Abstract
Why some elite athletes maintain or enhance performance under acute competitive pressure whereas others underperform remains a central problem in sport psychology. Mental toughness, clutch performance, challenge and threat appraisal, attentional control, and choking have each been proposed as explanatory constructs, yet their conceptual boundaries and behavioral manifestations remain contested. Professional tennis provides an unusually informative natural setting because its scoring architecture repeatedly generates discrete, consequential pressure episodes while preserving objective performance records. This comparative behavioral case study examines pressure-relevant performance in Novak Djokovic and Roger Federer. Rather than treating archival match outcomes as direct measures of latent psychological traits, break points, tie-breaks, deciding sets, match points, and major-championship encounters are conceptualized as behavioral indicators of performance under varying forms of competitive pressure. These indicators are interpreted within contemporary research on mental toughness, clutch performance, challenge-threat appraisal, and choking. The available evidence indicates meaningful differences in the distribution of performance across high-pressure contexts, with several indicators favoring Djokovic, particularly in prolonged deciding-set contests and exceptionally consequential match situations. However, the pattern is neither uniform nor sufficient to establish that one athlete possesses a globally superior latent psychological trait. Federer also demonstrates substantial evidence of pressure resilience, and observed differences may reflect technical, tactical, physiological, opponent, surface, age, and selection effects. The findings support a constrained conclusion: Djokovic's competitive record contains unusually strong behavioral evidence of successful performance in several high-pressure contexts, but retrospective performance statistics alone cannot identify the psychological mechanisms responsible. The study illustrates both the potential and limits of archival behavioral data for investigating performance under pressure and proposes a multilevel framework integrating point importance, baseline ability, opponent strength, match context, psychophysiology, and athlete-reported states.
Keywords: performance under pressure, mental toughness, clutch performance, choking, tennis, challenge and threat, Novak Djokovic, Roger Federer
1. Introduction
Elite sport is partly a problem of execution under consequence. At the highest levels, technical differences between opponents can be small, while the importance of particular moments can vary dramatically. A routine point early in a match and a championship point may require the same nominal motor skills but differ profoundly in perceived stakes, evaluative threat, attentional demands, and consequences. Understanding why performance is maintained, enhanced, or degraded as stakes rise is therefore a central problem in sport psychology and performance science (Hill et al., 2010; Schweickle et al., 2021, 2026).
The contemporary literature rejects a simple equation in which pressure necessarily impairs performance. Pressure can be debilitative, neutral, or facilitative, and athletes differ in both their appraisal of pressure and their behavioral response to it (Schweickle et al., 2026). The systematic review of clutch performance by Schweickle et al. (2021) found substantial definitional, conceptual, and measurement heterogeneity, including disagreement over whether clutch performance should be treated as an enduring ability or an isolated episode. Subsequent qualitative work suggests that clutch performance is situational and context dependent (Schweickle et al., 2022). A recent meta-study identified self-efficacy, demand appraisal, mental toughness, and task-focused resource allocation among recurring facilitators of successful performance under pressure (Hufton et al., 2024).
Mental toughness is frequently invoked to explain such differences, but its status as a construct remains debated. The influential 4C model describes control, commitment, challenge, and confidence as constituent dimensions (Clough et al., 2002). However, a five-study construct-validation program by Gucciardi et al. (2015) concluded that mental toughness may be better represented as a unidimensional construct and showed that it can display both enduring and situation-varying properties. A recent systematic review and meta-analysis nevertheless indicates a meaningful positive association between mental toughness and athletic performance, while also demonstrating heterogeneity across measures and sporting contexts. Thus, mental toughness is relevant to pressure performance but should not be inferred directly from a small number of dramatic competitive outcomes.
