Does Practicing the Wrong Way Really Improve Hitting? A Critical Examination of Skill Acquisition Claims

This article is an assessment of Rob Gray’s Perception-Action Podcast Episode 446: Wrong Way Study. It is not intended to serve as an Ad Hominem piece. While the lack of timing control in this study is a major flaw in its design, it stems from a gap in understanding rather than negligence. You don't know what you don't know....

Abstract

A recent study investigated whether introducing movement variability in hitting practice enhances perception-action coupling and improves swing decision-making. However, the study presents a false dichotomy by comparing only two structured training conditions, leaving a gap in testing that limits the validity of its conclusions. Without a third test group, it remains unclear whether movement variability and adaptability naturally emerge as part of a batter’s learning process or if they must be externally imposed. The design did not account for how hitters might adjust when left to self-organize, nor did it consider whether skill improvements could result purely from changes in timing rather than mechanical variability.

Instead, the study compared two externally guided training groups: a "right way" group and a "wrong way" group. The "right way" group was directed to hit the ball "the right way," which, if defined as line drives, imposed an unintended constraint on their adaptability. Additionally, this group received coaching feedback, introducing an external instructional bias that influenced their decision-making and swing adjustments. In contrast, the "wrong way" group was tasked with achieving various undesired outcomes—foul balls, pop-ups, and ground balls—but received no coaching on how to achieve these results, relying instead on self-guided exploration. The study concluded that the "wrong way" group exhibited greater improvements in slugging percentage, reduced out-of-zone swing rates, and increased variability in weight shift patterns—interpreted as evidence of enhanced adaptability.

A third test group, had it been considered, could have taken different forms. One possibility is a group that was given no prescribed objective, allowing hitters to apply their own intent without instruction or imposed constraints. Another possibility is a group that was directed to manipulate only their timing while keeping mechanics constant. This would have tested whether the "wrong way" objectives (foul balls, pop-ups, ground balls) could be achieved purely through timing errors rather than mechanical variability. If this were the case, it could suggest that timing, rather than movement variability, may be the primary driver of adaptability and improved swing decision-making. Without such a comparison, the study’s conclusions remain incomplete.

Furthermore, the study was conducted in a virtual environment, which removed real-world constraints that influence swing decisions. Notably, hitters did not face the threat of being hit by inside pitches or the confined swing space afforded by these pitches—factors that naturally affects decision-making and mechanical approaches and adjustments in live at-bats. In real-world conditions, novice batters often instinctively adjust their approach to inside pitches due to self-preservation instincts, which can impact both swing decisions and mechanics. In a VR setting, however, the inside pitch was simply a problem to solve, devoid of consequence. This means that the approach hitters developed in VR may not translate directly to real-world adaptability, further complicating the interpretation of the study’s findings.

Additionally, the absence of proprioceptive feedback in the virtual environment raises concerns about the real-world transferability of the observed improvements. This rebuttal will demonstrate that a timing-first approach, along with a more comprehensive experimental design, would have provided a more valid and scientifically sound method for optimizing hitting performance.

The study inadvertently suggests that movement variability, rather than timing stabilization, is the key to improved adaptability. This is a fundamental misunderstanding of skill acquisition. A hitter with excellent mechanics but poor timing will not make consistent contact, while a hitter with imperfect mechanics but stable timing can still produce results. There is no empirical or theoretical basis to suggest that refining mechanics without first stabilizing timing leads to better outcomes. Yet, the study fails to isolate timing as a distinct, first-order constraint, confounding movement variability with compensatory adjustments rather than true adaptability. This analysis will be examined in detail in the accompanying review, which evaluates the study’s methodology, the role of timing as a first-order constraint, and the broader implications for hitting skill development.

1. Introduction: The Core Issue with the Study

The study attempts to demonstrate that movement variability leads to improved perception-action coupling and decision-making in hitting. The study presents two training groups:

1. “Right-way” group: Coached using a predefined mechanical model and successful outcome attempts objective (hard hit line drives).
2. “Wrong-way” group: Allowed to explore movement solutions with guided mechanical variability (intentional foul balls, pop-ups, grounders).

