Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. The title begins with a redundant letter "T". This should be corrected.
2. In the Participants section, the authors state that ethics committee approval was not required, whereas the Ethics Statement and IRB Statement sections provide a protocol number approved by an institutional committee. These statements are contradictory and must be reconciled and clarified.
3. With n = 13, the absence of an a priori power analysis seriously undermines the reliability of the findings. Authors should either conduct such an analysis or provide a thorough discussion of its absence and the resulting interpretive limitations.
4. Obtaining values of ρ = –0.91 and R² = 0.83 with n = 13 raises legitimate concerns about overfitting. Authors should apply a validation procedure (e.g., bootstrapping) or explicitly state that all findings are strictly exploratory.
5. All measurements were conducted in the same fixed order (prone → kneeling → standing). The standing position was always last and shows the strongest associations with asymmetry, which may partly reflect cumulative fatigue rather than a true asymmetry effect. A sensitivity analysis should be conducted, or this issue should be discussed far more thoroughly as a limitation.
6. In rifle shooting, the dominant side does not necessarily correspond to the side supporting the weapon. Authors should clarify how handedness was translated into limb classification for the purposes of the asymmetry index.
7. Missing legend in Table 3 – The symbols * and ** are not explained. A footnote defining the significance levels should be added.
8. Unreadable heatmap (Figure 1) – The ρ values are invisible in many cells. The figure should be resubmitted at higher resolution with improved legibility.
9. The Discussion section is overly descriptive. Authors should provide a more in-depth explanation of the biomechanical mechanisms through which shoulder flexion asymmetry could lead to rifle instability in the standing position.
10. Separators between some author names are missing in the title page header and should be added.
11. In several places, spaces before citation brackets are missing. Formatting should be corrected throughout.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors investigated the association between upper-limb inter-limb strength asymmetry and rifle shooting performance in elite rifle athletes across prone, kneeling, and standing positions, with the aim of understanding whether strength imbalances could be meaningfully linked to performance outcomes in high-level shooting. Thirteen elite rifle shooters completed standardized 20-shot series in each position according to ISSF regulations, while the authors assessed bilateral maximal isokinetic strength of the wrist, elbow, and shoulder at 60°/s using a Biodex System 4 dynamometer, together with handgrip strength measured via a handheld dynamometer.

They then calculated inter-limb asymmetry indices for each joint action and recorded position-specific shooting scores as well as the number of good-10 hits (≥10.0). The authors examined associations between asymmetry indices and performance outcomes using Spearman correlation analyses and simple linear regression models.

Their results showed that greater asymmetry in selected shoulder actions was strongly associated with lower performance scores across prone, kneeling, and standing positions, while standing performance was additionally negatively associated with wrist flexion asymmetry. The number of good-10 hits was positively associated with standing score and negatively associated with both shoulder and wrist asymmetry. Regression analyses further indicated that inter-limb asymmetry explained a substantial proportion of the variance in position-specific scores (54–83%), whereas standing score accounted for 58% of the variance in good-10 hits.

Overall, the authors conclude that position-specific upper-limb strength asymmetry, particularly at the level of the shoulder and wrist, is meaningfully associated with rifle shooting performance and may represent an important factor to consider in the training and monitoring of elite rifle athletes.

 

The study is interesting and contributes to the scientific literature:

 

Strengths:
The topic is both intriguing and highly relevant, with clear practical implications for sport performance and training. The study is developed with solid scientific rigor and demonstrates good methodological control and domain expertise.

Weaknesses:
Some aspects related to editorial presentation and clarity of exposition could be improved to enhance readability. In addition, careful attention should be paid to the correct and consistent application of MDPI formatting and stylistic standards.

Specific comments:

There is a typographical error in the title, where an extraneous “T” appears and should be corrected.

In the abstract, the first part does not clearly correspond to a standard “Background” section and could be better aligned with conventional structure.

The introduction is clearly structured and easy to follow, with a logical progression connecting postural stability, upper-limb neuromuscular control, and rifle shooting performance in precision sports. However, some concepts are repeated across sections, and a more concise presentation would improve readability and help the main message emerge more clearly.

