The football throw-in: A scoping review

Abstract

Despite the publication of articles about set pieces in football, research has not provided much focus on throw-in actions. The aim of this review is to offer an overview of the current literature concerning football throw-ins and identify perspectives for future research. Following the PRISMA-ScR recommendations, PubMed, SPORTDiscus and Google Scholar databases were used, and a total of 1676 articles were found. After applying inclusion and exclusion criteria, 33 articles were included in the review. These publications were divided into three main categories: 1) studies on biomechanics, motor learning and training, 2) studies analyzing goals scored from these actions in different competitions and 3) studies analyzing the game and the involvement of throw-ins. This scoping review revealed that throw-ins play an important role in football, as it is the most frequent game interruption in matches, but it seems difficult for teams to retain possession and create goal scoring opportunities after throw-ins. Different variables can influence the outcome of throw-ins, and some key principles could be followed in order to learn and/or improve the throwing motion. Future research should focus on evaluating throw-in performance and finding training methods.

Keywords

Biomechanics motor learning performance analysis set pieces soccer

Introduction

The field of match analysis in football (soccer) provides objective information and feedback on performance enabling coaches to identify collective or individual strengths and weaknesses and notably impact technical behaviour and tactical decision-making.¹ Numerous studies using match analysis have focused on identifying successful models of playing performance including technical, tactical and physical components.² Yet, the most important single factor that determines results in football play is simply and evidently the number of goals scored (or conversely, conceded).

While not always recognized as an essential part of the game, set pieces have enjoyed increased recognition in recent years, particularly after an extremely prolific 2018 World Cup during which over 40% of goals arose from these situations.³ Testament to this is the hiring of specialized set-piece coaches and analysts in addition to a growing number of dedicated scientific studies.⁴ However, coaches and studies tend to focus mainly on corner kicks, free kicks and penalty kicks, arguing that these events lead directly to goal-scoring opportunities⁵ paying less attention to other set pieces such as goal-kicks and throw-ins.

Yet throw-ins are key tactical and technical actions in their own right and can be improved following specific principles and training. Scientific studies initially focused on biomechanical analyses of these actions with the aim to throw the ball as far as possible⁶ and using different techniques.⁷ Findings suggested that throwers should focus on their run-up before launching the ball and aim for a release angle of about 30°. This information can be useful for teams trying to use throw-ins as an attacking weapon by delivering the ball directly into the opponent's 18-yard box. Throwing distance can also be improved by following a specific training program, such as repeating the throwing pattern using either a football ball or a medicine ball over a period of 6 weeks.⁸ Throw-ins have also been studied from a motor learning perspective, highlighting the importance of the frequency and the content of feedback to learn the correct pattern, especially in younger age groups.⁹

The throw-in is also the set piece that occurs the most frequently during a football match.¹⁰ However, these actions rarely lead to a goal being scored; analysis of the 2018 World Cup showed that only ∼1.2% of goals occurred following a throw-in.³ Also, throw-ins can also prove quite problematic for many teams, as ball possession is notably lost by the team executing the throw-in almost one situation out of two.¹¹ In a sport where possession is a key element of performance,¹² it is problematic to lose the ball so frequently when the team has it initially in their possession. Different throw-in strategies are followed by teams, some of which seem more effective than others to retain possession or create goal scoring opportunities (e.g., throwing the ball backwards, taking the throw-in quickly after the ball has gone out of play).¹³ The aim of this work is to review current scientific knowledge on throw-ins by conducting a scoping review of the associated literature. Preferred Reporting Items for Systematic Reviews and Meta-analyses Extension for Scoping Reviews (PRISMA-ScR) recommendations¹⁴ will be followed to ensure a scientific frame. This review can include articles with various participants characteristics (youth, adults, novices, elites) to gather as much information as possible about throw-ins, some of which may be specific to a category. Then, it appears meaningful to conduct a scoping review over a systematic review as the purpose of the review is to scope a body of literature and summarize the evidence around this topic, consisting of studies presenting different methods, approaches and themes.¹⁵ Based on this overview, it will then be possible to identify gaps in the literature and subsequently suggest future avenues for research to inform and eventually help enhance this key feature of the game.

Method

Databases and search strategy

This scoping review followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analyses Extension for Scoping Reviews (PRISMA-ScR).¹⁴ The PubMed, SPORTDiscus and Google Scholar article databases were queried on 04/03/2024.

Articles identified via the various databases had to meet the selection criteria described below to be included in the review:

For PubMed, no filter was applied, and the following equation was used:

(“football” OR “soccer”) AND (“set piece*” OR “set play*” OR “throw in”)

For SPORTDiscus, the “Peer Reviewed” filter was ticked in relation to the selection criteria, and the following equation was used:

(“football” OR “soccer”) AND (“set piece*” OR “set play*” OR “throw in”)

For Google Scholar, the “Review articles” filter was ticked in relation to the selection criteria, and the following equations were used:

(1) soccer “set piece” OR “set play” OR “throw in”.

(2) football “set piece” OR “set play” OR “throw in”

In addition, reference lists of retained articles after full-text screening were examined to identify additional articles not identified by the initial search.