Professional tennis is particularly valuable for studying performance under pressure. Its nested scoring system creates repeated changes in point importance, from ordinary points to break points, set points, tie-breaks, match points, and deciding sets. Point-level research using twelve years of US Open data found persistent heterogeneity in players' "critical ability," defined as adjustment of performance to the importance of the situation, and showed that this ability was associated with career success (González-Díaz et al., 2012). Other tennis studies illustrate why simplistic assumptions are hazardous. Cohen-Zada et al. (2017), analyzing Grand Slam games, reported performance deterioration under competitive pressure among men, whereas Goff et al. (2025), using more than 136,000 professional tennis game instances, found that both male and female players improved service-hold probabilities under heightened pressure. These findings use different operationalizations, samples, time periods, and units of analysis; together, they imply that pressure response is conditional rather than universal.
The rivalry between Novak Djokovic and Roger Federer offers an unusually informative comparative case. Their careers overlapped extensively; both accumulated exceptional records across surfaces and tournament levels; and they repeatedly competed against one another in major events. Public discussion often describes Djokovic as the psychologically tougher competitor, while Federer is sometimes portrayed as comparatively vulnerable in a small number of highly salient closing situations. Such narratives are intuitively compelling but scientifically hazardous because they can transform memorable outcomes into post hoc personality diagnoses.
The purpose of the present study is therefore not to determine which player "possesses" more mental toughness as though a latent psychological trait could be read directly from match results. Instead, the study asks three narrower questions: (1) Do Djokovic and Federer show different descriptive patterns across multiple pressure-relevant performance contexts? (2) Are those patterns consistent across indicators, or do they depend on the type of pressure being examined? (3) What psychological interpretations are compatible with the observed record, and which interpretations exceed what archival behavioral evidence can establish? By answering these questions, the study seeks to contribute both substantive insight into elite tennis and a methodological framework for studying pressure performance without conflating outcome statistics with psychological measurement.
2. Theoretical Framework
2.1 Pressure, performance, and the problem of inference
Pressure is best understood as a property of the performer-situation relationship rather than a fixed property of an event. Competitive context, reward and loss contingencies, ego relevance, audience evaluation, uncertainty, and perceived consequences can increase the importance assigned to successful performance (Hill et al., 2010; Schweickle et al., 2026). Two athletes facing the same score can therefore experience different subjective pressure, and the same athlete may appraise nominally similar situations differently across time.
This distinction creates an inferential problem for archival research. Score states can identify situations that are objectively more consequential, but they cannot reveal subjective appraisal, anxiety, attentional focus, confidence, or perceived control. Accordingly, the present study uses the term pressure-relevant context rather than assuming that every break point or match point generated an identical psychological state. Behavioral records can establish what happened under objectively consequential conditions; they cannot, by themselves, establish why it happened.
2.2 Mental toughness: useful construct, imperfect proxy
Mental toughness has been conceptualized as the capacity to maintain high levels of functioning despite stress, adversity, or pressure. The 4C framework emphasizes control, commitment, challenge, and confidence (Clough et al., 2002), and sport-specific measures such as the Sports Mental Toughness Questionnaire have attempted to operationalize related attributes (Sheard et al., 2009). Yet conceptual debate remains substantial. Gucciardi et al. (2015) found stronger support for a unidimensional representation and demonstrated both trait-like and state-like properties. The implication is important for archival case analysis: even if one athlete repeatedly performs better in critical moments, those outcomes do not uniquely identify a particular mental-toughness dimension.
Meta-analytic evidence supports a positive association between mental toughness and athletic performance but does not license reverse inference from performance to trait level. Performance is multiply determined. Technical skill, tactical matchups, conditioning, injury status, age, fatigue, surface, opponent behavior, random variation, and scoring structure all contribute. Mental toughness is therefore treated here as one plausible explanatory resource within a broader system rather than as the dependent variable itself.
2.3 Clutch performance and choking
Clutch performance refers broadly to successful or improved performance under pressure, whereas choking denotes suboptimal performance under stressful conditions (Hill et al., 2010; Schweickle et al., 2021). Importantly, the clutch literature does not support a single uncontested definition. Schweickle et al. (2021) identified two major traditions: clutch as an ability and clutch as an episode. The episodic approach is particularly relevant here because a player can display clutch performance in one situation without being globally "clutch," and an athlete can experience both clutch and choking episodes across a long career.