The research gap is generally well identified and appropriately framed. Nevertheless, it could be made more explicit and compelling. In particular, inter-limb upper-limb asymmetry appears somewhat gradually, whereas it would benefit from being introduced earlier and more directly as the central unresolved issue, thereby strengthening the perceived novelty and motivation of the study.

The Methods section does not follow a fully standardised structure. While all necessary information is provided, the presentation is somewhat fragmented, with short subsections and a list-like format. Although this is not a methodological flaw per se, a more integrated structure would improve readability. A flowchart could further help readers better understand the study design.

The same issue is present in the Results section. Although the organisation into thematic subsections is appropriate, the rationale behind the structure could be clarified more explicitly. In addition, figures and tables appear at times to lack refinement; for example, Tables 2 and 3 could be improved in clarity and presentation.

Moreover, the manuscript relies heavily on figures and tables, while the accompanying narrative analysis is relatively limited. Expanding the textual interpretation of the results would strengthen the scientific depth and overall impact of the study.

The Discussion and Conclusions sections, however, are strong and effectively enhance the overall value and impact of the work. They significantly improve the interpretative depth of the study and contribute positively to the final evaluation.

 

Author Response

please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The paper presents a cross-sectional, observational study examining the association between upper-limb inter-limb strength asymmetry (assessed via isokinetic dynamometry at 60°/s and handheld dynamometry for handgrip) and rifle shooting performance across the three ISSF positions (prone, kneeling, standing) in 13 elite rifle athletes. Performance was indexed by 20-shot totals per position and the count of "good-10" hits (≥10.0). The authors report strong negative Spearman correlations between selected joint-specific asymmetry indices and position-specific scores and conclude that asymmetry - particularly at the shoulder and wrist - may represent a meaningful target for monitoring and conditioning in elite shooters.

The research question is timely and relevant, the access to an elite athlete cohort is genuinely valuable, and the manuscript addresses an under-investigated dimension of precision shooting. However, the study suffers from several substantive methodological, statistical, and reporting issues that, in their current form, prevent the manuscript from supporting the conclusions drawn.

These are detailed below:

1. Statistical inference is not commensurate with the sample size. With n = 13, the manuscript reports approximately 22 Spearman correlations (Table 3) and additional t-tests, with no correction for multiple comparisons (Bonferroni, Holm–Bonferroni, or FDR). At α = 0.05 uncorrected, several of the "significant" findings (e.g., correlations with p around 0.013–0.026) would not survive even a modest correction. The authors must (a) declare a primary, pre-specified analysis, (b) clearly label the remaining tests as exploratory, and (c) report adjusted p-values or, at minimum, 95 % bootstrap confidence intervals for the Spearman coefficients. Given the sample, percentile bootstrap CIs (10,000 resamples) would be more informative than point estimates.
2. Reported R² values of 54–83 % are very likely inflated by overfitting and selection bias. The regression models in Table 4 are not pre-specified; rather, they appear to have been selected from the predictors that yielded the strongest Spearman correlations in the same dataset. This circularity systematically inflates R². With n = 13 and one predictor, the adjusted R² and a leave-one-out cross-validated R² (or a stratified k-fold equivalent) should be reported. I would expect a non-trivial drop in explained variance, and the discussion must be re-calibrated accordingly.
3. Circularity in the good-10 hits ↔ standing score relationship. The regression "Good-10 hits ~ Standing score" (R² = 0.579) is structurally problematic: good-10 hits are derived from the same shot series that produces the standing score (and, presumably, also from the prone and kneeling series). Standing score is itself partly a count-based summary of high-scoring shots. The relationship is therefore tautological to a substantial degree. Please either remove this model or replace it with an analysis where good-10 hits are computed from shots OTHER than those contributing to the predictor (e.g., prone score predicting kneeling/standing good-10s).
4. Conceptual treatment of "asymmetry" is inadequate for the task. Rifle shooting is intrinsically a bimanually asymmetric task: one hand operates the trigger, the other supports the fore-end; the body is rotated; loading on the shoulder complex is left/right-different by design. The implicit assumption that "lower asymmetry → better performance" therefore requires explicit justification. Two specific issues:
   1. (a) The asymmetry index used, |F_dom − F_nondom| / max(F_dom, F_nondom), discards directional information. It is therefore impossible to tell from the analysis whether the supporting limb being stronger, or the trigger limb being stronger, drives the observed associations. A signed asymmetry index — or, equivalently, separate dominant/non-dominant strength values entered into a partial-correlation framework — would be far more informative and would also align with current recommendations (cf. Bishop et al. 2018, 2023, which the authors already cite).
   2. (b) The construct "dominant vs. non-dominant" is itself ambiguous for rifle shooters, because the "dominant" hand for self-report may not be the rifle-shouldering side. Please clarify how dominance was operationalized (handedness vs. shooting-shoulder), and report the distribution. The 7:6 left:right split is unusual and warrants comment, as it likely reflects shooting-side rather than general handedness.
5. Internally inconsistent pattern of correlations across positions. Across the three positions, the same joint action sometimes correlates positively, sometimes negatively, and sometimes not at all with score (e.g., shoulder flexion asymmetry: ρ = +0.45 prone, −0.17 kneeling, −0.91 standing). In a small sample without multiple-comparison control, such an inconsistent pattern is suggestive of chance findings rather than a stable underlying mechanism. The Discussion does not acknowledge this; instead, each isolated correlation is given a plausible biomechanical interpretation post-hoc. A more critical engagement is required — either a coherent, position-specific biomechanical model that predicts the observed sign reversals a priori, or an explicit recognition that the cross-positional pattern may not be robust.
6. Ethics statement is contradictory. On p. 4 the authors state that "formal approval from an institutional ethics committee was not required under local regulations" and that the procedures were "reviewed and authorized by the team's internal performance support department." However, on p. 5 (Ethics Statement) and again in the Institutional Review Board Statement (p. 11) they state that the study "was approved by the Institutional Ethics Committee of Beijing Sport University (protocol code: 202561211)." These statements are mutually exclusive and must be reconciled. If formal IRB approval was obtained, the earlier paragraph should be removed; if not, the IRB statement must be corrected. This is a non-negotiable point for an ethics-sensitive review process.
7. Confounding by sex is not addressed in primary analyses. Table 5 documents a large effect-size difference between male and female athletes in discipline-specific shoulder abduction asymmetry (d ≈ 1.7 by eye), yet the primary correlation and regression analyses pool sexes. Given n = 7 males / 6 females, formal stratification is impractical, but the authors should at minimum (a) present scatterplots with sex-coded markers for the main regressions of Table 4, and (b) report partial Spearman correlations adjusted for sex (where feasible), or explicitly discuss this confound as a major caveat.
8. Fixed testing order and missing fatigue control. The fixed prone → kneeling → standing order is acknowledged but not analytically addressed. Because standing yields the strongest associations and is also tested last, accumulated fatigue may inflate (or distort) standing-specific findings. Even within a 13-athlete sample, the authors could test whether time of testing relative to start correlates with any performance variable. Counterbalancing in a future study is essential.
9. Isokinetic 60°/s does not represent the "quasi-static" rifle-holding demand. The authors justify the choice of 60°/s on grounds of being "controlled, quasi-static." This is biomechanically inaccurate: 60°/s is a moderate isokinetic velocity, not isometric. Rifle holding is fundamentally an isometric (or near-isometric) postural task. Isometric peak torque or rate-of-force-development measurements would have been more ecologically valid. Please justify the velocity choice with reference to reliability or precedent in the shooting literature, or acknowledge this as a limitation.
10. "Discipline-specific shoulder abduction/adduction in a rifle-support posture" is under-described. This is presented as a methodological strength, but the joint angles, lever-arm setup, body stabilization, and exact posture geometry are not provided. Without this detail, the test is not reproducible. A figure showing positioning on the dynamometer would substantially strengthen the methods.
11. The Statistical Analysis subsection contains placeholder artefacts: "Model fit was evaluated using, F-statistics..." and "Effect sizes for between-group comparisons were expressed as Cohen’s. The significance level was set at ..." - symbols (R², d, α = 0.05) are missing. 
12. Table 1: "Mean SD Min Max" - please add the ± symbol or use proper column headers; "Training years" should read "Training experience (years)" for consistency with the text.
13. Figure 1 (Spearman heatmap): the figure is low-resolution, several cells are unreadable, axis labels are truncated ("SHFlex," "WrFlex"), and the color scale is not explicitly anchored to the value range. Please regenerate at ≥600 DPI, annotate all cells with ρ values to two decimals, mark statistical significance (e.g., asterisks for p < 0.05), and provide a clear, anchored diverging colormap (e.g., −1 to +1 centered at 0).
14. Figures 2 and 3: include 95 % confidence bands around the regression line, individual residuals, and identify sex with marker shape/color. Add the regression equation and R² to the panel.
15. Methods: please specify (i) whether isokinetic/handgrip testing and shooting assessment occurred on the same day or different days; (ii) the warm-up duration; (iii) test–retest reliability of the asymmetry indices in your laboratory or, failing that, published reliability estimates for the Biodex System 4 at 60°/s for the relevant joint actions.
16. Reference 14 (Bishop et al., 2018) is cited in the Methods section context that does not match the content of that paper. Please verify all in-text citations against the reference list; the Introduction also reads as if it occasionally pairs citations with claims they only partially support (e.g., the archery references in lines 80–85).
17. The Conclusions overstate the evidence: "Upper-limb inter-limb strength asymmetry is strongly associated with rifle shooting performance" should be tempered ("...was associated, in this small sample of elite shooters, with...") and the causal/training implications removed pending longitudinal data.
18. The Abstract conclusion uses the word "may" appropriately, but the phrase "strongly associated" in the Conclusions and Discussion should be matched to confidence-interval-bounded effect sizes rather than to point estimates from n = 13.
19. Abbreviations list includes "COP" but center-of-pressure is not actually used as a measured variable in the manuscript (only cited from prior work). Either include COP-related discussion explicitly or remove from abbreviations.