Data extraction and article selection

The results of the database queries were extracted and saved in CSV format spreadsheets (containing author name, title, year of publication, journal and publisher) in order to be processed according to the different selection phases, as illustrated in Figure 1. The first phase consisted of the removal of duplicates. In the second phase, the title screening was conducted manually. Thereafter, the full texts of the remaining studies were screened manually, and inclusion and exclusion criteria presented in Table 1 were applied. This part was conducted by the first author of this review (RN). Finally, the references of the selected articles were examined, and all the articles meeting inclusion criteria were included in this review. The inclusion of the studies was discussed by all the researchers (RN, CC, OD).

Figure 1.

PRISMA flow chart.

Table 1.

Selection criteria for the review.

Number	Inclusion criteria	Exclusion criteria
1	The article studies throw-ins in 11-a-side football (soccer), including the men's, women's and youth games from all levels.	The article studies futsal, small-sided games, small-sided football, or another sport.
2	The article is from a peer-reviewed journal or edited book published in any language. It must be considered a primary source, and therefore reliable and validated.	The article is a systematic review, an article review, a thesis or dissertation, a conference abstract, or any other type of non-evaluated source.
3	The article contains data relevant to the review (e.g., indicators and observation criteria, involvement of throw-ins in the game, biomechanical study, link with training).	The article does not contain data relevant to the review (e.g., no link with throw-ins, no involvement of throw-ins in the game, no biomechanical or game analysis).

Results

Search results

The initial search identified 1687 potentially relevant articles for this review: 48 via PubMed, 1048 via SPORTDiscus and 591 via Google Scholar. After excluding duplicate articles (n = 11), the first stage involved reading the titles of the articles: 1563 articles were excluded as being irrelevant to the research theme (dealing with other sports, review articles, no words related to the topic). The full texts of the remaining 113 articles were then read, and those that did not meet the conditions for inclusion were excluded. The main reason for the exclusion was the absence of data relevant to the study (n = 91), as articles did not contain content about throw-ins and game analysis, motor learning, biomechanics, or just quote “throw-ins” in their text but with no further information. The other was the nature of the articles, some taking the form of an article review paper or a systematic review (n = 8). This resulted in the retention of 14 articles, the references of which were examined to identify additional articles relevant to the study: 18 articles from the references were deemed relevant to the study. These were added to the list of articles retained, bringing the total number of articles included in this scoping review to 32. Finally, an additional article deemed relevant for inclusion was identified by manual search after the database search phase, and so was included in the review, which now comprises 33 articles. The steps involved in selecting the articles are presented in the flow diagram, following the PRISMA-ScR recommendations (Figure 1).

Characteristics of articles

The articles included in this review were published between 1986 and 2025. Figure 2 shows the cumulative distribution of studies by year of publication. It shows that a large proportion of the studies (n = 16, 48%) were published in the last 10 years, reflecting the recent emergence of interest in this specific phase of the game.

Figure 2.

Cumulative distribution of studies included in the review by year of publication.

The studies came from a wide range of countries and cover a variety of topics. Their distribution is shown in Figure 3 below.

Figure 3.

Diagram presenting the themes and origin of the studies included in the review.

Topics of articles

In this section, the results of the various studies will be presented according to the topics to which they relate.

Biomechanics, motor learning and training

The review highlighted studies providing biomechanical analyses of the throw-in to determine the optimal movement patterns to achieve the greatest distance. Maximal throwing distance can be improved by following a specific training program based on the repetition of throws with a soccer or a medicine ball.^8,16 Motor learning of the throw-in movement and the role of different types of feedback has also been investigated, notably in novice players.^9,17 Finally, a study questioned the reproducibility of an offensive pattern starting from a throw-in during training (Table 2).

Table 2.

Articles studying biomechanics, motor learning or training for throw-ins.

Study	Sample	Objective	Protocol	Main results
Messier & Brody (1986)¹⁸	15 university football players (18-21 years old; 1.753 ± 0.02 m; 768.6 ± 11,1 N) 2 teenage footballers (13 years old; 1.580 ± 0.075 m; 434.9 ± 2.7 N)	Compare the conventional throw-in technique and the handspring throw-in technique, while throwing as far as possible	13 players (13 university students) for the traditional throw, 10 trials filmed at 100 frames per second, with the best 3 trials retained. 4 players (2 university students and 2 teenagers) for the handspring throw, 10 trials filmed at 200 frames per second, with the best 3 trials retained	Throwing with the handspring throw-in technique allows one to throw further, thanks to a higher take-off speed
Weeks & Kordus (1998)¹⁷	34 boys (12.3 years old on average) not used to the throw-in movement	Compare the effects of varying the frequency of knowledge of performance (feedback) on the learning process of the throw-in movement	2 groups of 17 participants, 30 throws in the acquisition phase, then 5 throws in the testing phases 5 minutes, 24 hours and 72 hours after acquisition. During the acquisition phase, a group receives feedback after each trial (100%), the other every 3 trials (33%)	The 2 groups obtained similar results in the tests, but the group receiving knowledge of performance less frequently retained better movement during the testing phase
Ferrario et al. (1999)¹⁹	2 teams of young players (15-17 years), one semi-professional (junior national championship) and one amateur	Study players’ ability to reproduce attacking patterns	2 offensive patterns, 25 repetitions per pattern for the semi- professional team (5 sets of 5 repetitions) and 10 repetitions per pattern for the amateur team (2 sets of 5 repetitions) Video recording using a camera Both teams trained those patterns for 2 weeks prior to registration	The semi-professional team repeated the 2 patterns more precisely than the amateur team. Nevertheless, the amateur team reproduced the pattern 1 correctly (pattern starting with a throw-in)
Lees et al. (2005)⁷	10 football players accustomed to performing throw-ins (22 ± 1.78 years; 1.779 ± 0.059 m; 75.2 ± 6.1 kg)	Study movement and forces in the upper limb during throws seeking to reach the greatest distance, with and without run-up	5 throws with a run-up, 5 throws without a run-up (with the 2 feet staggered / not parallel). Indoor throws with biomechanical markers and motion capture system	Throwing with a run-up allows the ball to take off at greater speed, and thereby reach a greater distance
Linthorne & Everett (2006)⁶	1 university footballer (21 years old; 1.77 m)	Find the release angle to achieve the greater distance when performing a throw-in	5 throws at the thrower’s preferred angle, then 20 throws at higher and lower angles	The launcher gives the ball greater speed at a release angle of around 30°
Van Den Tillaar & Marques (2009)¹⁶	64 sport science students (46 men; 18 females; 21.1 ± 2.1 years; 71.1 ± 11 kg; 1.75 ± 0.09 m) novices to this type of training	Determine whether two different throwing training programs increase the speed of the throw	2 mixed groups: one carrying out 2 sets of 8 throws with a 5 kg medicine ball per session, the other doing 6 sets of 14 throws with a football. 2 sessions per week for 6 weeks. Tests before and after 6 weeks	The 2 groups increased their throwing speed at the end of the 6-week program: +3.2% for the football group and +5.1% for the medicine-ball group
Wulf et al. (2010)⁹	48 children (18 boys; 30 girls; aged 10-12)	Measure the effects of feedback frequency and feedback nature (internal vs. external) on learning the throw-in movement	4 groups of 12 participants, 2 groups with internal feedback and 2 groups with external feedback. For each type, 1 group each with feedback after each attempt, the other every 3 attempts. 30 throws in the acquisition phase, 5 throws without feedback in test phase just after and 24 hours after the acquisition phase	Externally oriented feedback given after each attempt enables better motor learning than when given every 3 attempts. Compared to feedback with an internal orientation, feedback with an external enabled retention of a better movement during the maintenance and test phases. No significant differences were observed in terms of throwing accuracy
Van Den Tillaar & Marques (2011)⁸	63 students (39 females; 24 men; 16.5 ± 1.8 years old; 57.8 ± 12 kg; 1.646 ± 0.09 m) new to this type of training	Determine whether different throwing training programs increase the speed of the throw	3 mixed groups: one performing 3 sets of 6 throws with a 3 kg medicine ball per session, another 6 sets of 14 throws with a football, the last 9 throws with a 3 kg medicine ball and 3 sets of 14 throws with a football. 2 sessions per week for 6 weeks. Tests before and after the 6 weeks	The 3 programs resulted in an increase in throwing speed after 6 weeks/ + 8.5% for the group with the football, +5% for the group with the medicine ball and +7.1% for the group with the combined program

Goals arising from throw-ins

The various studies aim to analyze how goals are generally scored in competitions, including attacking plays, time periods, area or body part used. Regarding set pieces, a large proportion of the total number of goals were shown to arise from these actions (17.4% to 42%), albeit with a small percentage arising from throw-ins (0.1% to 7.1%) (Table 3).

Table 3.

Articles analyzing goals scored in various international and domestic competitions.

Study	Competition	Number of matches / goals	Main results
Yiannakos & Armatas (2006)²⁰	2004 UEFA European Championship in Portugal	32 matches	35.6% of the goals scored from set pieces, including 1.8% from throw-ins
Acar et al. (2008)²¹	2006 FIFA World Cup in Germany	64 matches / 147 goals	66 goals (39.1%) scored from set pieces, including 6 from throw-ins (4.1%)
Wright et al. (2011)²²	English Premier League Football Championship	169 goals, 1788 attempts	60 goals (36%) scored from set pieces, including 12 from throw-ins (7.1%). 152 goal attempts came from throw-ins (8.5%)
Njororai (2013)²³	2010 FIFA World Cup in Africa South	64 matches / 145 goals	35 goals (24.1%) scored from set pieces, including 1 from throw-ins (0.7%)
Mitrotasios & Armatas (2014)²⁴	2012 UEFA European Championship in Poland and Ukraine	31 matches / 76 goals	21 goals (27.6%) scored from set pieces, including 1 from throw-ins (1.3%)
Göral (2019)³	2018 FIFA World Cup in Russia	64 matches / 169 goals	68 goals (40.2%) scored from set pieces, including 2 from throw-ins (1.2%)
Vergonis et al. (2019)²⁵	2018 FIFA World Cup in Russia	64 matches / 169 goals	70 goals (41.4%) scored from set pieces, including 2 from throw-ins (1.2%)
Kubayi (2020)²⁶	2018 FIFA World Cup in Russia	64 matches / 169 goals	66 goals (39.1%) scored from set pieces, including 2 from throw-ins (1.2%)
Vergonis et al. (2021)²⁷	2018 FIFA World Cup in Russia	64 matches / 169 goals	71 goals (42.0%) scored from set pieces, including 2 from throw-ins (1.2%)
Anzer et al. (2021)²⁸	German First and Second Division Football Championship, 2017/2018 and 2018/2019 seasons	136 matches / 3457 goals	1277 goals (36.9%) scored from set pieces, including 20 from throw-ins (1.6%)
Akyildiz et al. (2022)²⁹	Turkish First Division Football Championship, 2015/2016, 2016/2017 and 2017/2018 seasons	918 matches / 2559 goals	446 goals (17.4%) scored from set pieces, including 2 from throw-ins (0.08%)

Game analysis

Studies demonstrate that throw-ins are the most frequent game interruptions and are used by teams as a means to buy time.^33,35 From a tactical point of view, the outcomes of throw-ins are influenced by a range of variables such as location, direction or opposition pressure (Table 4).

Table 4.

Articles dealing with game analysis.

Study	Sample	Objective	Indicators / Variables	Main results
Taylor et al. (2004)³⁰	22 league and cup matches for one professional British team, 2002/2003 season	Create position-specific performance profiles for professional football players	Aerial duels, clearances, crosses, dribbles, interceptions, lost balls, passes, tackles for, tackles against, shots, penalties, corners, throw-ins, direct free kicks, indirect free kicks and fouls.	Full-backs are the players who perform most throw-ins (14.1 ± 4.1 on average per match), followed by central defenders (1.4 ± 3.1), forwards (1.0 ± 1.0) and midfielders (0.5 ± 0.4)
Taylor et al. (2008)³¹	40 league matches for a professional British team, 2002/2003 and 2003/2004 seasons	Study the effects of the match venue, the quality of the opposition and the score on technical aspects of the performance	Aerial duels, clearances, crosses, dribbles, interceptions, lost balls, passes, shots, tackles for, tackles against, corners, free kicks and throw-ins.	The frequency of throw-ins does not depend on the location of the match, the quality of the opposition or the score. However, the outcome of throw-ins is dependent on these variables
Siegle & Lames (2012)³²	16 matches in the German first division, 2009/2010 season	Identify game interruptions and their use as a tactical means	Type of interruption, number, duration, location, score, period	Throw-ins are the most frequent game interruptions (40 per match on average). Their average duration is 8.56 ± 5,80 s
Wallace & Norton (2014)¹⁰	12 World Cup finals between 1966 and 2010	Quantify changes in the structure of the game, speed of play and patterns of play	Type, number and duration of game interruptions, ball speed, number and accuracy of passes, player density, effective playing time	Between 26 and 50 throw-ins observed per match (37 on average), with the number being quite constant over the years. Their average duration is between 8 and 12 s and has tended to increase over time.
Augste & Cordes (2016)³³	32 knockout matches at the 2006 and 2014 FIFA World Cups	Examine the use of game interruptions as a tactical mean	Type, number and duration of game interruptions, effective playing time, additional time, score and location	Throw-ins are the most frequent game interruptions (average 44.3 ± 9.4 in 2006 and 44.8 ± 9.6 in 2016). Their average duration was 10.2 ± 5.4 s in 2006 and 11.5 ± 7.3 s in 2014
García-Paúl et al. (2019)³⁴	23 semi-professional footballers from Spain (4^th division)	Investigate attacking throw-ins and crosses received in official matches	Throw-in outcome, player losing the ball, type of loss, time of loss, direction, area of loss, continuation, area	64% of throw-ins result in a loss, which occurs mostly during the following actions (not on the throw)
Greve et al. (2019)³⁵	2231 matches in a European first division, seasons 2009/2010 to 2014/2015	Characterize the time saved by teams during game interruptions	Type, number and duration of game interruptions, score, team level	Teams take longer to play throw-ins when leading than when drawing or losing
Velasco et al. (2019)³⁶	40 Champions League and Spanish league matches from first to third division (10 matches per category), season 2014/2015	Analyse the number and the influence of stoppages on the game	Game interruptions type and number	There were 51.2 ± 8.2 throw-ins per match, which represent between 35.1% and 46.3% of stopped phases, with an average of 124.5 game interruptions per match.
Håland et al. (2020)³⁷	960 matches in the Norwegian Premier League, seasons 2014 to 2017	Evaluate players’ passing ability through a series of models	Number of passes in the sequence, pass type, previous phase, event type, location and player	Throw-ins were deemed difficult to negotiate, since they are the subject to a lot of pressure from the opposition.
Augste & Prestel (2021)³⁸	265 throw-ins from the German first division championship, 2019/2020 season	Study the tactical actions performed during opponent's throw-ins	Location, direction, intensity of pressure, outcome, continuity of play, period, score	54.0% of throw-ins are retained. They are played predominantly forward in the low and mid third. Applying big pressure helps to recover the ball.
Stone et al. (2021)¹¹	16154 throw-ins from 380 matches in the English Premier League, 2018/2019 season	Determining the link between throw-ins and performance, and the parameters influencing reception, possession retention and goal-scoring opportunities creation	Ranking, direction, length, location, reception, retention, retention time, creation of goal-scoring opportunities	Top-ranked teams are the most effective on throw-ins. 83% of throw-ins were received by a teammate, 54% were retained and 8.8% led to a shot
Casal et al. (2023)⁵	2658 throw-ins from 80 Spanish league matches	Identify tactical indicators linked to the outcome of the throw-in, predicting their outcome and analysing the influence of certain variables on their effectiveness	Match venue, location, ranking, result, score, period, duration, pressing, mobility, length, direction, side and outcome	Throw-ins played quickly, out of pressure, short and backwards present the highest probability of retaining possession
Epasinghege Dona & Swartz (2024)³⁹	237 matches in the Chinese Premier League, 2019 season	Investigate the causal relationship between certain variables on the outcome of throw-ins, and determine the optimum target to aim for	Teams, thrower, direction, length, outcome, location of the throw and the reception, period, score, pressure intensity, betting odds, red cards	Lower-ranked teams throw more forward while higher-ranked teams throw less forward and more backwards. Losing teams use more long lateral throw-ins
Stone et al. (2025)¹³	71220 throw-ins from 1826 matches across five European leagues, season 2022/2023	Examine throw-in strategy used by teams and how that strategy affected first contact success, possession retention, and attacking outcomes	Direction, length, time out of play, reception, retention, retention time, shot creation, match state.	Throwing the ball quicker and backwards increases chances of successful first contact, retaining possession and creating goal scoring opportunities