The study concludes that the "wrong way" group outperformed the "right way" group in SLG% and swing decision quality, suggesting that mechanical variability enhances adaptability. However, this conclusion is based on a fundamental misinterpretation of motor learning principles and fails to account for timing as an independent, first-order constraint and how timing is at the core of all swing outcomes, including those tested in these two groups. Specifically:

• It presents a false dichotomy by assuming timing emerges from mechanics rather than as a solvable constraint. Timing has an equal or greater influence on batted-ball outcomes than mechanics. The same mechanical swing can produce a ground ball, pop-up, line drive, or foul ball depending on when contact is made. Because timing errors are the primary cause of mis-hits, poor swing decisions, and mechanics disruptions, improvements in decision-making should not be attributed to movement variability unless timing is isolated as a variable. The study reports that the “wrong way” group improved swing decision-making, but it does not establish whether this was due to movement variability or simply the result of fewer constraints on timing exploration. As we explore later with Bayesian learning, batters will self-organize their mechanics to optimize intent (a second-order objective) once the first-order constraint (timing) is stabilized.
• It fails to control for timing as an independent variable, confounding movement changes with compensatory adaptations. The study does not distinguish whether the "wrong way" group improved because of movement variability or simply because they experienced greater exposure to timing variation. A properly designed experiment could have included a group that manipulates only timing while keeping mechanics constant. If this group achieved the same performance gains, it would indicate that timing—not movement variability—was the driving factor behind improved adaptability. Without a third test group allowing for self-organized, intent driven adaptation or one that isolated timing as a variable, the study creates a misleading causal link between movement variability and improved decision-making.
• It relies on outcome-based performance metrics that do not validate actual movement optimization but instead reflect potential improvements enabled by greater timing and collision point exploration. The study evaluates adaptability based on SLG% and swing decision metrics, assuming these outcomes reflect improved movement adaptability, but it does not measure the process by which a batter arrives at these results The 'wrong way' group may not have improved due to movement variability—they may have simply benefitted from greater freedom to explore their own optimal timing strategies without coaching intervention, or the restrictive objectives imposed on the 'right way' group, such as the emphasis on line drives. Without biomechanical markers such as swing efficiency, attack angle refinement, or optimal contact consistency, SLG% alone does not confirm whether hitters actually improved their mechanics or simply benefited from a more adaptable timing strategy.
• It removes proprioceptive feedback by using VR, raising concerns not just about skill transfer but also about whether the study’s conditions create an apples-to-oranges comparison to real-world hitting. Hitters rely on cross-sensory integration (visual, proprioceptive, and kinesthetic feedback) to refine swing decisions. Without real bat-ball contact in VR, hitters in the study lacked proprioceptive confirmation, a critical factor in optimizing timing and mechanical efficiency. The study’s conclusions assume movement variability improves adaptability, but without real-world impact validation, it is unclear whether hitters truly improved or simply adapted to VR-specific conditions.

To truly optimize hitting performance, had timing variability been controlled as the first-order constraint, mechanics could have self-organized around a stable decision-making reference, allowing for intent-driven mechanical adjustments rather than reactionary compensations

?

2. A Missing Control Group

The study could have considered a third, more scientifically valid approach. One option might have been solving the timing constraint first, then allowing mechanical refinements and intent-driven outcomes to follow naturally.

Two objectives exist in hitting development:

1. Accurate intersections – The ability to consistently time the swing to meet the ball at the correct moment in space. This is the first-order objective of all hitters.
2. Intent-driven outcomes – The ability to manipulate attack angle, contact depth, and barrel trajectory to influence batted-ball results. This is a second-order objective, only possible once the first-order constraint is resolved.

A properly designed study could have included a third group, one instructed to achieve different batted-ball outcomes purely through timing adjustments, without altering mechanics. This group would have demonstrated that undesired outcomes—such as pop-ups, ground balls, or foul balls—often result from misted swings rather than ascribing them solely to mechanical variability. Furthermore, hitters establish early/late contact parameters through timing experimentation, which directly influences batted-ball trajectories. The natural curvature of the swing arc means that small timing shifts can significantly alter attack angle and launch angle without requiring separate mechanical modifications. While mechanical adjustments can fine-tune outcomes, decision timing is the first-order variable that determines whether contact occurs at an optimal depth.

