Goalkeepers benefit from using deceptive actions in the soccer penalty kick

ABSTRACT

Most studies on deception in soccer penalty kicks have focused on the deceptive actions used by penalty takers. However, it is worth noting that deception can also be played out by goalkeepers. To examine the effectiveness of goalkeepers’ deceptive actions in professional competition, we analysed 714 penalty kicks taken during matches in the English Premier League and German Bundesliga, spanning the seasons from 2016–2017 to 2019–2020. We scored whether goalkeepers used deception, and if so, what type of deception, the outcome of the penalty and the kicking strategy of the penalty taker. The results showed that goalkeepers used deception in half of the penalty kicks, resulting in significantly less goals compared to penalties without deception. This advantage was similar for the different types of deception, but larger when penalty takers paid attention to goalkeepers. We propose that the deceptive actions by goalkeepers are effective, mainly because it leads the penalty taker to lose focus. The practical implications of these findings are discussed for both goalkeepers and penalty takers.

KEYWORDS:

• Penalty kick
• deception
• attention
• affordance
• penalty kick strategy

Introduction

The penalty kick plays a pivotal role in professional soccer or association football. Notably, both during the European Championship 2020 and the FIFA World Cup 2022, the champions – Italy and Argentina, respectively – emerged victorious via penalty shootouts. In the encounter between penalty taker and goalkeeper, one way to obtain the advantage is by deceiving the opponent, inducing false perceptions of the forthcoming actions (Zheng et al., 2023). Most of the research on deception in soccer penalty kicks focused on the deceptive actions produced by penalty takers’ and goalkeepers’ susceptibility to those actions (for reviews, Guldenpenning et al., 2017; Jackson & Cañal-Bruland, 2019; Zheng et al., 2023). Yet, deception also manifests itself in the opposite direction. Goalkeepers can and do use deceptive actions to influence penalty takers’ decisions and aiming, but research is relatively scarce (cf. Furley et al., 2017; Masters et al., 2007; van der Kamp & Masters, 2008). The current study aimed to examine what deceptive actions professional goalkeepers produce during penalty kicks in competition and to evaluate the degree to which they are beneficial.

Researchers in sport science literature have not reached consensus on how to define and classify the various types of deceptive action. On the one hand, Guldenpenning et al. (2017) identified two mutually exclusive types, misleading and disguise, which they collectively refer to as deceptive actions (also see Panten et al., 2019; Zheng et al., 2023). On the other hand, Jackson and Cañal-Bruland (2019) have argued that deception, which equates to misleading in the classification of Guldenpenning et al. (2017), must be considered a subset of disguise. We do not intend to resolve this classification controversy. Instead, we follow Guldenpenning et al. (2017) and use deceptive actions as a generic term. Further to this point, we also draw on research in magic that demonstrates that magicians deceive the audience by directing attention away from critical action areas (G. Kuhn & Tatler, 2005; Phillips et al., 2015) and refer to this as misdirection. Consequently, we distinguish three types of deceptive actions, misleading, disguise and misdirection (Zheng et al., 2023).

Using ecological psychology as our theoretical backdrop, we take deceptive action as an attempt to actively hinder the opponent’s perception of the true affordances or possibilities for action by controlling the information that the opponent attends to (Ramsey et al., 2022; Zheng et al., 2023). Accordingly, misleading actions (including feints) are understood as directing the opponent’s attention to false affordances or possibilities for actions. Disguise involves concealing the true possibilities for action, while misdirection refers to redirecting the opponent’s attention away from specifying information to less relevant, non-specifying information that makes the perception of the possibilities for action imprecise (Zheng et al., 2023).