Discussion

The aim of this work was to review current scientific knowledge on throw-ins in football to identify gaps in the literature and subsequently suggest future avenues for research. Main findings from this scoping review show that i) the quantity of research on this topic is increasing, with a major increase in the number of studies published in the last five years, ii) studies tend to focus on a range of themes, such as biomechanics, training interventions, match analysis and goal scoring, iii) the throw-in movement can be improved by learning and using specific techniques, and by following a training program based on throw repetitions, iv) even if a large proportion of goals scored arise from set pieces, only a small percentage of these are from throw-ins, v) throw-in outcome (ball retention, goal-scoring opportunity creation) is influenced by a range of factors.

Biomechanics, motor learning and training

In the present scoping review, the earliest studies that were published more than 15 years ago examined the biomechanics of throw-in actions. Messier and Brody¹⁸ notably compared throws performed using the conventional technique and those using a handspring. They found that there is an advantage to throwing with a handspring, as this technique allows the ball to take off at a higher speed, thereby increasing the length and radius of the throw. As such, the ball can reach teammates or zones further away, thereby making it more difficult for the opponent to defend, especially as the offside rule does not apply during this phase. Lees et al.⁷ focused on the movement patterns of the upper limbs during the throw-in and also identified the usefulness of performing a run-up before executing the action. This enables improved propulsion and therefore a higher take-off speed, a factor that greatly influences the length of the throw. Finally, Linthorne and Everett⁶ determined that the optimum ball release angle seems to be around 30° to the horizontal, although this value varies from one thrower to another (depending on individual body shape and technique). In addition, the authors discussed giving a backspin effect to the ball to produce a lifting force, thereby increasing the length of the throw. They also mentioned the possibility of reducing the ball flight time by reducing the flight angle, which could surprise the opponent and create goal scoring opportunities and recommended players improve their release speed by developing explosive strength and throwing techniques. However, two of these studies were conducted in male collegiate players^6,18 and it is questionable whether such results would be observable in players at a higher level. Finally, these studies only consider the movement required to reach the greatest possible distance, and do not focus on the movement required to precisely attain a target, such as a teammate.

Regarding studies conducted on motor learning, Weeks and Kordus¹⁷ investigated the effects of variations in the relative frequency of feedback on the learning of the throw-in skill in children. Feedback was based on kinematic aspects of the movement pattern. The authors reported that receiving feedback less frequently (after one attempt out of three) during the acquisition phase enabled novices to retain a better pattern of movement for the test phases. In contrast, there was no significant difference in terms of accuracy than when receiving feedback more frequently (after each attempt). This may be explained by the fact that feedback following each attempt leads the participant to focus on correcting a precise aspect of the movement and thus modifying their motor pattern, whereas in the case of less frequent feedback they may focus on trying to apply the correction during several successive trials. Following on from this study, Wulf et al.⁹ studied the difference between two types of feedback (internal vs external) and provided at two frequencies (after each attempt vs after one attempt out of three), also in novice children aiming to learn the throw-in skill. They revealed that receiving externally oriented feedback (focus on the movement effect) after each trial led to improved motor learning of movement than when receiving externally oriented feedback every 3 trials. Another result was that when compared with the groups receiving internally oriented feedback (focus on the performer's body movements), the groups receiving externally oriented feedback maintained better movement during the maintenance and test phases. In contrast, no significant difference was reported for the accuracy of the throws. Hence, frequency and type of feedback seem to influence motor learning of the throw-in skill. Nevertheless, this is only observable at the movement level with no notable difference being identified regarding the precision of the throw, and the populations in these studies were comprised of novice children, so it is difficult to extrapolate these results to other populations.

