.2587.85
7th division (a,b,c)1266.135.8231.68162.431369.831255.5374.90
Avg. shot speed NDL (km/h)1st division (d)1070.223.0250.95768.058272.386865.5374.70
2nd division (d)1768.235.7671.39965.263171.192856.6876.38
5th division (d)1865.007.0361.65861.505168.503253.9877.83
7th division (a,b,c)1251.386.7841.95847.066855.687438.5063.75
Avg. shot speed/max shot speed (%)1st division (c,d)1078.373.7301.18075.7081.0473.8185.33
2nd division1772.707.2631.76268.9776.4360.8286.84
5th division (a)1871.434.3811.03369.2573.6162.7380.18
7th division (a)1271.373.8461.11068.9373.8164.4877.40
Total score1st division (c,d)101219.5149.21347.1851112.711326.191035.101467.70
2nd division (c,d)171138.9168.13140.7781052.461225.34821.301364.20
5th division (a,b)18976.3194.48445.840879.591073.01701.601469.00
7th division (a,b)12841.8166.78048.145735.82947.76528.501175.80
Score lower targets1st division (c,d)10745.9371.48122.604694.80797.06658.60890.10
2nd division (c,d)17728.46146.01835.415653.38803.53474.40909.00
5th division (a,b)18607.34137.68532.453538.88675.81407.80918.90
7th division (a,b)12501.46117.14833.818427.03575.89292.60689.70
Score upper targets1st division10473.52152.27348.153364.59582.45278.40691.70
2nd division17410.44124.83830.278346.26474.63255.20786.80
5th division18368.9691.77421.631323.32414.59230.00550.10
7th division12340.3392.16026.604281.78398.89228.40526.20
Score DL1st division10626.27107.00033.836549.73702.81450.40791.50
2nd division17616.11103.51325.106562.88669.33411.80777.40
5th division18519.50116.66327.498461.48577.52261.00767.70
7th division12515.08170.57949.242406.69623.46266.60860.70
Score NDL1st division (c,d)10593.1877.92824.643537.43648.93459.80694.30
2nd division (d)17522.7996.16223.323473.35572.24340.70659.70
5th division (a,d)18456.80107.89225.431403.15510.45256.70701.30
7th division (a,b,c)12326.7284.34824.349273.12380.31192.30447.20
DL = dominant leg; NDL = nondominant leg; Differs significantly from (a) 1st division (b) 2nd division; (c) 5th division; (d) 7th division.
Table 2. Results of ANOVA.
Table 2. Results of ANOVA.
VariablesF (3.53)pη2
Avg. shot speed NDL23.599<0.001 **0.572
Avg. shot speed19.641<0.001 **0.526
Score NDL16.535<0.001 **0.483
Max. shot speed13.959<0.001 **0.441
Total score11.391<0.001 **0.392
Score lower targets10.124<0.001 **0.364
Avg. shot speed DL6.997<0.001 **0.284
Shots on target4.8290.005 **0.215
Avg. shot speed/max. shot speed4.4330.007 **0.201
Shots missed3.6020.019 *0.169
Score DL3.1880.031 *0.153
Score upper targets2.8910.044 *0.141
Shots at post0.1120.9530.006
** High significance; * medium significance; DL = dominant leg; NDL = nondominant leg. Effect sizes (ƞ2): 0.01 ≤ small, 0.06 ≤ medium, 0.14 ≤ strong.
Table 3. Spearman correlation analysis of test variables and coach ranking.
Table 3. Spearman correlation analysis of test variables and coach ranking.
Coach RankingMax. Shot SpeedAvg. Shot SpeedAvg. Shot Speed DLAvg. Shot Speed NDL% of Max. Shot SpeedTotal ScoreScore Lower TargetsScore Upper TargetsScore DLScore NDLShots on TargetShots at PostShots Missed
Coach ranking-−0.421 **−0.684 **−0.462 **−0.720 **−0.409 **−0.666 **−0.619 **−0.402 **−0.427 **−0.678 **−0.433 **−0.0140.387 **
Max. shot speed -0.691 **0.552 **0.668 **−0.338*0.411 **0.525 **0.0620.277 *0.409 **0.0570.185−0.167
Avg. shot speed -0.854 **0.926 **0.444 **0.570 **0.533 **0.340 **0.368 **0.584 **0.143−0.027−0.027
Avg. shot speed DL -0.594 **0.420 **0.337 *0.262*0.269 *0.299 *0.261−0.046−0.0800.092
Avg. shot speed NDL -0.381 **0.637 **0.634 **0.330 *0.352 **0.713 **0.2550.016−0.236
% of max. shot speed -0.2450.0570.369 **0.1400.269 *0.135−0.280 *0.057
Total score -0.842 **0.716 **0.837 **0.828 **0.840 **0.132−0.828 **
Score lower targets -0.2260.662 **0.740 **0.705 **0.040−0.650 **
Score upper targets -0.654 **0.536 **0.603 **0.187−0.653 **
Score DL -0.385 **0.728 **0.131−0.728 **
Score NDL -0.669 **0.088−0.648 **
Shots on target -−0.166−0.781 **
Shots at post -−0.487 **
Shots missed -
** Correlation is significant at the 0.01 level (2-tailed); * Correlation is significant at the 0.05 level (2-tailed); DL = dominant leg; NDL = nondominant leg.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Engler, F.; Hohmann, A.; Siener, M. Validation of a New Soccer Shooting Test Based on Speed Radar Measurement and Shooting Accuracy. Children 2023, 10, 199.

AMA Style

Engler F, Hohmann A, Siener M. Validation of a New Soccer Shooting Test Based on Speed Radar Measurement and Shooting Accuracy. Children. 2023; 10(2):199.

Chicago/Turabian Style

Engler, Felix, Andreas Hohmann, and Maximilian Siener. 2023. "Validation of a New Soccer Shooting Test Based on Speed Radar Measurement and Shooting Accuracy" Children 10, no. 2: 199.

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop