Scanning When Passing: A Reliable and Valid Standardized Soccer Test

Abstract

Background/Objectives: In soccer, scanning before receiving the ball helps players better perceive and interpret their surroundings, enabling faster and more effective passes. Despite its importance, no standardized tests currently incorporate scanning actions into assessments of passing abilities. In this study, we test the reliability and validity of a battery of passing tests that assess a player’s ability to control and pass the ball while also scanning for the appropriate target. Methods: We designed three passing tests that reflect different scanning demands that are routinely placed upon players during matches. Using players from the first and reserve teams of two professional clubs in Ghana (Club A, first-team n = 11, reserve-team n = 10; Club B, first-team n = 16, reserve-team n = 17), we: (i) tested the repeatability of each passing test (intraclass correlations), (ii) assessed whether the tests could distinguish between first and reserve team players (linear mixed-effects model), and (iii) examined whether players who were better in the passing tests had higher performances in 3v1 Rondo possession games (linear models). Results: All passing tests were significantly repeatable (ICCs = 0.77–0.85). Performance was highest in the 120-degree test (30.11 ± 7.22 passes/min), where scanning was not required, and was lowest in the 360-degree test (25.55 ± 5.94 passes/min), where players needed to constantly scan behind them. When players were scanning through an arc of 180 degrees, their average performance was 27.41 ± 6.14 passes/min. Overall passing performance significantly distinguished first from reserve team players (β = −1.47, t (51) = −4.32, p < 0.001)) and was positively associated with 3v1 Rondo possession performance (R² = 0.51, p < 0.001). Conclusions: Our results show that these passing tests are reliable, distinguish players across competitive levels, and correlate with performance in possession games. These tests offer a simple, ecologically valid way to assess scanning and passing abilities for elite players.

1. Introduction

Passing the ball is a fundamental skilled action in soccer. Compared to other actions such as shooting, dribbling, or tackling, players pass the ball far more frequently than any other activity during games [1,2,3]. Unsurprisingly, teams that pass the ball more often and more effectively than their opponents tend to win games [3,4,5,6]. Players in elite competitions must be able to rapidly control the ball when receiving a pass, quickly identify a teammate in a better or more advanced position, and then deliver an accurate pass to that teammate. Rapid and effective ball control relies on a player’s ability to receive the ball with either foot while positioning their body correctly to set up a subsequent pass or shield the ball from opponents. The ability to visually scan the surroundings while controlling the ball also enables a player to quickly identify the most appropriate teammate for the next pass. Players who scan their surroundings before receiving the ball tend to make faster subsequent passes [7,8], execute more switches of play and forward passes [9], and successfully complete a greater percentage of passes [10], all of which promote further attacking opportunities. Despite the benefits of effective scanning for attacking play, no standardized test currently incorporates this action into metrics of passing and control. Existing tests do so without involving or depending on scanning [11,12,13,14,15]. For example, Wilson and colleagues [15] designed a battery of passing tests using only paired rebound boards, where players alternated passes between boards that varied in distance or angle. Despite the lack of scanning requirements in this test, higher performances were associated with greater passing success in 3v1 [15] and 4v3 possession games [16]. Similarly, the Loughborough Soccer Passing Test (LSPT) [13] is a valid and reliable assessment of passing in adult male [13], female [17], and youth [18,19] players. It can discriminate between players of different ability [18] and predict those most likely to be selected for elite training programs [20]. In the LSPT, players must pass the ball among four boards while responding to auditory cues that indicate the next target. However, because these auditory cues eliminate the need for players to visually scan their surroundings, the test does not assess scanning as part of passing ability. This absence of scanning in the LSPT may partly explain the lack of correlation between performance on the test and actual passing ability in 11v11 matches [21].