Regarding training interventions, Ferrario et al.¹⁹ compared the ability of young semi-professional and amateur players to reproduce two attacking action patterns, one starting from a throw-in, and other starting from a right-wing position. Patterns were performed in training without opposition. The authors concluded that the semi-professional team was able to repeat the 2 patterns more accurately than the amateur team, but that the amateur team was nevertheless able to reproduce correctly the attacking pattern starting with a throw-in. They explained this by the fact that this pattern was easier to reproduce than the other pattern, which was more technically and physically demanding. The repetition of empty attacking patterns starting with a throw-in thus seems appropriate for players to retain defined positions and movements.

Results showed that it also seems possible to develop throwing speed via different physical training programs using either a medicine ball (5 kg or 3 kg), a traditional football or both: Van Den Tillaar and Marques^8,16 compared different training programs where 2 sessions were performed per week over 6 weeks. The program was based on the repetition of throws and induced the same workload within the groups. Tests were performed before and after the 6-week follow-up programs, and the intervention showed an improvement in throwing speed, but there was no significant difference between the groups. The authors confirmed the importance of a sufficient and appropriate workload to improve throwing speed. Once again, these experiments were conducted in novices (students), so it is difficult to determine whether following these types of programs would produce the same results in other populations.

Goals arising from throw-ins

Articles analyzing goals scored during competitions (Table 3) show that between 17.4% and 42% of total goals are scored from set pieces. Of these, only 0.08% and 7.1% of goals arise from throw-in events. Regarding the number of goals scored following a throw-in as a proportion of the total number of goals scored at set pieces, this accounts for between 0.5% and 20% of set piece goals. Yiannakos and Armatas²⁰ indicated in an early article that goals scored from throw-ins accounted for 8% of the goals scored from set pieces at the 1982 World Cup and 9.75% of the goals scored from set pieces at the 2002 World Cup. Wright et al.²² reported that in the English Premier League the ratio of goals scored to goal attempts from throw-ins was close to those from corners and free-kicks (1:12.6 from throw-ins, 1:11 from corners and 1:12 from free-kicks). The authors also pointed out that for the 2006 World Cup, teams reaching the semi-finals of the competition were more likely to score from set pieces (ratio of 1:7.5) than teams who were eliminated (ratio of 1:14). The proportions vary considerably from one study to another and seem to indicate a lack of uniformity in observations and analyses: 4 studies notably examined the same competition (the 2018 World Cup), and each reported a different number of goals from set pieces (66 to 71 over the 169 goals scored).^3,^25–27 Moreover, no study has concretely defined what constitutes a goal scored following a set piece and what criteria are used to define it as such (e.g., a number of seconds after the execution of the set piece, a number of passes, if goals scored during the second phase of the set piece are counted).

Game analysis

Most studies relating to this theme are among the most recent to be included in the review, indicating that game analysis is a topical area of research. These generally concur on the number of interruptions (108 to 119), the number of throw-ins (40 to 51) and the average duration of throw-ins (8 to 13 s on average) taking place during matches. They also agree on the fact that throw-ins are used as a means of buying time by teams leading or obtaining a favorable result against a theoretically superior team, with the duration of these actions increasing significantly towards the end of the match.^10,11,32,35

A consensus seemingly exists regarding the variables influencing the outcome of throw-ins: taking quick, short, and backward throws increases the likelihood of the ball being received, conserved and scoring chances to be created. The location of the throw-in, the intensity of opposition pressure, the level of the team, the period, the score, the location of the match and the time taken to put the ball back into play also influence the outcome of throw-ins.^{5,11,13,34,38,39} Studies show that from a defensive point of view, it makes sense to prevent the opponent from playing laterally or towards the back and to apply strong pressure in order to maximize the chances of recovering the ball.^{5,11,13,37,38}

Research by Taylor et al.³⁰ showed that the majority of throw-ins are performed by full-backs. As such, it would seem useful to work specifically with these players to enhance throwing technique and tactical plays, via specific training interventions. The authors report successful throw-in rates of between 70% and 82% for full-backs and between 52% and 100% for centre-backs, and that throw-in success is dependent on the location of the match, the quality of the opposition or the score.^30,31 Some studies also used a random sample and did not include all throw-ins or all goals in a competition. Depending on the sample selected, this can alter the proportions of goals scored from set pieces or throw-ins, as well as the rates of the different outcomes following a throw-in.^5,28,34,36 This limitation may explain the differences observed: Stone et al.^11,13 respectively report a 46% and 36,7% loss of possession in a 7-s period following a throw-in while Casal et al.⁵ report a 37% loss with the same definition, and García-Paúl et al.³⁴ a 64% loss, although the standards of the leagues investigated were different.