A few laboratory studies have examined how deceptive actions of the goalkeeper can affect penalty taker actions (e.g., Masters et al., 2007; van der Kamp & Masters, 2008; Weigelt et al., 2012), albeit that the researchers did not explicitly refer to “deception”. For example, in studying the aiming of penalty takers (in handball), van der Kamp and Masters (2008) reported that goalkeepers were perceived taller and shorter than their actual height when they held their arms raised up or lowered them down, respectively. Consequently, the participants threw the ball further or closer to the goalkeeper’s body. They were misled about the target areas the goalkeeper could cover (see also Shim et al., 2014). Wood and Wilson (2010) showed that penalty takers kicked the ball more centralized when goalkeepers waved their arms. Presumably, the waving directed the penalty takers’ attention to the goalkeeper, distracting it away from the uncovered goal areas or the ball, which adversely affects the aiming of the kick (Bakker et al., 2006; Noël & van der Kamp, 2012). In addition, a few notational studies have suggested that goalkeepers’ deceptive actions impact penalty takers’ choice of which side of the goalkeeper to kick (Masters et al., 2007; Park et al., 2022). Masters et al. (2007) examined how the exact position of goalkeepers on the goal line affects the decision of penalty takers to kick to the left or right of the goalkeeper, using a convenience sample of 200 penalty kicks from the World Cup and European Championships. Penalty takers were (mis)led to aim at the goal side with more uncovered area when goalkeepers stood obviously off centre, but also when the goalkeeper stood marginally off centre. In addition, further laboratory studies showed that penalty takers were not consciously aware that their decisions to kick left or right are affected by the goalkeeper standing marginally off centre (i.e., disguise) (see also Noël et al., 2015).

Unlike the more conventional classification of deceptive actions into misleading, disguise and misdirection, which is based on how performers manipulate the information (Guldenpenning et al., 2017; Zheng et al., 2023), we divide the deceptive actions of goalkeepers functionally, that is, in terms of the false possibilities for action they invite: deceptive actions that invite penalty takers to kick to the goalkeeper’s preferred side (i.e., side deception) versus deceptive actions that invite penalty takers to aim for target areas that the goalkeeper can cover (i.e., target area deception).¹ The former indicates actions by which goalkeepers manipulate the information about the side to which penalty takers kick (e.g., standing clearly off-centre²), while the latter refers to actions that manipulate the information related to the area penalty takers’ aim for regardless of the side of the goal (e.g., waving the arms). Since the evidence for goalkeepers’ deceptive actions in competition is fragmented (Furley et al., 2017), it has yet to be determined what deceptive actions goalkeepers use in competition and whether they derive any advantage from producing them, as is suggested by laboratory studies.

To our knowledge, only one notational study explicitly addressed the question of whether goalkeepers take advantage of producing deceptive actions. Furley et al. (2017) focused exclusively on deliberate misdirection actions by the goalkeepers, which was referred to as distraction. They analysed 322 penalty kicks from World Cup and UEFA European Football Championships shootouts between 1984 and 2012. Two experienced coders independently rated whether goalkeepers attempted to “distract the penalty taker in any way” (Furley et al., 2017, p. 875), and also whether the penalty taker “predominantly gazes at the goalkeeper or away from the goalkeeper prior to and during the run-up” (Furley et al., 2017, p. 875). Furley et al. (2017) reported that goalkeepers intentionally distracted the penalty takers in 15% of penalty kicks. This resulted in higher no-goal rates (i.e., 27%) compared to kicks where goalkeepers did not try to distract the penalty taker (i.e., 17%), although that difference failed to reach significance (i.e., p = 0.056). Goalkeepers were also found to be more successful if the penalty taker predominantly gazed at the goalkeeper rather than away, but the authors did not report whether distraction was associated with increased gaze at the goalkeeper. Accordingly, the observations of Furley et al. (2017) suggest a substantial benefit for goalkeepers to distract the penalty taker, but a more comprehensive and precise assessment of the deceptive actions of goalkeepers is needed to be able to distinguish what deceptive actions (also including misleading and disguising actions) goalkeepers use in competition and to know their effectiveness.

Since goalkeepers are more successful when penalty takers predominantly gaze at the goalkeepers (Furley et al., 2017), it stands to reason to assume that the penalty kick strategy would moderate the effectiveness of goalkeeper deception. Two distinct penalty-taking strategies have been identified (W. Kuhn, 1988; Van der Kamp, 2006, 2011). With the keeper-independent strategy, penalty takers decide the side and target area to aim for before the run-up, regardless of the goalkeeper’s actions during the run-up. By contrast, in the keeper-dependent strategy penalty takers decide the side based on the goalkeeper’s actions during the run-up and to aim the ball at the side opposite of the goalkeeper’s dive. Penalty takers thus pay more attention to the goalkeeper when using the keeper-dependent than using the keeper-independent strategy (Noël et al., 2015). This potentially makes a goalkeeper’s deceptive action more effective when kickers employ the keeper-dependent strategy, perhaps especially when made in the run-up to the ball, because penalty takers are likely to monitor the goalkeeper’s actions during the run-up to use the latest available information before they execute the kick (Noël & van der Kamp, 2012). In addition, penalty takers that use the keeper-independent strategies scarcely focused their attention on goalkeepers during their run-up, while those who employed keeper-dependent strategies looked at goalkeepers both before and during the run-up (Noël & van der Kamp, 2012). It is therefore expected that deceptive actions during the run-up are less effective in penalties with keeper-independent strategies.

In sum, our objective was to examine what deceptive actions professional goalkeepers use in competition and how effective the different deceptive actions are. In addition, we assessed whether effectiveness was moderated by the penalty kick strategy and when the goalkeepers make the deceptive action. We hypothesized that no-goal rates would increase when goalkeepers use deceptive actions, and more so when penalty takers use a goalkeeper-dependent strategy than a goalkeeper-independent strategy. In addition, we expected that goalkeepers would dive to the correct side more often when using the side deception compared to the target area deception. Conversely, we expected a higher conversion rate (i.e., the rate of successful saves when goalkeepers dive to the correct side) in penalties with the target area deception than those with the side deception.³ Lastly, it was hypothesized that deception occurring predominantly before the start of the run-up would be beneficial for goalkeepers in both penalty kick strategies. However, deception actions produced predominantly during the run-up would only be effective in penalties with a keeper-dependent strategy.

Methods

Stimuli

The current dataset included all the in-game penalty kicks of the English Premier League and the German Bundesliga that were taken in the seasons 2016–2017 to 2019–2020. This sample size was calculated using G*Power 3.1 with an estimated medium effect size w of 0.3 and a significance level of 0.05 at the desired power of 0.8. The calculation resulted in a sample size of 88. Since goalkeepers used deception in at least 15% of the penalty kicks (Furley et al., 2017), the minimum required sample size was 586. A total of 714 penalty kick video clips were obtained from OPTA ⁴(https://optaplayerstats.statsperform.com). The videos had a frame rate of 30 Hz. For 62 videos, certain actions of the goalkeeper and/or penalty taker from the start of the run-up to the moment the ball crossed the goal line were not visible. These videos were excluded, resulting in a final video collection of 652 penalty kicks. All videos started at the moment the penalty was awarded by the referee and ended a few seconds after the ball crossed the goal line or was stopped by the goalkeeper.

Procedure and measures

As previous studies did not systematically describe all possible deceptive actions of goalkeepers, we first consulted a panel of experts, including three goalkeeper coaches and two sports scientists of the Dutch national soccer team. The information from this meeting combined with the information obtained from the scientific literature was used to construct an observation list of possible deceptive actions by the goalkeeper (see Table 1). The video clip of each penalty kick was first analysed at normal speed to identify the presence of a deceptive action, and if so, its type (i.e., actions that intended to influence penalty takers’ perception of possibilities for choosing side or target area),⁵ and the time of occurrence was noted (i.e., the first frame in which the goalkeeper started and ended the deceptive action and the onset of penalty taker’s run-up was obtained. The deceptive action was then classified as either predominantly before or during the run-up.). If none of the deceptive actions were observed, the penalty kick would be defined as non-deception. Subsequently, for all clips, penalty outcome (i.e., goal or no-goal), the side of the dive (i.e., the same or opposite side as the kick) and penalty kick strategy (i.e., keeper dependent or keeper independent) were derived from the video clip, with a frame-by-frame analysis using Kinovea software (version 0.9.1, Zheng et al., 2022). To identify the penalty kick strategy, we followed the three-predictor logistic regression model of Noël et al. (2015) (Equation 1). Specifically, the penalty taker’s attention to the goalkeeper (attention – no attention) and the fluency of the run-up (stagnant – fluent in terms of the pace of the run-up) were rated on an 11-point Likert scale and then converted to a score between 0 and 1. Kicking technique was categorized as instep, inside, or outside. The cut-off point of the model was 0.5, with a value of p (Equation 1) below 0.5 indicating a keeper-independent strategy, and a value of 0.5 and above implying a keeper-dependent strategy:

Table 1. Classification and observation list of deceptive actions.

Deception type	Indicator	Behavioral details
Deceptive actions intended to influence penalty takers’ perception of possibilities for choosing side	1. Making body feints during run-up penalty taker. 2. Clearly standing off centre 3. Pointing towards a specific corner 4. Standing completely still until the point of no return.	1.Angle of knee extension < 150 degree or angle of knee flexion < 100 degree Sánchez et al. (2005). 2. More than 10 cm off centre, measured with Kinovea before the run-up onset Masters et al. (2007). 3. One hand points towards one corner Weigelt et al. (2012). 4. Stands still until penalty taker initiates the kicking action (i.e., approx. 220 ms before foot-ball contact, Navarro et al., 2013). (Standing still all the time, no dive, is considered inaction and not deception.)
Deceptive actions intended to influence penalty takers’ perception of possibilities for aiming	1. Trying to make visual contact with the penalty taker (i.e., attract visual attention) verbally, non-verbally or physically 2. Dictating the pace of the penalty 3. Making themselves look smaller or larger	1. Move sideways back and forth on the goal line or dynamically waving the arms. Talk before/during run-up. Make eye contact or physical contact before the onset of the run-up. Walk backwards to the goal while maintaining eye contact. Shout when apparently losing contact. Continue talking before the run-up. Raise two arms and quickly put them down (waving). This may be repeated several times. Touch the crossbar or post of the goal 2. Postpone the moment to stand on the goal line. 3. Keep the arms up or down and maintain this posture for at least 2 seconds van der Kamp and Masters (2008).

$\mathit{p}=\frac{1}{1+{\mathit{e}}^{-\mathit{x}}}$

where,

$\mathit{x}=-11.67+13.15\times \left(\mathit{Attention}\right)+20.98\times \left(\mathit{Runup}\right)-1.13\times \left(\mathit{Kickingtechnique}\right)$

Inter-rater reliability (Cohen’s Kappa) of two independent observers (BvdZ and RZ) based on 10% of all the penalty kicks was larger than 0.82 for every dependent variable except for the behaviour details of the deceptive action (i.e., 0.74). Disagreements were resolved by discussion.

Data analyses

First, chi-squared tests were used to examine the distribution of penalty kick outcome and dive side as a function of the presence of deceptive actions. Next, additional chi-squared tests compared the effectiveness of the two types of deception (i.e., side deception versus target area deception). Third, two separate 3-way log-linear analyses were conducted to investigate whether the distribution of penalty kick outcome (goal, no-goal) and dive side (same side, opposite side) was influenced by deception (deception, no deception) and penalty kick strategy (keeper-dependent, keeper-independent). Furthermore, another two separate 3-way log-linear analyses were conducted to investigate whether the distribution of penalty kick outcome (goal, no-goal) and dive side (same side, opposite side) across two deception types (side deception, target area deception) was moderated by penalty kick strategy (keeper-dependent, keeper-independent). Lastly, for the penalties with deception, two separate 3-way log-linear analyses were used to examine the distribution of penalty kick outcome (goal, no-goal) and dive side (same side, opposite side) across penalty kick strategy (keeper-dependent, keeper-independent) and deception timing (before run-up, during run-up).

Results

Descriptive analysis

Of the 652 penalty kicks, 77.6% were scored, 16.7% were saved, and 5.7% were missed. Goalkeepers used deception in half (i.e., 50%) of the penalties. Of these deceptive actions, 34% were side deceptions, and 66% were target area deceptions. Finally, penalty takers adopted the keeper-dependent and keeper-independent strategies in about one-quarter (24.7%) and three-quarters (75.3%) of penalty kicks, respectively.

Effectiveness of deception

As shown in Table 2, there is a significant association between goalkeeper deception and penalty kick outcome, χ² (1) = 7.94, p = .005 and Cramer’s V = 0.11 and between goalkeeper deception and dive side, χ² (1) = 9.11, p = .003 and Cramer’s V = 0.12. No-goal rates were higher and goalkeepers dived to the same side more often when goalkeepers produced deceptive actions than when they did not. In addition, one proportion tests on dive side indicated that goalkeepers dived to the same side as the ball at chance level, z = 1.2, p = 0.23, when using deception, but below chance level, z = 5.095, p < 0.001, when they did not produce deceptive actions.

Table 2. Absolute frequency (and percentages) of penalty kick outcome and dive side for no deception and deception penalties.

	No deception	Deception
Penalty kick outcome
Goal	268 (82.2%)	238 (73%)
No-goal	58 (17.8%)	88 (27%)
Dive side
Same side	117 (35.9%)	157 (47.5%)
Opposite side	209 (64.1%)	179 (52.5%)

Next, we compared the effectiveness between side deception and target area deception. As shown in Table 3, the distributions of penalty kick outcome, dive side and conversion rate were comparable between the two types of deception, all ps >0.603.

Table 3. Absolute frequency (and percentages) of penalty kick outcome and dive side for side deception and target area deception.

	Side deception	Target area deception
Penalty kick outcome
Goal	81 (73%)	157 (73%)
No-goal	30 (27%)	58 (27%)
Dive side
Same side	55 (49.5%)	100 (46.5%)
Opposite side	56 (50.5%)	115 (53.5%)
Conversion rate
Converted	30 (54.5%)	58 (58%)
Not converted	25 (45.5%)	42 (42%)

Association between deception and penalty kick strategy

First, two separate 3-way log-linear analyses were conducted to investigate whether the distribution of penalty kick outcome (goal, no-goal) and dive side (same side, opposite side) across deception (deception, no deception) was moderated by penalty kick strategy (keeper-dependent, keeper-independent). The analysis for penalty kick outcome produced a final model that retained two 2-way interaction terms, including deception by penalty kick outcome, χ² (1) = 7.704, p = .006 and deception by penalty kick strategy χ² (1) = 6.682, p = .01 (Table 4). The likelihood ratio of this model was χ² (2) = 0, p = 1. The deception by penalty kick outcome interaction indicated that the no-goal rate was 1.71 times higher when goalkeepers used deception compared to when they did not. The deception by penalty kick strategy interaction suggested a positive association between penalty kick strategy and goalkeepers’ deception. Penalty takers used the keeper-dependent strategy 1.62 times more often when goalkeepers used deception than when goalkeepers did not.

Table 4. Absolute frequency (percentages) of penalty kick outcome and dive side as a function of penalty kick strategy and deception.

	No deception		Deception
	Keeper dependent	Keeper independent	Keeper dependent	Keeper independent
Penalty kick outcome
Goal	53 (80.3%)	215 (82.7%)	69 (72.6%)	169 (73.2%)
No goal	13 (19.7%)	45 (17.3%)	26 (27.4%)	62 (26.8%)
Dive Side
Same side	19 (28.8%)	98 (37.7%)	44 (46.3%)	111 (48.1%)
Opposite side	47 (71.2%)	162 (62.3%)	51 (53.7%)	120 (51.9%)

A similar three-way log-linear analysis for dive side also revealed significant interactions of deception by strategy effect, χ² (1) = 7.569, p = 0.006 and deception by dive side, χ² (1) = 9.735, p = 0.002. The likelihood ratio of this model was χ² (2) = 0, p = 1. As shown in Table 4, there were positive associations between deception and dive side and between deception and penalty kick strategy. The deception by dive side interaction indicated that goalkeepers dived to the same side 3.57 times more frequently when deception was used than when they did not use deception. The deception by penalty kick strategy interaction suggested that penalty takers used the keeper-dependent strategy 1.62 times more often when goalkeepers used deception than when they did not.

Next, we further examined the effect of different types of deception and penalty kick strategy on penalty outcome and dive side. The three-way log-linear analysis for penalty kick outcome produced a final model without any interactions. The likelihood ratio of this model was χ² (4) = 0, p = 1. The three-way interaction was not significant, χ² (1) = 0.315, p = .58. Further, none of the two-way interactions were significant, all ps >0.61. Similarly, the 3-way log-analysis for dive side produced a final model without any interactions. The likelihood ratio of this model was χ² (4) = 0, p = 1. The three-way interaction, χ² (1) = 0.035, p = .85, and two-way interactions, all ps >0.78, were not significant. These results indicated that the interaction between deception type and penalty kick strategy did not influence the penalty kick outcome or dive side (Table 5).

Table 5. Absolute frequency (and percentages) of penalty kick outcome and dive side as a function of penalty kick strategy and deception condition.