Incorporating scanning into standardized tests of passing and control performance is likely to enhance the ecological relevance of such assessments. Although scanning enables players to perceive and interpret their surroundings—thereby facilitating the rapid selection and execution of the most appropriate pass [7]—it may also negatively affect a player’s ability to effectively control the ball. When receiving a pass from a teammate, players typically track the ball’s trajectory and adjust their body position to intercept it. If scanning requires a player to shift their gaze away from the approaching ball, it can significantly reduce their ability to make subtle adjustments for optimal control. In this sense, scanning functions as a secondary task that may interfere with the performance of the primary task: ball control [22,23,24]. The impact of secondary tasks on primary motor skill performance has been examined across several sports, including ice hockey [25], soccer [22,26], rugby league [23], and netball [24]. Notably, the negative effects of a secondary task are more pronounced in less experienced players. For example, when novice soccer players were asked to dribble a ball through a series of cones while simultaneously completing a visual detection task, their performance declined [26]. However, experienced players were unaffected by the secondary task, suggesting that less skilled players require constant visual contact with the ball, whereas more skilled players exhibit greater automation in ball manipulation, needing less visual input. This automation of skill in high-performing players enables them to perform multiple tasks simultaneously, which is advantageous in team sports as it allows for a broader functional visual field [27]. Consequently, compared to single-task assessments that evaluate only the target motor skill, dual-task assessments may be more effective at identifying elite athletes by better capturing their level of skill automation [27]. It is this idea that forms our motivation for designing a passing test that incorporates the secondary task of scanning.

In this study, we test the reliability and validity of three passing tests that reflect different passing situations that players commonly encounter during games and that vary in complexity and scanning demands. High performance in each test requires the use of both feet and appropriate body positioning when receiving and delivering the pass. The first test required players to alternately pass the ball between two rebound boards. With only one target option per pass and no visual search required, this was considered a single-task test—evaluating the player’s ability to control the ball and deliver an accurate pass. This basic task resembles situations in a match where the player has already predetermined the direction of the next pass before receiving the ball and is simply aiming to execute it quickly and accurately. The second test required players to pass and control the ball among four rebound boards distributed across an arc of 180°. This reflects common match situations, such as when a wide player receives the ball on the sideline and all available passing options lie within a 180° arc. For this test, players had to visually search between each pass to identify the next target, making it a dual-task test. The third test was also a dual-task and required players to control and pass the ball while scanning among three rebound boards evenly spaced around a 360° arc, with the player positioned at the centre. This simulates match situations in which a player receives the ball while scanning and identifying passing options behind them. While the 180- and 360-degree passing tests do not require players to identify the next target before receiving the ball, top performance depends on doing so through scanning. First, we tested the repeatability of each of the three passing tests. Second, we examined whether these tests could discriminate between players of different abilities by comparing the performances of first- and reserve-team players from two professional clubs competing in the same league. Finally, we tested whether performance in the passing tests was positively associated with an individual’s ability to maintain possession in 3v1 Rondo possession games. Rondo possession games require attacking players to make numerous passes in a short period of time to maintain possession against a single defender. These games are used in training to improve the passing and control skills of players [28,29] and can improve players’ problem solving and creativity [30]. Rondo possession games offer repeated exposures to situations commonly encountered during matches. Therefore, a robust assessment of our tests’ construct validity can be achieved by determining the association between performance in our passing tests and 3v1 Rondo performance. We predicted that the passing tests would (i) provide repeatable metrics of individual performance, (ii) differentiate between players from the first and reserve squads of a professional team, and (iii) be positively associated with passing ability in small-sided possession games. In addition to these main objectives, we also predicted that (iv) compared to the single-task test, performance would decrease on both dual-task tests with the introduction of target uncertainty and scanning requirements.

2. Methods

2.1. Subjects

Participants were adult males from the first and reserve teams of two professional soccer clubs that compete in the second division of Ghana’s Accra National Leagues (Club A, first-team n = 11, reserve-team n = 10; Club B, first-team n = 16, reserve-team n = 17). Consent was provided from all individual participants and the study was conducted with institutional ethical approvals.

2.2. Study Design

The passing tests were completed across five one-hour sessions across three consecutive days, with each player attending one session. Each session was attended by players from both teams, from a single club. Players were assessed on three passing tests (120-degree, 180-degree, and 360-degree tests) and had two attempts (trials) on each test, with at least 5 min rest between trials. All players completed the 120-degree test followed by the 180-degree test, then the 360-degree test. We also measured the passing performance of all players from Club A in 3v1 Rondo possession games across three sessions. Although most individuals attended all three Rondo sessions, some players only attended one or two sessions. Of the 40 players who competed in the 3v1 Rondo possession games, 21 individuals also completed the passing tests.

2.3. Passing Tests

We designed three passing tests of varying complexity, based on the number of available passing options and the difficulty of visual search required to determine the next target. All tests used custom-built rebound boards that were 1.1 m wide and 0.5 m tall, each with a 5 cm diameter hole in the centre. Behind each hole, a RoxPro vibration/light sensor (A-Champs Inc., Barcelona, Spain) was mounted, which emitted a light to indicate the next target board based on a pre-programmed sequence.