Further complicating the study’s conclusions is that both groups were subject to a timing bias, though in different ways. The "wrong way" group’s timing adjustments were an intentional part of their prescribed variability, meaning they were actively encouraged to explore different timing patterns. The "right way" group, however, was influenced by an unrealized timing bias, imposed by the requirement to optimize for line drives in Phase 1.

Since hitting a line drive requires precise timing—as it depends on specific contact depth and swing arc alignment—these batters likely carried this reinforced timing approach into Phase 2, even when the task no longer required it. This is critical because it means the study did not truly isolate movement variability as an independent factor. The "wrong way" group had already experimented with varied contact depths, timing shifts, and movement variability, which made them more adaptable in Test 2—not necessarily because of movement variability, but because their approach was not constrained by a fixed timing strategy.

Without a true control group (i.e., one without imposed constraints or one that imposed timing changes rather than intentional mechanical changes), it is impossible to determine whether either method was superior to natural skill acquisition. The results cannot distinguish whether adaptability was truly a function of movement variability or if it was simply a byproduct of timing flexibility in the "wrong way" group.

Additionally, the "right way" group was not just restricted in movement variability but also in attack angle adaptability. Because their training forced them to optimize for line drives, their ability to adjust to higher launch angles may have been inherently limited in Test 2. The "wrong way" group, by contrast, had already explored a range of attack angles and timing strategies, which biased them toward higher SLG% outcomes—not because of superior movement exploration, but because of a specific timing variability and mechanical advantage.

This means the study’s conclusion—that mechanical variability improved perception-action coupling—is misleading. The results do not necessarily reflect improved adaptability but instead highlight an unintended launch angle bias that naturally favored the "wrong way" group in Test 2. If the "right way" group had trained across a broader range of attack angles, their SLG% would likely have improved as well.

The Correct Approach

• Establishing a stable timing reference first would have eliminated unnecessary mechanical variability, allowing for natural, intent-driven optimizations rather than reactionary compensations.
• A properly structured study could have included a third control group that isolated decision timing as the only manipulated variable. This group could have been instructed to mistime swings (too early, too late) without altering their mechanical approach. The purpose of this control would be to determine whether undesired outcomes—foul balls, pop-ups, and grounders—could be induced purely through decision timing errors, rather than through mechanical variability.

Why This Matters

If undesired outcomes can be created without movement variability, then the claim that mechanical exploration improves perception-action coupling is incomplete. It would suggest that timing adaptations alone may be sufficient to produce the observed performance improvements in the "wrong way" group. Without this control, the study fails to distinguish whether mechanical variability was necessary or if timing exploration was the real driver of improved decision-making and SLG% outcomes.

3. Failure to Isolate Timing as an Independent Variable

Major Methodological Flaw: The study never defines nor stabilizes timing as a measurable, first-order constraint.

Why This is a Problem:

• Timing in hitting is not just a biomechanical issue—it is an interception problem.
• A hitter must predict and align their movement with an external event (pitch arrival) before refined mechanics or intent driven outcomes even become relevant.
• Hitters do not rely solely on visual input—they use internal proprioceptive reference points to confirm their decision timing.

The Study’s Logical Fallacy:

The study assumes that movement variability is the primary driver of improved batted-ball outcomes, when in reality, timing may have a greater influence on these outcomes, as it is the mechanism that allows for successful mechanical adjustments. This misinterpretation stems from a failure to recognize the role of timing variability in the learning process. By treating movement variability as an independent factor, the study overlooks how adjustments in timing can directly influence mechanics and overall adaptability.

Skill acquisition follows an iterative refinement process governed by Bayesian learning principles and the calculus of variations. In this process, hitters adjust their swing in response to prior successful intersections, progressively reducing unnecessary variability while retaining functionally useful adaptations. Timing serves as the fundamental stabilizing factor in this process. Without properly accounting for timing variability, observed mechanical variability may not indicate improved adaptability—it may simply reflect a reaction to a resolved or unresolved first-order constraint rather than a deliberate mechanical refinement.