Choking likewise should not be diagnosed from an unfavorable outcome alone. Losing a match after holding match point may involve anxiety-driven attentional disruption, but it may also reflect an opponent's exceptional return, an unreturnable shot, fatigue, tactical error, or ordinary probabilistic variance. Reviews of choking emphasize both distraction and self-focus mechanisms, while intervention research shows that pre-performance routines, quiet-eye training, acclimatization, and related strategies can protect execution under pressure (Gröpel & Mesagno, 2019; Hill et al., 2010). A 2026 meta-analysis further reported beneficial effects of pressure-counteracting interventions across motor-accuracy and execution outcomes, while noting variation between closed- and open-skill contexts (Zeng et al., 2026).
2.4 Challenge-threat appraisal
The biopsychosocial approach to challenge and threat offers a complementary account. Challenge states arise when perceived coping resources are sufficient relative to demands, whereas threat states occur when demands are perceived to exceed resources. An updated meta-analysis of 62 studies involving 7,418 participants found that challenge states were associated with better performance than threat states across domains, although effects were small (Hase et al., 2025). Sport-specific theory also cautions against treating challenge and threat as simple opposites; both may coexist, and appraisal can change rapidly during competition (Meijen et al., 2020).
For the present comparison, challenge-threat theory provides plausible mechanisms but not directly observed variables. A high five-set win rate, for example, is compatible with effective challenge appraisal, but it does not demonstrate that the athlete subjectively appraised the match as a challenge. The distinction between compatibility and demonstration is maintained throughout the analysis.
2.5 A multilevel model of pressure performance
Integrating these literatures suggests a multilevel sequence: objective match context influences perceived demands and consequences; athletes appraise those demands relative to resources; appraisal interacts with confidence, attentional control, routines, experience, and mental-toughness-related resources; these processes affect execution; and execution interacts with opponent behavior and tennis scoring to produce outcomes. This sequence prevents two common errors: reducing pressure performance to personality, and reducing psychology to final scores.
3. Method
3.1 Design
The study uses a comparative archival case-study design. Djokovic and Federer are treated as information-rich cases rather than as a random sample from a population. The purpose is analytic comparison and theory development, not population-level estimation. This design is appropriate for historically exceptional performers for whom experimental manipulation and retrospective psychometric assessment are impossible.
3.2 Evidence hierarchy and source selection
Evidence was organized hierarchically. Peer-reviewed research was used to establish theory and empirical findings about pressure performance. Official tour and tournament records were preferred for match results and event-specific statistics. Reputable tennis statistical compilations and journalism were used only for descriptive records not maintained as official ATP summary variables, and such evidence is explicitly identified as secondary. Autobiographical or media statements were not treated as psychometric evidence.
Several statistics contained in the earlier version of this manuscript were excluded because their derivation could not be independently reconstructed from the stated methods. In particular, pressure-versus-baseline break-point deltas were not retained. This conservative decision reduces apparent evidential breadth but increases reproducibility.
3.3 Pressure-relevant indicators
Five classes of indicators were considered: (a) break-point performance, because break points alter the probability of winning a game and potentially a set; (b) tie-break performance, because each point has unusually high leverage within a compressed scoring sequence; (c) deciding-set and five-set performance, because prolonged matches combine consequence, fatigue, and repeated recovery from setbacks; (d) match-point episodes, because the next point can terminate the contest; and (e) direct major-championship encounters, which combine high objective stakes with matched opponents. These indicators should not be assumed to measure the same psychological process. Their convergence is more informative than any single statistic.
3.4 Analytical logic
The analysis is primarily descriptive and theory-guided. No inferential test is used to claim a stable latent trait difference because the episodes are non-random, nested within players and opponents, and confounded by era, age, surface, tournament stage, and playing style. The emphasis is therefore on pattern consistency, counterevidence, and alternative explanations. A result is described as behaviorally consistent with successful pressure performance when execution or match outcome remains favorable in a context of elevated objective consequence; psychological mechanisms are discussed as hypotheses rather than established causes.