To begin each trial, the player kicked the ball against the board with the first light illuminated. As soon as the ball hit that board, the light would turn off and a new light would immediately activate on a different board, indicating the next target. This sequence continued until the trial was complete. Players were required to take at least one touch between passes but could choose which foot and technique to use. If a pass missed the target or an error occurred (see below), the player received another ball and continued the trial. The RoxPro system recorded the time elapsed between each successful strike of the boards.

In the 120-degree test, two boards were placed 120° apart, both facing a shared focal point located 5 m from each board (Figure 1). Players began each trial standing at the focal point. The starting board was randomized between players, and each performed ten passes in each direction, excluding the first pass (20 total). As only two boards were used and players always knew the next target, this was considered the least complex test, requiring no visual search. However, strong performance still demanded efficient ball control and accurate passing.

In the 180-degree test, four boards were positioned across a 180° arc, with each adjacent board 60° apart and all facing a common focal point (Figure 1). Each board was 5 m from the focal point. Depending on the light sequence, players had to turn left or right through angles of 60°, 120°, or 180° before executing the next pass. Each type of directional turn was performed three times (18 total). In addition to the technical requirements of the 120-degree test, high performance in this test required players to scan the boards to identify the next target. The most efficient strategy involved scanning immediately after striking the current target and before receiving the ball, enabling a first touch that facilitated a rapid, accurate pass. Alternatively, players could wait until after receiving the ball to scan, but this required more versatile control to allow passing in multiple directions.

In the 360-degree test, three boards were spaced 120° apart around a full circle, all facing a central focal point (Figure 1) and each positioned 5 m away. Players performed ten passes in each direction (20 total). This test incorporated the technical demands of the 120-degree test and required players to identify the next target before receiving the ball. If players delayed scanning until after receiving the ball, their first touch could either help or hinder performance—either positioning the ball toward the next target for a quick pass or requiring adjustment of body or ball position to complete the pass.

All tests were video recorded, allowing each pass to be categorized as a hit, error, or invalid by a single researcher. An invalid pass was any instance where the player executed a first-time pass, and these were excluded from analyses. Errors occurred when the player missed the target boards or mis-controlled the ball and required a replacement ball. Intra-rater reliability on these categorisations was estimated on a random subset of twelve trials (four from each test). With the “psych” package of R, intra-class correlations (two-way mixed effects model) for this subset were estimated as 1.00. Because these errors reflected individual performance, we applied a standard time penalty. Specifically, we used the median pass-time (4.60 s) across all tests and trials where a replacement ball was required. This approach ensured that players were consistently penalized with a time value that reflected the typical duration taken to miss the target (or lose control), receive a new ball, and then successfully complete the next pass.

2.4. 3v1 Games

A 3v1 “Rondo” possession game requires attacking players to make numerous passes in a short period of time to maintain possession against a single defender. For each player, this setup offers repeated exposures to situations commonly encountered during matches.

Players were placed in groups of four, with one player randomly selected to begin as the defender. The game was played within a 5 × 5 m field, where the attacking players started with possession and aimed to maintain it by passing the ball among teammates while the defender attempted to gain possession. If the ball went out of play or the defender won possession, an attacker restarted play by passing the ball to a teammate. Each round lasted 2 min and consisted of four games where each player in the group took a turn as the defender for 30 s. After each round, players were randomly reassigned into new groups of four, and the games continued.

All games were video recorded, and for each player, the number of successful passes and the number of times they committed an error (i.e., lost possession) were recorded by a single researcher. A player was considered to have lost possession if they were tackled by the defender, had a pass intercepted, or took a poor touch that caused the ball to leave the playing area. Intra-rater reliability for these categorizations was estimated on a random subset of 10 games, with data extracted for three attackers from each game. With the “psych” package of R, intra-class correlations (two-way mixed effects model) for this subset were all >0.94.

We developed a performance metric for each individual in each game based on the number of successful passes relative to their number of errors. However, in cases where a player made no errors during a game (e.g., 8 passes and 0 errors), it was not possible to calculate a passes-per-error ratio. To address this, we created a success metric based on each player’s number of successful passes relative to the number of expected errors, which was derived from the average number of passes per error across all players in each of the three sessions. These averages were 7.61 in Session 1, 8.16 in Session 2, and 8.58 in Session 3. Using these values, we calculated each player’s expected number of errors based on the number of passes they made. This allowed us to calculate a success metric for each player in each game, defined as their total number of successful passes relative to their expected errors.