The study also fails to recognize that timing instability produces both positive and negative mechanical changes. When timing is inconsistent, hitters usually compensate with mechanical adjustments, rather than changes in their timing decisions, some of which may be inefficient rather than productive. This means that observed movement variability may have been a reaction to timing adaptations rather than an independent driver of adaptability.

For instance, a batter consistently making contact too early on inside pitches may pull the ball foul. This can occur in two ways, both grounded in timing. First, the same mechanics that drive the ball to center field—when mistimed—could instead result in a pulled foul ball. Second, the same too-early timing, combined with an over-rotated torso ("flying open"), could also produce the same foul-pull outcome.

This distinction is critical: in the first case, timing alone changes the outcome, while in the second case, timing errors induce a mechanical flaw. If timing was first stabilized, the second-order mechanical issues may never arise. This highlights why mechanics cannot be meaningfully analyzed without first isolating timing variability.

Without a control group designed to test timing as the only manipulated factor, the study cannot determine whether decision-making improvements resulted from movement variability or simply greater exposure to timing variations.

The study only demonstrates that a less constrained group outperformed a more constrained group, but this does not confirm that movement variability is inherently a better training method. Group 1 was trained to optimize for line drives, reinforcing a stable timing strategy that resulted in a rewarding outcome they were unlikely to abandon in Phase 2. Group 2, however, was encouraged to explore different methods—including both mechanical and timing adjustments—to achieve their assigned outcomes. As a result, adaptability improvements in Group 2 may have stemmed from timing flexibility rather than movement variability, but the study does not test this distinction.

The study’s control group followed conventional training methods, which inherently involve both mechanical and timing adaptations. However, this does not isolate timing as an independent factor. A properly structured control group would have tested timing as the only manipulated variable, allowing a clearer distinction between the roles of movement variability and timing flexibility in improving decision-making.

Without a third group that isolates timing adjustments while keeping mechanics constant, the study’s conclusions remain incomplete. It does not determine whether movement variability was a necessary factor in improved decision-making or if timing calibration alone influenced outcomes or could have produced similar or superior results.

How This Could Have Been Handled:

• Establish a consistent decision-timing reference before introducing mechanical refinement. This would ensure that all batters in both groups were equally comfortable with their intersection timing before being asked to manipulate secondary factors like movement variability.
• Measure mechanical efficiency under a stabilized timing condition rather than testing movement variability under unstable constraints. This would isolate the effects of movement variability, preventing compensatory mechanical adjustments caused by timing uncertainty.

4. The Missing Role of Proprioceptive Reference Points in Hitting

Critical Oversight:

• By conducting the study in a virtual environment, this reduces the hitters to relying solely on visual information (TAU, optic flow) to predict pitch timing.
• However, hitting is not purely a visual-perceptual task—it is a multisensory process where proprioceptive feedback plays a decisive role in skill acquisition.
• Proprioceptive Confirmation in Swing Learning:
• When a ball contacts the bat, hitters immediately register where impact occurred relative to the sweet spot as well as the hands’ and bat’s orientation within the swing’s arc.
• This proprioceptive confirmation allows hitters to encode successful reference points, memorizing optimal swing execution while discarding unsuccessful attempts.
• Future swings are not calibrated purely against external visual cues but against an internalized spatial-temporal reference. This allows for fine-tuned, subconscious swing adjustments and aligns with Bayesian models of skill acquisition, where the brain optimizes movement based on successful past interactions.

The study does not fully account for this process, as it assumes that movement variability itself is the primary learning mechanism. However, if movement variability were the key factor, then the right-way group—who successfully met their outcome objective (line drives in fair play) and demonstrated improvements in batting average and strikeout rate—should have seen comparable improvements in SLG% and swing decision-making in Phase 2. However, because this group was constrained to an idealized model in Phase 1, they were unlikely to deviate from that learned pattern in Phase 2. Their reduced adaptability suggests that exposure to varied launch angles and timing adjustments—rather than mechanical variability alone—was the true differentiator in the wrong-way group's superior performance.