4. Results
4.1 Pressure performance is not uniformly distributed
The archival record does not support a simple contrast in which Djokovic always excels under pressure and Federer does not. Federer's career contains abundant evidence of pressure success, including an historically strong aggregate tie-break record and numerous victories after saving match points. The relevant question is therefore comparative and conditional: whether Djokovic's outcomes are more concentrated in particular high-consequence contexts, not whether Federer lacked pressure resilience.
4.2 Tie-breaks: aggregate excellence versus situational concentration
Aggregate tie-break performance complicates simplistic narratives. Federer finished his career with an exceptional overall tie-break record, evidence that he was highly effective in compressed, high-leverage scoring sequences. This is important counterevidence against any general characterization of Federer as a poor pressure performer. Djokovic likewise has an elite tie-break record. However, career tie-break percentage is not a pure pressure measure: stronger players are generally favored in tie-breaks, serving quality matters, opponent composition differs, and tie-breaks vary substantially in consequence.
The theoretically interesting issue is not simply who won a larger percentage of all tie-breaks, but whether performance changes as the consequence of the tie-break increases. The original manuscript's claim of a perfect 7-0 record for Djokovic versus 0-3 for Federer in a narrowly defined class of deciding tie-breaks was not retained as a primary quantitative result because the sampling rule and complete denominator were insufficiently documented. Selective high-salience subsets can produce compelling narratives while concealing the larger opportunity structure. A confirmatory analysis should predefine the event class and reconstruct every eligible episode.
4.3 Five-set performance and prolonged adversity
The original dataset reported Grand Slam five-set records of 38-11 for Djokovic and 33-24 for Federer, corresponding to win rates of 77.6% and 58.2%, respectively. Treated descriptively, this is one of the clearest differences in the manuscript. Five-set contests require repeated recovery from setbacks and sustained technical and physical execution over long durations, and they frequently contain multiple shifts in momentum and score pressure.
Yet the psychological meaning of this difference is underdetermined. Five-set success depends heavily on conditioning, age, surface, opponent quality, tactical adaptability, injury, and the circumstances that caused a match to reach a fifth set. Selection effects are particularly important: a dominant player may avoid fifth sets against weaker opponents, leaving a disproportionately difficult set of five-set cases. The observed difference is therefore behaviorally consistent with exceptional resilience under prolonged competitive adversity, but it cannot be attributed uniquely to emotional control, commitment, or challenge appraisal.
4.4 Match points: extreme consequence, extreme inferential risk
Match points are intuitively attractive pressure indicators because a single point can terminate the contest. They are also methodologically treacherous. The original manuscript reported that Federer lost 24 matches after holding match point and Djokovic only four. A journalistic reconstruction after the 2019 Wimbledon final independently reported 24 such Federer losses and emphasized that the ATP did not maintain this as a standard official summary statistic. Because the underlying denominator is not an official standardized variable, these figures should be treated as descriptive archival counts rather than validated psychological metrics.
The same caution applies to comeback wins after saving match points. Federer himself accumulated an unusually large number of such victories. This symmetrical evidence matters. If losing after holding match point were treated as direct evidence of weak mental toughness, then winning repeatedly from match point down would imply the opposite conclusion about the same athlete. The coexistence of both patterns demonstrates why match-point outcomes are better conceptualized as rare high-consequence episodes than as direct measures of a stable trait.
4.5 The 2019 Wimbledon final as a case of performance concentration
The 2019 Wimbledon final provides an especially informative illustration of why aggregate match performance and performance at outcome-critical moments should be distinguished. Federer won 218 total points to Djokovic's 204 and produced 94 winners to Djokovic's 54, yet Djokovic won all three tie-breaks and the championship. Federer served for the title at 8-7 in the fifth set and held two championship points at 40-15; Djokovic saved both, broke back, and eventually won the deciding tie-break.