It is also possible that the parameters used to define the different throw-in outcomes were not the same. Stone et al.^11,13 and Casal et al.⁵ proposed and used identical definitions of the variables analyzed and the outcomes of throw-ins: for example, Possession was defined as “The team retained the ball in possession for 7 s from the point in which the ball was thrown” and Shot Creation defined as “When a player attempted a shot at goal, which resulted from the throw-in possession”. These definitions should be reused in future studies to extend knowledge on this topic, although arguably some of them need to be developed further for greater clarity and reproducibility.

Limitations

We aimed to conduct this scoping review as rigorously as possible, following the precise and clear recommendations of reference publications.¹⁴ Nevertheless, other databases such as Web of Science, Web of Knowledge or Scopus could have been queried in order to obtain more articles on the topic, and querying the databases in other languages, such as French, Spanish or Portuguese, might have yielded more articles.

Perspectives

A number of research questions and directions have emerged from the present work. Stone et al.^11,13 suggest that future studies should investigate i) explain what constitutes a successful throw-in and why throwing sideways and backwards results in a higher percentage of success, ii) which behaviors lead to a loss and iii) whether the score line, individual match outcome or the team's rank is influenced by throw-in strategy. In line with the last suggestion, determining how throw-ins can influence the overall performance of a team could give further information to coaches about which strategies are pertinent.

Finally, even if studies included in this scoping review present guidelines to perform better on throw-ins, the lack of scientific knowledge concerning throw-ins and set pieces training from both a collective and individual point of view suggests that this aspect deserves to be explored in greater depth, particularly with experienced populations. Following on from this work, and in the event that a training plan is put in place, it could be relevant to measure the effects of throw-in training on goals scored and conceded, in order to observe whether training is effective and whether it modifies the outcomes of successive actions following a throw-in.

Conclusion

The aim of this scoping review was to examine the current scientific knowledge related to the football throw-in. Following PRISMA-ScR recommendations, the PubMed, SPORTDiscus and Google Scholar databases were queried and 33 relevant articles were identified and included in the review. The articles covered a wide range of topics, including biomechanics, motor learning and training, as well as goal analysis and game analysis. This topic seems to have gained in importance recently, with a large proportion of the articles published in the last five years.

Analysis of the various articles shows that set pieces and throw-ins play an important role in football: in fact, throw-ins are the most frequent set piece following a game interruption. The articles also present detailed variables influencing the outcome of throw-ins, such as the distance and direction of the throw, the context of the match, the time between the ball leaving play and the throw-in, and the area of the pitch. Finally, key information is provided for practicing throw-ins and achieving a more effective throwing motion, as well as shedding light on the best options for both attacking and defensive throw-ins.

Footnotes

Acknowledgments

Not applicable

ORCID iDs

Romain Nony

Christopher Carling

Olivier Degrenne

Ethical considerations

This article does not contain any studies with human or animal participants.

Consent to participate

Not applicable

Consent to publication

Not applicable

Author contributions

RN contributed to the design, search strategy, interpretation of the results, and writing up of the paper. CC contributed to the interpretation of the results and writing up of the paper. OD contributed to the design, search strategy, interpretation of the results, and writing up of the paper. All authors read and approved the final version of the manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data used is available here:

References

McRobert

Fernández-Navarro

Seth

. Technical and tactical match analysis. In: Williams

Ford

Drust

(eds) Science and soccer. 4th ed. Oxon: Routledge, 2023, pp.273–291.

Sarmento

Marcelino

Anguera

, et al. Match analysis in football: a systematic review. J Sports Sci 2014; 32: 1831–1843.

Göral

. The importance of set-pieces in soccer: Russia 2018 FIFA world cup analysis. J Hum Sc 2019; 16: 885–893.

Sarmento

Clemente

Araújo

, et al. What performance analysts need to know about research trends in association football (2012–2016): a systematic review. Sports Med 2018; 48: 799–836.

Casal

Armatas

Losada

, et al. Effects of tactical dimension and situational variables in throw-ins on the offensive performance in football. PLoS ONE 2023; 18: e0294317.

Linthorne

Everett

. Soccer: release angle for attaining maximum distance in the soccer throw-in. Sports Biomech 2006; 5: 243–260.

Lees

Kemp

Moura

. Biomechanical analysis of the soccer throw-in with a particular focus on the upper limb motion. In: Reilly

Cabri

Araújo

(eds) Science and football V: the proceedings of the fifth world congress on sports science and football. 1st ed. Oxon: Routledge, 2005, pp.89–94.

Van Den Tillaar

Marques

. A comparison of three training programs with the same workload on overhead throwing velocity with different weighted balls. J Strength Cond Res 2011; 25: 2316–2321.

Wulf

Chiviacowsky

Schiller

, et al. Frequent external-focus feedback enhances motor learning. Front Psychol 2010; 1: 190–190.

10.

Wallace

Norton

. Evolution of world cup soccer final games 1966–2010: game structure, speed and play patterns. J Sci Med Sport 2014; 17: 223–228.