	Side deception		Target area deception
	Keeper- dependent	Keeper- independent	Keeper- dependent	Keeper- independent
Penalty kick outcome
Goal	26 (70.3%)	62 (74.7%)	43 (74.1%)	107 (72.3%)
No goal	11 (29.7%)	21 (25.3%)	15 (25.9%)	41 (27.7%)
Dive side
Same side	16 (44.4%)	40 (47.6%)	28 (47.5%)	71 (48.3%)
Opposite side	20 (55.6%)	44 (52.4%)	31 (52.5%)	76 (51.7%)

Associations between timing of deception and penalty kick strategy

Finally, we examined, for the penalty kicks in which goalkeepers produced a deceptive action, whether penalty kick outcome was related to (the combination of) penalty kick strategy and the timing of deception. The three-way log-linear analysis produced a final model that retained all effects. The likelihood ratio of this model was χ² (0) = 0, p = 1. This indicated that the removal of highest-order (three-way) interaction was significant, χ² (1) = 4.086, p = .04. Next, separate chi-squared tests on deception timing and penalty kick outcome were performed for the keeper-dependent and keeper-independent strategies, respectively. This showed that only penalties with a keeper-independent strategy were affected by deception timing, χ² (1) = 5.237, p = .02. Goalkeepers saved penalties 2.21 times more often when they used deception predominantly before the start of the run-up compared to during the run-up (Table 6). By contrast, in penalties with keeper-dependent strategies, the timing of deception did not affect penalty outcome, χ²(1) = 0.705, p = .40 (Table 6). Finally, a similar three-way log-linear analysis for dive side with penalty kick strategy and the timing of deception did not reveal any significant interaction, all ps >0.15.

Table 6. Absolute (and percentages) frequency of penalty kick outcome and dive side as a function of penalty kick strategy and deception timing.

	Keeper-dependent		Keeper-independent
	During	Before	During	Before
Penalty kick outcome
Goal	34 (70.8%)	33 (78.5%)	63 (82.8%)	103 (68.7%)
No goal	14 (29.2%)	9 (21.4%)	13 (17.2%)	47 (31.3%)
Dive side
Same side	23 (47.9%)	20 (47.6%)	37 (48.7%)	72 (48%)
Opposite side	25 (52.1%)	22 (52.4%)	39 (51.3%)	78 (52%)

Discussion

The present study investigated the effectiveness of goalkeeper deceptive actions for in-game penalty kicks in competitive professional soccer matches. The deceptive actions of a goalkeeper can hinder the penalty taker’s perception of the true possibilities for action or affordances, especially regarding the choice of side and aiming at a target area. In doing so, we also examined how effectiveness is mediated by the penalty kick strategy and the timing of deception. To this end, we analysed video footage of in-game penalty kicks taken during the 2016–2017 to 2019–2020 seasons from the English Premier League and the German Bundesliga. The results revealed that goalkeepers used deception in half of the penalties and significantly benefited from this (i.e., there were more dives to the same side as the kick and, critically, fewer goals ensued). This benefit was regardless of the type of deception. Notably, goalkeepers’ deceptive actions were most effective when they are employed simultaneously with the penalty taker who is closely monitoring the goalkeeper, that is, during and before the run-up for the keeper-dependent strategy and only before the run-up in the case of a keeper-independent strategy.

Confirming that deceptive actions are effective is not trivial. For penalty takers, for instance, a recent review raised reasonable doubts about the effectiveness of deceptive actions produced during the run-up, at least within the context of experimental studies (Zheng et al., 2023). By contrast, the current study unequivocally indicated that goalkeepers’ deceptive actions during the penalty kick can be successful. Dives to the same side the ball is kicked and the no-goals increased by 12% and 9%, respectively, in penalties with deception compared to those without. This is of similar magnitude as previously reported by Furley et al. (2017).