It would be inappropriate to infer from this match that Federer "choked," that Djokovic possessed greater mental toughness as a latent trait, or that the result was caused primarily by psychological differences. Those interpretations cannot be established from retrospective match statistics. The match nevertheless illustrates a theoretically important phenomenon: performance advantages distributed across an entire contest need not translate into victory when a scoring system assigns disproportionate outcome significance to a relatively small number of points.
This pattern can be described as performance concentration under pressure: the extent to which successful execution is preserved or enhanced as the conditional importance of individual points increases. The construct is related to González-Díaz et al.'s (2012) critical ability but emphasizes the distribution of successful execution across a pressure gradient within a nonlinear scoring system. The 2019 final does not validate this construct by itself; rather, it demonstrates why point-level models are preferable to retrospective labels such as "mentally tougher."
4.6 Direct major-championship encounters
Djokovic and Federer played repeatedly at Grand Slam tournaments, including four finals, three of which Djokovic won. Direct encounters offer a partial control for opponent quality because the same two players face one another, but they remain confounded by age, surface, tournament, and career phase. Federer was nearly six years older than Djokovic, making late-career meetings especially difficult to interpret as trait comparisons. Nevertheless, Djokovic's success in several close major matches contributes to the descriptive pattern that fuels his reputation for high-consequence resilience.
The correct inference is narrower than the popular narrative. These matches show that Djokovic repeatedly produced sufficient performance to survive imminent defeat against Federer in major championships. They do not reveal whether the causal mechanism was lower anxiety, better attentional control, superior return skill, greater physical reserve, more favorable tactical choices, chance, or some combination of these factors.
5. Discussion
5.1 What the evidence supports
Across the pressure-relevant domains considered here, Djokovic's record contains a notable concentration of successful outcomes in prolonged and extremely consequential contexts. The five-set record reported in the archival dataset, repeated survival of match points in major matches, and direct victories over Federer in tightly contested Grand Slam encounters are jointly consistent with unusually effective performance under pressure. The convergence across qualitatively different contexts is more informative than any single famous match.
However, the evidence does not support the stronger claim that Djokovic has been psychometrically demonstrated to possess more mental toughness than Federer. Neither athlete completed a validated mental-toughness instrument for this study; neither athlete's moment-to-moment appraisal, anxiety, attentional focus, cardiovascular response, or perceived coping resources was observed. The data are behavioral outcomes, not measurements of latent psychological states.
This distinction improves the scientific contribution. Rather than asking an unanswerable retrospective personality question, the analysis identifies a behavioral pattern and specifies the mechanisms that future research would need to measure to explain it. Djokovic's record can legitimately be described as containing stronger evidence of successful execution in several high-pressure contexts. Whether that pattern arises from mental toughness, challenge appraisal, confidence, attentional control, superior tennis skills, physical durability, or interactions among them remains an empirical question.
5.2 Why the 4C model should not be the sole explanation
The original manuscript organized nearly every result around the 4C dimensions of control, commitment, challenge, and confidence. That mapping was too deterministic. For example, a five-set victory was interpreted as evidence of commitment and challenge orientation, while failure to convert match point was interpreted as attentional vulnerability. Such one-to-one mappings are not justified because the same behavioral outcome can arise from multiple processes.
The broader literature favors a pluralistic framework. Gucciardi et al. (2015) challenged the assumption that mental toughness necessarily decomposes into the 4C dimensions and found evidence for both enduring and state-varying properties. Schweickle et al. (2021) showed that clutch performance itself is conceptually heterogeneous. Hase et al. (2025) found that challenge states are associated with better performance than threat states, but with small average effects. Hufton et al. (2024) identified multiple interacting facilitators of successful pressure performance. Finally, the 2026 ESSA position statement explicitly adopts a systems-led approach and emphasizes that pressure may facilitate or impair performance depending on context and response (Schweickle et al., 2026).