Furthermore, without proprioceptive confirmation in the VR environment, hitters in both groups lacked a fully formed internalized reference, further reducing the real-world applicability of the study’s conclusions. This raises concerns about whether observed improvements in adaptability were genuine or if they were simply adjustments to the artificial constraints of the VR setting.

Why This Matters:

• Hitters develop an internalized map of optimal contact points—something the study does not account for.
• This reference system allows decision timing to be quantitatively mapped against known spatial-temporal markers, making timing a solvable constraint.
• Without proprioceptive validation, the study assumes that hitters were purely reacting to visual cues when, in reality, the brain integrates multiple sensory inputs to optimize movement precision.

Correct Approach:

• Recognize that proprioception is not just supplementary—it is essential for decision calibration.
• Include real bat-ball interaction or haptic feedback devices to measure how decision-timing cues translate into real-world movement patterns.
• Acknowledge that movement variability should be analyzed in relation to stable timing references, not as an uncontrolled adaptation to unstable constraints.

4. The Missing Role of Proprioceptive Reference Points in Hitting

Critical Oversight:

• By conducting the study in a virtual environment, this reduces the hitters to relying solely on visual information (TAU, optic flow) to predict pitch timing.
• However, hitting is not purely a visual-perceptual task—it is a multisensory process where proprioceptive feedback plays a decisive role in skill acquisition.
• Proprioceptive Confirmation in Swing Learning:
• When a ball contacts the bat, hitters immediately register where impact occurred relative to the sweet spot as well as the hands’ and bat’s orientation within the swing’s arc.
• This proprioceptive confirmation allows hitters to encode successful reference points, memorizing optimal swing execution while discarding unsuccessful attempts.
• Future swings are not calibrated purely against external visual cues but against an internalized spatial-temporal reference. This allows for fine-tuned, subconscious swing adjustments and aligns with Bayesian models of skill acquisition, where the brain optimizes movement based on successful past interactions.

The study does not fully account for this process, as it assumes that movement variability itself is the primary learning mechanism. However, if movement variability were the key factor, then the right-way group—who successfully met their outcome objective (line drives in fair play) and demonstrated improvements in batting average and strikeout rate—should have seen comparable improvements in SLG% and swing decision-making in Phase 2. However, because this group was constrained to an idealized model in Phase 1, they were unlikely to deviate from that learned pattern in Phase 2. Their reduced adaptability suggests that exposure to varied launch angles and timing adjustments—rather than mechanical variability alone—was the true differentiator in the wrong-way group's superior performance.

Furthermore, without proprioceptive confirmation in the VR environment, hitters in both groups lacked a fully formed internalized reference, further reducing the real-world applicability of the study’s conclusions. This raises concerns about whether observed improvements in adaptability were genuine or if they were simply adjustments to the artificial constraints of the VR setting.

Why This Matters:

• Hitters develop an internalized map of optimal contact points—something the study does not account for.
• This reference system allows decision timing to be quantitatively mapped against known spatial-temporal markers, making timing a solvable constraint.
• Without proprioceptive validation, the study assumes that hitters were purely reacting to visual cues when, in reality, the brain integrates multiple sensory inputs to optimize movement precision.

Correct Approach:

• Recognize that proprioception is not just supplementary—it is essential for decision calibration.
• Include real bat-ball interaction or haptic feedback devices to measure how decision-timing cues translate into real-world movement patterns.
• Acknowledge that movement variability should be analyzed in relation to stable timing references, not as an uncontrolled adaptation to unstable constraints

?

5. Why the VR Model Fails to Replicate Real-World Hitting

1. VR Creates an Artificial Training Bias

• In real-world hitting, decision timing is refined through cross-sensory integration—a blend of visual, proprioceptive, and kinesthetic feedback.
• The VR model removes proprioceptive reinforcement, forcing hitters to rely solely on visual cues. This creates an artificial training environment where solution strategies may be specific to the VR setting rather than transferable to real-world hitting.
• Without real bat-ball contact, swing adaptations may have been influenced by the limitations of the VR system rather than genuine improvements in adaptability.
• Additionally, the absence of real danger and consequences eliminates a key real-world constraint that influences swing decisions. In live at-bats, inside pitches create an inherent self-preservation instinct that can lead to mechanical changes, even when unintended. This factor was entirely absent in the VR environment, meaning the approach to inside pitches was not influenced by the threat of getting hit or the reduction of physical swing space. Instead, hitters simply solved the problem of the inside pitch in a consequence-free setting, which may not translate to real-world adaptability.