Based on ecological psychology, we classified the deceptive actions in terms of the false opportunities for action they present, either inviting penalty takers to kick on the goalkeeper’s chosen dive side (i.e., side deception) or to aim for target areas that the goalkeeper is more likely to cover, regardless of goal side (i.e., target area deception). Based on logic and some laboratory findings (e.g., Masters et al., 2007; van der Kamp & Masters, 2008; Wood & Wilson, 2010), we expected that goalkeepers dived to the correct side more frequently when using side deception and converted or saved more kicks to the correct dive side when using target area deception. However, the findings did not support the hypotheses related to the type of deception. Penalty kick outcome and dive side were similar for the two deception types. This indicates a non-specific effect of deception, irrespective of the type of deception.⁶ More importantly, penalty takers reduced kicks to the empty side from approximately 65% in penalties without goalkeeper deception to 50% (i.e., chance level) in penalties with deception. This indicates that with goalkeeper deception penalty takers may guess the side to which the goalkeeper dives, while without deception, they are enabled to more accurately perceive the side to which the goalkeeper is going to dive and thus manage to kick to the opposite side more often.

Accordingly, we propose an alternative hypothesis to account for the effect of goalkeepers’ deception in the penalty kick. Goalkeepers are considered as the biggest source of stress for penalty takers (Jordet & Elferink-Gemser, 2012; Wood et al., 2015). Possibly, when confronted with a goalkeeper producing a deceptive action, penalty takers tend to lose focused attention. Perhaps, this loss is induced by the mere movement of goalkeepers and irrespective of the type of affordance hindered or (mis)information generated in the deception. As a result, and especially in high-pressure situations, which are known to reduce attentional resources, problem-focused coping may suffer (Lazarus & Folkman, 1984; Nicholls & Polman, 2007), resulting in increased guessing and/or poor kick execution (Nieuwenhuys & Oudejans, 2012). This hypothesis implies that goalkeeper deception disrupted the accuracy of affordance perception in a more generic sense rather than inviting the penalty takers to perceive specific false affordances (cf. Ramsey et al., 2022). To further test this alternative hypothesis, a qualitative approach could be adopted by interviewing goalkeepers supplemented by a video-recall method (e.g., Jordet & Elferink-Gemser, 2012).

The current study also demonstrated that the effectiveness of goalkeeper deception was moderated by the interplay of the timing of the deceptive action and the penalty taker’s kicking strategy. For the keeper-dependent strategy, the timing of the deceptive action, whether before or during the run-up, did not affect its effectiveness. Conversely, in penalties with a keeper-independent strategy, a goalkeeper’s deception was less effective when it was predominantly used during the run-up, as opposed to before it. This finding is in line with the results from a study by Noël and van der Kamp (2012) that investigated differences of gaze behaviours of penalty takers in relation to keeper-dependent and keeper-independent strategies. This study found that penalty takers who used a keeper-independent strategy spent 40% of the time fixating the goalkeeper before the run-up but barely looked at the goalkeeper during the early run-up and almost never during the late run-up. By contrast, penalty takers employing a keeper-dependent approach fixated the goalkeeper both before and during the run-up, with 50% of the viewing time during the early run-up and 25% of the viewing time during the late run-up. As a result, it appears that the effectiveness of goalkeeper deception hinges on whether the penalty takers direct their attention towards the goalkeeper at the moment the deceptive action is made (see also Furley et al., 2017).

There are several limitations of the current study. First, the deceptive actions that we included in our analysis were identified as such by a panel of experts. However, identifying that goalkeepers produce these actions, as we did in the current study, does not prove that the goalkeepers were actively hindering the opponent’s perception of affordance. In this respect, we are biased to overestimate the frequency that goalkeepers deliberately use deceptive actions. In addition, as mentioned in footnote 3, we were unable to accurately determine the exact location where the ball crossed the goal line due to the quality of available videos (i.e., viewing angle). Future studies are encouraged to examine whether this location is indeed influenced by target area deception in penalty kicks during competition, as suggested by lab-based studies. Moreover, for each penalty kick, the type of deception was classified based on the most salient deceptive action that had the longest duration. Yet, we recognize that goalkeepers sometimes used more than one deceptive action during a penalty kick. For example, goalkeepers first tried to distract penalty takers by moving arms vigorously for a long period before the start of the run-up (i.e., target area deception) and then performed body feint (i.e., side deception) around the moment the penalty taker initiated the kicking action. Therefore, it is pertinent to further assess whether the efficacy of deception is heightened when multiple deceptive actions are produced or when the various types of deceptive actions are strategically combined. If indeed deception primarily operates by perturbing the focus of attention, then we expect that the number of deceptive actions is a stronger predictor for the efficacy of deception than other more intricate factors related to the type of the deceptive action. Finally, studies have shown that defenders use situational information (e.g., information of the opponent’s action preferences) to anticipation the opponents’ actions decision-making (Abernethy et al., 2001), and at least one experimental study has reported that soccer goalkeepers are able to use situational information when facing penalty kicks (Navia et al., 2013). Consequently, goalkeepers may exploit situational information (e.g., information about the penalty taker’s preferred side to kick) or knowledge about the penalty taker using situational information (e.g., knowing that the penalty taker may use information about the goalkeeper’s dive preferred side to dive) to deceive, as is suggested for the interactions of the bowler and batter in cricket (Marshall et al., 2024). Qualitative studies may shed light on the use of such deceptive strategies, but for now and based on the anecdotical evidence available, we doubt whether elite goalkeepers are aware of these strategies, and if so, if they are able to employ them systematically (Lehmann & Siemes, 2017; van der Kamp & Savelsbergh, 2014).