11.

Stone

Smith

Barry

. The undervalued set piece: analysis of soccer throw-ins during the English premier league 2018–2019 season. Int J Sports Sci Coach 2021; 16: 830–839.

12.

Jones

James

Mellalieu

. Possession as a performance indicator in soccer. Int J Perform Anal Sport 2004; 4: 98–102.

13.

Stone

Kubayi

Smith

, et al. Analysis of throw-ins strategy on performance metrics in five men’s European football leagues. Int J Perform Anal Sport 2025; 25: 1013–1027.

14.

Tricco

Lillie

Zarin

, et al. PRISMA Extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med 2018; 169: 467–473.

15.

Munn

Peters

MDJ

Stern

, et al. Systematic review or scoping review? Guidance for authors when choosing between a systematic or scoping review approach. BMC Med Res Methodol 2018; 18: 143.

16.

Van Den Tillaar

Marques

. Effect of two different training programs with the same workload on soccer overhead throwing velocity. Int J Sports Physiol Perform 2009; 4: 474–484.

17.

Weeks

Kordus

. Relative frequency of knowledge of performance and motor skill learning. Res Q Exerc Sport 1998; 69: 224–230.

18.

Messier

Brody

. Mechanics of translation and rotation during conventional and handspring soccer throw-ins. Int J Sport Biomech 1986; 2: 301–315.

19.

Ferrario

Sforza

Dugnani

, et al. Morphological variation analysis of the repeatability of soccer offensive schemes. J Sports Sci 1999; 17: 89–95.

20.

Yiannakos

Armatas

. Evaluation of the goal scoring patterns in European championship in Portugal 2004. Int J Perform Anal Sport 2006; 6: 178–188.

21.

Acar

Yapicioglu

Arikan

, et al. Analysis of goals scored in the 2006 world cup. In: Reilly

Korkusuz

(eds) Science and football VI. Oxon: Routledge, 2008, pp.261–268.

22.

Wright

Atkins

Polman

, et al. Factors associated with goals and goal scoring opportunities in professional soccer. Int J Perform Anal Sport 2011; 11: 438–449.

23.

Njororai

WWS

. Analysis of goals scored in the 2010 world cup soccer tournament held in South Africa. J Phys Educ Sport 2013; 13: 6–13.

24.

Mitrotasios

Armatas

. Analysis of goal scoring patterns in the 2012 European football championship. Sport J 2014: 1–11.

25.

Vergonis

Michailidis

Mikikis

, et al. Technical and tactical analysis of goal scoring patterns in 2018 FIFA World Cup in Russia. Facta Univ Ser Phys Educ Sport 2019; 17: 181–193.

26.

Kubayi

. Analysis of goal scoring patterns in the 2018 FIFA World Cup. J Hum Kinet 2020; 71: 205–210.

27.

Vergonis

Balasas

Michailidis

, et al. The significant role of scoring from set plays in the 2018 FIFA World Cup. J Sports Med Phys Fitness 2021; 61: 47–51.

28.

Anzer

Bauer

Brefeld

. The origins of goals in the German Bundesliga. J Sports Sci 2021; 39: 2525–2544.

29.

Akyildiz

Nobari

González-Fernández

, et al. Variations in the physical demands and technical performance of professional soccer teams over three consecutive seasons. Sci Rep 2022; 12: 2412.

30.

Taylor

Mellalieu

James

. Behavioural comparisons of positional demands in professional soccer. Int J Perform Anal Sport 2004; 4: 81–97.

31.

Taylor

Mellalieu

James

, et al. The influence of match location, quality of opposition, and match status on technical performance in professional association football. J Sports Sci 2008; 26: 885–895.

32.

Siegle

Lames

. Game interruptions in elite soccer. J Sports Sci 2012; 30: 619–624.

33.

Augste

Cordes

. Game stoppages as a tactical means in soccer – a comparison of the FIFA World Cups™ 2006 and 2014. Int J Perform Anal Sport 2016; 16: 1053–1064.

34.

García-Paúl

ÁJ

Valera-Sánchez

Castillo-Rodríguez

. Evaluación de los centros laterales en contra y saques de banda a favor en fútbol semiprofesional. Estudio de caso. Rev Iberoam Cienc Act Física El Deporte 2019; 8: 73.

35.

Greve

Rudi

Walvekar

. Strategic rule breaking: time wasting to win soccer games. PLOS ONE 2019; 14: e0224150.

36.

Velasco

JMI

Morán

Porres

, et al. Análisis de acciones a balón parado en el fútbol profesional español y Liga de Campeones. Lect Educ Física Deport 2019; 24: 49–65.

37.

Håland

Wiig

Stålhane

, et al. Evaluating passing ability in association football. IMA J Manag Math 2020; 31: 91–116.

38.

Augste

Prestel

. Tactical actions during opponent throw-ins in soccer in the German Bundesliga. J Phys Educ Sport 2021; 21: 1846–1852.

39.

Epasinghege Dona

Swartz

. Causal analysis of tactics in soccer: the case of throw-ins. IMA J Manag Math 2023; 35: 111–126.