5.3 Pressure response as an athlete-by-situation interaction
One of the most important implications of the tennis literature is that pressure response should be modeled as an interaction between athlete and situation. González-Díaz et al. (2012) found persistent individual heterogeneity in adjustment to point importance, supporting the possibility of stable between-player differences. Yet Schweickle et al. (2022) found that clutch performance may be situational and context dependent, and Gucciardi et al. (2015) showed that mental toughness can vary across situations and time. These findings are compatible if athletes have characteristic tendencies whose expression depends on context.
This interactionist perspective also helps reconcile apparently conflicting tennis studies. Cohen-Zada et al. (2017) found evidence of choking under competitive pressure among men, whereas Goff et al. (2025) found improved service-hold probabilities under heightened pressure. Differences in score definition, serving context, player sample, era, and statistical model can all alter the estimated pressure effect. The scientific question should therefore move from "Does pressure hurt performance?" to "For whom, under which pressure, on which component of performance, and relative to what baseline?"
5.4 Performance concentration under pressure
The present analysis proposes performance concentration under pressure as a useful descriptive concept for sports with nonlinear scoring systems. Two athletes can produce similar aggregate performance yet differ in how successful execution is distributed across states of increasing outcome leverage. Tennis is especially suitable because point importance varies continuously and can be modeled probabilistically.
This concept should not be operationalized by counting only memorable points. A rigorous measure would estimate each point's importance as the change in match-win probability associated with winning versus losing that point, conditional on server, score, surface, opponent, and baseline ability. Player performance could then be modeled as a function of point importance. A positive slope would indicate that execution improves as leverage increases; a negative slope would indicate deterioration. Random slopes could estimate individual differences, following the logic of critical ability research.
Such a model would improve substantially on career summaries. It would use every eligible point rather than selected dramatic episodes; adjust for opponent and serve advantage; distinguish pressure intensity from tournament prestige; and permit uncertainty estimates around individual pressure-response parameters. Applied to Djokovic and Federer, it could test whether the descriptive narrative survives statistical adjustment.
5.5 Alternative explanations
Several nonpsychological explanations must be considered. First, return skill matters. Djokovic's return quality can be disproportionately valuable on break points and when an opponent serves for a match. Second, physical conditioning may contribute to deciding-set success independently of psychological resilience. Third, playing style affects exposure: aggressive, serve-dominant tennis and return-oriented baseline tennis generate different distributions of tie-breaks, break points, and long rallies. Fourth, age and career phase are inseparable from the rivalry. Federer's early prime preceded Djokovic's, while many later meetings occurred when Federer was substantially older.
Fifth, opponent effects are endogenous. Reaching a fifth set or match point against another all-time great is not equivalent to doing so against an average tour player. Sixth, survivorship and selection biases can distort career aggregates. Seventh, nonlinear tennis scoring magnifies a small number of points; a player can win more points and games yet lose the match, as the 2019 Wimbledon final vividly demonstrates. Finally, random variation remains important. Even elite athletes sometimes lose high-leverage points without any psychologically meaningful failure.
5.6 Implications for applied sport psychology
Although the present study cannot diagnose the mechanisms underlying either player's record, contemporary evidence suggests several pathways through which pressure performance can be trained. A meta-analysis of pre-performance routines found benefits under pressurized conditions (Gröpel & Mesagno, 2019). The ESSA position statement highlights pressure training, implicit attentional strategies, psychological skills training, mindfulness-based approaches, and pre-performance routines as evidence-informed options (Schweickle et al., 2026). The 2026 meta-analysis of choking interventions likewise found improvements across several performance domains (Zeng et al., 2026).
The applied lesson is not to imitate a famous athlete's presumed personality. It is to identify an athlete's pressure-response profile, determine where execution deteriorates or improves, and train mechanisms appropriate to the task. In tennis, this could involve pressure simulations for serving out sets, return routines on break points, between-point attentional resets, and repeated exposure to score states that reproduce competitive consequence.