In conclusion, this study delved into the effectiveness of goalkeepers’ deceptive actions during in-game penalty kicks in competitive soccer matches. Although goalkeepers might be in a disadvantaged position during penalty kick situations, they are not without opportunities to save the shot. Results revealed that goalkeepers significantly benefited from producing deceptive actions, leading to fewer scored penalties. In this respect, the findings also emphasized the importance of the timing of the deception. However, the underlying process for the effect of deception may have been different from that reported in laboratory studies. Inviting false perception of possibilities for side and target area cannot account for the general or non-specific effects we observed. Alternatively, we hypothesize that penalty takers may lose control over attentional focus when confronted with goalkeepers making deceptive actions.

Goalkeepers are advised to use deception before the run-up when penalty takers looked at them. However, the International Football Association Board (IFAB, 2023) has started to limit what they consider unfair use of goalkeepers’ deception, stipulating that “the goalkeeper must not behave in a way that unfairly distracts the kicker, e.g., delay the taking of the kick or touch the goalposts, crossbar or goal net” (IFAB 23/24. Pg. 119). Goalkeepers thus should be aware of the boundary between proper and unfair distraction, avoiding intimidating behaviours when employing deceptive actions. This boundary is not precise and likely varies between referees and competitions. Goalkeepers are therefore advised to cautiously explore a referee’s tolerance in each match. Penalty takers are suggested to practice in a representative environment and confront various types of deceptive actions employed by goalkeepers, thereby bolstering problem-focused coping strategies. In addition, rather than only practicing with goalkeepers, penalty takers can compete with fellow players or even with audiences to disrupt their attentional focus, or increase the level of anxiety (Wood et al., 2015). Even low levels of anxiety during practice can be effective to improve the performance in competition (Oudejans & Pijpers, 2009).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the China Scholarship Council.

Notes

¹ If we follow the conventional classification of deceptive actions, it is noteworthy that identical categories of deceptive actions can exert different effects on opponents. For example, both making body feints and making goalkeepers look smaller are misleading actions. Yet, they affect the possibilities for action in a different way. They aim to invite penalty takers to choose the same side as the goalkeeper intends to dive and to aim for target areas that are within reach of the goalkeeper, respectively.

² Masters et al. (2007) argued that goalkeepers did not stand off centre intentionally since they did not dive to the side with greater area more often than the side with smaller area. However, goalkeepers likely do not simply dive to the side with greater area, but will also take the kinematic information from the kickers into account. Only using the goalkeepers’ dive direction may not be sufficient to speculate about the goalkeepers’ intention.

³ We aimed to calculate the exact location where the ball crossed goal line for each penalty kick. However, it appeared that unlike in lab-based study where the penalty kick can be filmed from a frontal perspective (Wood & Wilson, 2010, Navarro et al., 2013), the video footage from competition typically is not frontal, with different degrees of deviation. In the current study, therefore, we were unable to accurately measure the exact location where the ball crossed the goal line, and consequently, did not directly assess the, possibly subtle, differences in aiming.

⁴ OPTA has been proven to be a reliable professional source for analysing soccer penalty kicks (e.g., Jamil et al., 2021; Liu et al., 2013).

⁵ If more than one deceptive action was identified during a kick, then only the most salient and prolonged action was considered.

⁶ We also compared the effectiveness of different deception types following the conventional classification: misleading, disguise, and misdirection. Also, no differences were found in effectiveness between these three deception types.