A player’s expected errors in any given game were calculated using the following equation:

Expected errors = [number of passes by individual in the game]/[average passes per error *]

* Average passes per error is calculated across all individuals in a session.

For each player, we then calculated their performance score in each game using the following formula:

Player performance = [Expected errors]−[number of errors]

For example, if during a game in Session 1 a player made 14 passes and committed 1 error, their player performance score would be calculated as follows: (14/7.61) − 1 = 0.84. This score indicates that the player made 0.84 fewer errors than expected, given the number of passes they made. Each player’s performance score was then averaged across all games to generate an overall metric of 3v1 passing performance. We selected this metric because it accounts for the number of passing opportunities each player had during a game and accommodates situations where a player made zero errors. Furthermore, this performance metric aligns intuitively with the results of our passing tests, as higher positive values consistently indicate better performance across all variables.

2.5. Statistical Analysis

First, all pass-times (s) were converted to passes per minute (passes/min). Having each pass quantified in this manner was more intuitive, as larger numbers indicate better performance. For estimates of repeatability on each test (prediction (i)), we first calculated the mean passes/min for each trial, then intra-class correlation coefficients (two-way mixed effects model) were estimated with the “psych” package of Ref. [31]. For each test, we estimated repeatability under two conditions, including only hits, and including both hits and errors.

To estimate correlations among the three passing tests for descriptive purposes, we first calculated a metric of performance for each player in passes/min. To do this for each passing test, we first calculated each player’s average passes/min separately for turns to the left and turns to the right, and the average of these two measures was the performance metric (average-passes/min). Some players were faster in one direction compared to the other, and this protocol accounted for unbalanced individual data caused by removing invalid data points. Pearson correlations were then estimated with the “Hmisc” package of Ref. [32].

To determine if each passing test could discriminate between teams (prediction (ii)), we used a linear mixed-effects model with the “lmerTest” package of Ref. [33]. Using all passes coded as either a hit or error, we estimated the main effects of test (120-degree, 180-degree, or 360-degree test) and team (firsts, reserves) on passes/min, including an interaction. Player ID nested in club and trial number were included as random factors. The “emmeans” package [34] was used for pairwise Tukey comparisons. This model also estimated variance in performance among the three passing tests (prediction (iv)).

To calculate an overall metric of performance on the passing tests for each player, we used Principal Components Analysis (PCA) to characterize variation among the correlated performance metrics from each test. Prior to PCA, each variable was standardized to a mean of zero and a standard deviation of one. The first principal component of passing performance (PC_P1) accounted for 69% of the variation. All vectors were loaded in the same direction, with larger positive numbers indicating better performance. To determine whether the metric of overall passing performance could discriminate between teams (prediction (ii)), we used a linear mixed-effects model to estimate the main effect of team on PC_P1, including club as a random factor.

Separate linear models [35] were used to determine if our metrics of performance for each passing test could predict 3v1 passing performance (prediction (iii)). A linear model was also used to determine if overall performance on our passing tests (PC_P1) could predict 3v1 passing performance (prediction (iii)).

The statistical program R was used for all analyses and alpha levels set at 0.05 for statistical significance.

3. Results

Each of the three passing tests were significantly repeatable and were all positively correlated with each other (

Table 1. Pearson’s correlations among the three passing tests (n = 54).

	120-Degree	180-Degree
180-degree	r = 0.66 *
360-degree	r = 0.47 *	r = 0.46 *

* p < 0.001.

). When excluding passes that missed the target, the 120-degree (ICC = 0.77, CI = 0.63–0.85, p < 0.001), 180-degree (ICC = 0.85, CI = 0.76–0.90, p < 0.001), and 360-degree (ICC = 0.80, CI = 0.69–0.87, p < 0.001) tests were all significantly repeatable. Repeatabilities were lower for all three tests when errors were included and standardized penalties applied: 120-degree (ICC = 0.54, CI = 0.28–0.71, p = 0.003), 180-degree (ICC = 0.61, CI = 0.38–0.75, p < 0.001), and 360-degree tests (ICC = 0.66, CI = 0.47–0.79, p < 0.001).