6. Limitations
The first limitation is construct validity. Archival score states are proxies for objective consequence, not direct measurements of perceived pressure. The second is causal ambiguity: behavioral outcomes cannot identify the psychological mechanism that produced them. Third, several career-level tennis statistics are not maintained in standardized official databases, requiring reliance on secondary reconstructions. Those statistics should be replicated from point- or match-level data before confirmatory publication.
Fourth, the two-case design limits generalizability. Djokovic and Federer are extreme performers whose careers cannot be assumed to represent professional tennis generally. Fifth, age, era, surface, opponent, tournament stage, injury, and playing style are incompletely controlled. Sixth, outcome measures such as five-set win percentage confound pressure response with the process that selected a match into a deciding set. Seventh, memorable matches invite hindsight bias and narrative overfitting.
These limitations define the next research design. The strongest follow-up would use complete point-level data, preregistered pressure metrics, hierarchical models with player-specific pressure slopes, and sensitivity analyses across alternative definitions of point importance. Where feasible, prospective research should pair behavioral data with athlete-reported appraisal, anxiety, confidence, and psychophysiological measures.
7. Future Research
Future work should construct a longitudinal pressure-performance database covering complete careers and multiple surfaces. Point importance should be calculated from conditional match-win probabilities rather than categorical labels alone. Multilevel models should include server, returner, opponent strength, surface, age, fatigue proxies, tournament round, and score state. The key parameter would be the player-specific change in performance as point importance rises.
A second priority is multimethod validation. Behavioral pressure-response parameters should be correlated with validated measures of mental toughness, challenge-threat appraisal, self-efficacy, and attentional control in active players. Such work would test whether archival "critical ability" actually corresponds to the psychological constructs often used to explain it.
Third, researchers should distinguish closing pressure from survival pressure. Serving at match point, returning against match point, protecting a break lead, and playing a deciding tie-break may create different appraisals and attentional demands. Federer's history itself illustrates why this distinction matters: an athlete can repeatedly save match points while also losing some matches after holding them. Treating both situations as a single pressure category obscures potentially different mechanisms.
Finally, comparative work should extend beyond Djokovic and Federer to other elite performers. The goal should not be a league table of mental toughness. It should be a taxonomy of pressure-response profiles and the conditions under which those profiles predict performance.
8. Conclusion
The Djokovic-Federer comparison is scientifically valuable precisely because it resists a simple verdict. Djokovic's competitive record contains unusually strong behavioral evidence of successful performance in several high-consequence contexts, including prolonged deciding-set contests and major matches in which defeat was imminent. Federer, however, also displays extensive evidence of pressure resilience, including elite aggregate tie-break performance and numerous recoveries from match point down. The archival record therefore does not justify portraying one player as psychologically strong and the other as psychologically weak.
The more defensible conclusion is that the two careers appear to differ in the distribution of successful performance across certain pressure-relevant contexts. Djokovic's record is particularly notable for successful outcomes in several of the most consequential situations. That pattern is compatible with explanations involving mental toughness, challenge appraisal, attentional control, confidence, and self-regulation, but it does not uniquely demonstrate any of them.
This distinction between behavioral evidence and psychological inference is the paper's central contribution. Mental toughness should not be diagnosed from scorelines, and choking should not be inferred from a missed opportunity alone. Elite performance emerges from interactions among psychological resources, technical skill, physiology, tactics, opponent behavior, context, and chance. Tennis, with its repeated and quantifiable changes in point importance, offers an exceptional natural laboratory for studying those interactions.
A rigorous science of pressure performance should therefore move beyond retrospective labels toward point-level, multilevel, and multimethod analysis. The enduring question is not simply who was "mentally tougher." It is how exceptional performers allocate, preserve, or enhance execution as consequence rises, and which psychological and contextual mechanisms explain those differences.
References
Clough, P., Earle, K., & Sewell, D. (2002). Mental toughness: The concept and its measurement. In I. Cockerill (Ed.), Solutions in sport psychology (pp. 32-43). Thomson.