Author Response

please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the careful revision. Reading the v2 alongside my round-1 comments, the authors have engaged substantively with most of the points I raised: the abstract now consistently frames the analyses as exploratory; the methods follow STROBE more closely and include a recruitment flow diagram; the order effect (prone → kneeling → standing), the dominance-vs-supporting-limb interpretation issue, the absence of an a priori power calculation, and the reliability situation are all acknowledged openly; an FDR correction has been added to Table 3; and the good-10 hits / standing score overlap is now flagged. Taken together, the manuscript is considerably more honest about what the data can and cannot support than the v1 was, and it reads more like an exploratory profiling study of a small elite cohort — which is, in my view, the appropriate framing.

Outstanding issues

1. Selective reporting in the correlation table. Table 2 lists 11 asymmetry indices, but Table 3 and Figure 4 report correlations for only five of them (shoulder flexion, extension, adduction, wrist flexion, wrist extension). The omitted indices — including handgrip, elbow flexion/extension, shoulder internal/external rotation, and notably the discipline-specific shoulder abduction — were measured and would be informative regardless of whether their correlations reached significance.
2. The discipline-specific shoulder abduction is highlighted as a key measure but not analysed against performance. The authors describe the rifle-support-posture abduction/adduction test in some detail (Figure 3 and surrounding text), and it features prominently in the sex-difference analysis, yet its association with prone/kneeling/standing scores is never reported. Given that this is the most ecologically valid joint action in the protocol, its absence from the main analysis is conspicuous. Please add these correlations.
3. The standing-score / good-10 hits regression remains problematic. The authors now warn that the two variables are "partially derived from overlapping shot outcomes," which is welcome, but the model is still presented in Table 4 with R² = 0.579 as if it conveys predictive information. Because good-10 hits is, by construction, a count derived from the same shot decimals that produce each position score, this is closer to a structural identity than to an empirical association — particularly for the standing component, which contributes one third of the good-10 count. I would either remove this model or replace it with the alternative test that actually adds information (e.g., good-10 hits regressed on standing score after partialling out the standing series itself, or good-10 hits from prone + kneeling positions only against standing score).
4. The handedness ratio is unusual and worth a sentence. Seven of thirteen participants are reported as left-handed, against a population base rate of roughly 10%. In an elite shooting cohort this may genuinely reflect selection — left-handers may be over-represented at elite level, or "handedness" here may have been conflated with shooting side. Please clarify how handedness was determined and, if it diverges from shooting side, which one was used to assign dominant/non-dominant in the asymmetry index.

Author Response

Please see the attachment

Author Response File: Author Response.docx