Players from the first teams executed more passes per minute than reserve-team players on the 120-degree (β = 2.02, t = 4.73, p = < 0.001) and 180-degree tests (β = 2.26, t = 5.12, p = < 0.001), but no differences were detected between the groups in the 360-degree test (β = 0.47, t = 1.11, p = 0.878) (

Table 2. Linear mixed-effect model of passes/min, predicted by passing test and team. Intercept refers to the average first-team player on the 120-degree test.

	Estimate	Std. Error	df	t-Value	p
(Intercept)	30.137	0.490	3.032	61.473	<0.001
180-degree test	−2.659	0.291	6063.633	−9.148	<0.001
360-degree test	−4.565	0.279	6054.511	−16.353	<0.001
Reserve team	−2.020	0.427	98.521	−4.730	<0.001
180-degree test: Reserve team	−0.238	0.410	6061.768	−0.581	0.561
360-degree test: Reserve team	1.547	0.395	6054.682	3.914	<0.001

and

Table 3. Pair-wise comparison from linear mixed-effects model of passes per minute, predicted by passing test and team.

Contrast	Estimate	Std. Error	df	t-Value	p
120 Firsts–120 Reserves	2.02	0.427	98.2	4.725	<0.001
180 Firsts–180 Reserves	2.258	0.441	111	5.121	<0.001
360 Firsts–360 Reserves	0.473	0.427	98.1	1.106	0.878
120 Firsts–180 Firsts	2.659	0.291	6063.5	9.147	<0.0001
120 Firsts–360 Firsts	4.565	0.279	6054.4	16.353	<0.0001
180 Firsts–360 Firsts	1.906	0.291	6063.6	6.563	<0.0001
120 Reserves–180 Reserves	2.897	0.289	6059.3	10.018	<0.0001
120 Reserves–360 Reserves	3.018	0.28	6054.7	10.788	<0.0001
180 Reserves–360 Reserves	0.121	0.289	6059	0.419	0.9994

; Figure 2). First-team players made 30.11 ± 7.22 passes/min on the 120-degree test, while reserve-team players 27.97 ± 7.51 passes/min. On the 180-degree test, the first-team players produced 27.41 ± 6.14 passes/min and reserve-team players 25.16 ± 6.48 passes/min. Players from the first teams were also better performers than players from the reserve team in the overall metric of passing performance (β = −1.47, t (51) = −4.32, p < 0.001) (Figure 2).

For first-team players, the number of passes per minute significantly varied across the three tests, with the highest of 30.11 ± 7.22 passes/min on the 120-degree test and the lowest on the 360-degree test of 25.55 ± 5.94 passes/min (Table 3, Figure 2). For reserve-team players, performance in the 120-degree test was higher than for other tests, but no difference was detected between the 180-degree and 360-degree test (Table 3, Figure 2).

Across all 3v1 Rondo games, individuals made an average of 9.07 ± 2.65 passes and 1.26 ± 1.07 mistakes per game. For those individuals that completed the passing tests (n = 21), players competed in 29.95 ± 7.69 Rondo games while in possession (Range: 13–39). Average performance in the 3v1 games was 0.13 ± 0.36 and varied from the lowest score of −0.69 to a highest of +0.74. Performance in the 3v1 games was positively associated with their performances in the 120-degree (R² = 0.50, β = 0.10, F(1, 19) = 21.21, p < 0.001) and 180-degree (R² = 0.55, β = 0.19, F(1, 19) = 25.53, p < 0.001) passing tests, but not the 360-degree passing test (R² = 0.01, β = 0.08, F(1, 19) = 1.15, p = 0.297) (Figure 3). Overall passing performance (PC_P1) for individuals was also positively associated with their possession performance in 3v1 games (R² = 0.51, β = 0.20, F(1, 19) = 22.14, p < 0.001) (Figure 3).

4. Discussion

Using custom-built rebound boards with integrated light systems, we designed three tests of varying complexity that assessed players’ ability to receive, control, and pass the ball while visually scanning for the correct target. We evaluated whether the passing tests (i) provide a repeatable metric of passing performance, (ii) discriminate between players of differing competitive levels, and (iii) were associated with performances in small-sided possession games. We also tested whether (iv) the presence of multiple targets and scanning demands negatively impacted performance. The passing tests demonstrated strong within-individual repeatability and could distinguish between first- and reserve-team players for two clubs competing in a professional league in Ghana. Passing test performance was also associated with success in 3v1 Rondo possession games. Repeatability was highest when passing errors were excluded, reflecting each player’s “best” possible performance. However, including errors, despite lowering repeatabilities due to trial variability, is likely to be a better representation of an individual’s performance. As expected, we found the presence of multiple targets and increased scanning demands negatively impacted the players’ ability to control and pass the ball.