Cohen-Zada, D., Krumer, A., Rosenboim, M., & Shapir, O. M. (2017). Choking under pressure and gender: Evidence from professional tennis. Journal of Economic Psychology, 61, 176-190. https://doi.org/10.1016/j.joep.2017.04.005
Goff, B., Wilson, D., & Zimmer, D. (2025). Performing in high-pressure situations: The case of tennis players up a break and serving for the set. Applied Economics, 57(39), 6138-6147. https://doi.org/10.1080/00036846.2024.2381129
González-Díaz, J., Gossner, O., & Rogers, B. W. (2012). Performing best when it matters most: Evidence from professional tennis. Journal of Economic Behavior & Organization, 84(3), 767-781. https://doi.org/10.1016/j.jebo.2012.09.021
Gröpel, P., & Mesagno, C. (2019). Choking interventions in sports: A systematic review. International Review of Sport and Exercise Psychology, 12(1), 176-201. https://doi.org/10.1080/1750984X.2017.1408134
Gucciardi, D. F., Hanton, S., Gordon, S., Mallett, C. J., & Temby, P. (2015). The concept of mental toughness: Tests of dimensionality, nomological network, and traitness. Journal of Personality, 83(1), 26-44. https://doi.org/10.1111/jopy.12079
Hase, A., Nietschke, M., Kloskowski, M., Szymanski, K., Moore, L., Jamieson, J. P., & Behnke, M. (2025). The effects of challenge and threat states on performance outcomes: An updated review and meta-analysis of recent findings. EXCLI Journal, 24, 151-176. https://doi.org/10.17179/excli2024-7995
Hill, D. M., Hanton, S., Matthews, N., & Fleming, S. (2010). Choking in sport: A review. International Review of Sport and Exercise Psychology, 3(1), 24-39. https://doi.org/10.1080/17509840903301199
Hufton, J. R., Vella, S. A., Goddard, S. G., & Schweickle, M. J. (2024). How do athletes perform well under pressure? A meta-study. International Review of Sport and Exercise Psychology. Advance online publication. https://doi.org/10.1080/1750984X.2024.2414442
Meijen, C., Turner, M., Jones, M. V., Sheffield, D., & McCarthy, P. (2020). Challenge and threat: A critical review of the literature and an alternative conceptualization. Frontiers in Psychology, 10, 1255. https://doi.org/10.3389/fpsyg.2019.01255
Schweickle, M. J., Appaneal, R. N., Jackman, P. C., Minahan, C. L., Moyle, G. M., Pickering, C., Stewart, D., Vella, S. A., & Swann, C. (2026). Performance under pressure in sport: A position statement from Exercise and Sports Science Australia (ESSA). Sports Medicine, 56(8), 1851-1867. https://doi.org/10.1007/s40279-026-02472-x
Schweickle, M. J., Swann, C., Jackman, P. C., & Vella, S. A. (2021). Clutch performance in sport and exercise: A systematic review. International Review of Sport and Exercise Psychology, 14(1), 102-129. https://doi.org/10.1080/1750984X.2020.1771747
Schweickle, M. J., Vella, S. A., & Swann, C. (2022). Was it a clutch performance? A qualitative exploration of the definitional boundaries of clutch performance. Psychology of Sport and Exercise, 62, 102238. https://doi.org/10.1016/j.psychsport.2022.102238
Sheard, M., Golby, J., & van Wersch, A. (2009). Progress toward construct validation of the Sports Mental Toughness Questionnaire (SMTQ). European Journal of Psychological Assessment, 25(3), 186-193. https://doi.org/10.1027/1015-5759.25.3.186
Zeng, Z., Yan, W., Mohammadi, A., Toumi, Z., Sun, H., & Soh, K. G. (2026). Counteracting choking under pressure in sport: A meta-analysis of interventions. Psychology of Sport and Exercise, 85, 103130. https://doi.org/10.1016/j.psychsport.2026.103130