The 120-degree and 180-degree tests, as well as the composite passing metric, could discriminate between first- and reserve-team players. Unlike typical comparisons between novice and expert players [18,36], our results show discrimination between players at similar competitive levels. In both clubs, the first and reserve teams train together, and players often transitioned between starting teams based on current form. Among the three tests, the 180-degree test showed the greatest performance gap between first and reserve players, with only one reserve-team player possessing a performance that was greater than the average performance of the first-team players. The composite metric of passing performance was also highly effective at discriminating between first- and reserve-team players. In contrast, the 360-degree test did not significantly discriminate between first- and reserve-team players. In this test, players needed to scan behind them before receiving the ball in order to reliably turn in the correct direction to set up the next pass. Few players consistently scanned before receiving the ball when conducting the 360-test, which may explain the lack of differentiation between first- and reserve-team players. This test may be more appropriate for elite professional players or may be an activity in which only central midfielders excel.

Greater target uncertainty and scanning demands decreased passing performance. For example, when comparing the performances of individuals on the 120-degree and 180-degree tests, the introduction of multiple targets and scanning requirements reduced passing rates (passes/min) from 30.11 to 27.41 (Δ = 2.7) for first-team players, and from 27.97 to 25.16 (Δ = 2.81) for reserve-team players. While the reserve-team players experienced a slightly greater performance drop than first-team players, our statistical model indicates that this difference was not statistically significant (Table 2). This may be due to the small sample size, particularly from Club A, which makes it inherently difficult to detect significant differences for small effect sizes. However, the greater performance drop experienced by reserve-team players may help explain why the 180-degree test better differentiated between the first- and reserve-team players. First-team players may have scanned earlier, enabling cleaner first touches and faster passes, or may have more automated motor control, allowing them to scan and execute passes with less performance loss. As with dribbling [26], higher skill levels may reduce the need for visual contact with the ball. In contrast, reserve players may have delayed scanning until after controlling the ball or exhibited poorer control when scanning concurrently—both of which would delay passing. Future video analysis could examine these scanning behaviours and potentially clarify the mechanisms underlying performance differences. Regardless, by incorporating scanning, the tests offer an ecologically valid dual-task assessment capable of identifying skill differentials.

Higher scores on the 120- and 180-degree tests, as well as the overall metric, were associated with better passing performances during 3v1 Rondo games. The 180-degree test alone explained 55% of the variance. Rondo drills, commonly used in training, require attackers to maintain possession by passing quickly around defenders while adjusting body orientation and scanning their environment. Thus, it is unsurprising that players who excelled at our scanning-based passing tests also performed better in this possession-game format. Although the link between 3v1 and full 11v11 match performance remains unclear, prior research supports the use of small-sided games for both training and talent identification [37,38,39].

Our tests quantify a player’s ability to receive, scan, and pass quickly, which is a core skill in both small-sided and 11v11 play. This skill is executed repeatedly across all field positions. For example, a central defender receiving a pass (Figure 1B, board A) must rapidly evaluate and execute a pass to a left winger, central midfielder, or right back (boards B–D). Delays in decision-making provide opponents time to block passing lanes, reducing attacking potential. While we found strong associations between our tests and 3v1 performance, future work should explore links with performances in 11v11 matches. If these tests also predict passing performances in 11v11 matches, they offer a simple, scalable method for coaches to assess a fundamental skill critical to match success and may be adaptable to individual training programs.

One should be mindful that our results are from only one group of athletes. Although we suggest that the metrics of performance produced from these passing tests are likely appropriate for all competitive soccer players regardless of age, sex, or physical stature, this still remains to be tested. Using similar tests on a group of elite players ranging in age from 10 to 20 years old, Camata and colleagues [40] found that age alone was the best predictor of performance, not the combined effect of age, height, and mass. It is therefore unlikely that performance on our tests is confounded by physical attributes such as muscular power or sprinting speed, but this remains to be tested.

5. Conclusions

In summary, we developed three passing tests that assess players’ ability to control and pass while scanning for targets. Individual performance in these tests was correlated with performances in 3v1 Rondo games and could reliably discriminate between first- and reserve-team players in Ghana’s 2nd Division. These tests offer coaches a practical tool to efficiently evaluate a key component of soccer performance across large player pools.