1. How This Affects the Study’s Conclusions

• If decision-making was adjusted based on VR-specific cues, then the observed improvements may not transfer to real-world hitting.
• Performance changes in the variable training group may have resulted from adaptations to VR’s constraints rather than from movement variability itself.
• In a real hitting environment with full sensory feedback, hitters may have developed entirely different movement solutions—meaning conclusions about adaptability may not be ecologically valid.
• Any claim that movement variability enhanced adaptability is uncertain if the training environment itself imposed artificial constraints that shaped the observed results.

?

6. Conclusion: Addressing the Missing Variables in Hitting Skill Acquisition

The study’s conclusions regarding adaptability and skill acquisition remain incomplete because it never established a control group of any kind. Without an unstructured baseline for comparison, it is impossible to determine whether the observed improvements resulted from movement variability, timing flexibility, or a combination of both. Additionally, the conditions imposed on both test groups inherently shaped their outcomes, introducing unavoidable biases into the results. Because Group 1 was constrained to an idealized hitting model while Group 2 was encouraged to explore multiple swing variations, the study did not evaluate movement variability in an objective context—it merely tested how hitters adapted within the limitations imposed on them. This means the results reflect the influence of external constraints rather than a pure measure of adaptability. Whether recognized or not, timing played an inadvertent role in shaping the study’s outcomes, further complicating the interpretation of its findings. What was framed as a ‘revolutionary’ approach was, in reality, a structured experiment that predetermined the conditions under which variability would be considered successful, rather than an unbiased examination of its effectiveness.

Final Key Takeaways:

1. The study never established a control group, making it impossible to isolate the true drivers of adaptability. Without an unstructured test group or a timing-only group, the study cannot determine whether movement variability, timing flexibility, or a combination of both led to improved outcomes. Because the right-way group was constrained to an idealized model while the wrong-way group explored multiple attack angles, the conditions shaped the outcomes rather than objectively evaluating adaptability.
2. Timing is a quantifiable constraint that should have been controlled first. If timing was inconsistent, any mechanical variability observed may have been a reaction to unstable constraints rather than an indicator of improved skill. Hitters naturally adjust mechanics in response to timing errors, but this does not mean mechanics themselves were driving adaptability. Without a test group that manipulated only timing, it remains unclear whether timing flexibility—not movement variability—was the primary factor behind improved decision-making.
3. Outcome metrics (SLG%, swing decisions) do not validate mechanical improvement. SLG% reflects the outcome of batted balls but does not measure whether a hitter’s swing became more efficient. A hitter could mistime a pitch but still produce a high SLG% if the misalignment happened to result in a favorable launch angle. The right-way group was artificially constrained to line drives, while the wrong-way group explored multiple attack angles. This means the wrong-way group was already exposed to a broader range of batted-ball trajectories, which may have given them an advantage in Phase 2 that had nothing to do with movement variability itself. Without biomechanical markers such as swing time efficiency, attack angle refinement, or optimal contact consistency, SLG% cannot be considered proof of mechanical improvement.
4. The study’s VR-based approach lacks proprioceptive feedback, raising concerns about real-world transferability. Hitters in the study did not receive real bat-ball contact, meaning they never experienced the vibrational node feedback necessary to reinforce precise barrel accuracy or proprioceptive orientation confirmation. The fine motor adjustments required to optimize collision efficiency are absent in a virtual setting, meaning the movement solutions identified in the study do not reflect real-world constraints. Without proprioceptive reinforcement, hitters were forced to rely solely on visual cues, which do not provide the full sensory feedback required for precise swing calibration. Movement adjustments that would naturally emerge in a real-world environment may have been suppressed or distorted because hitters lacked impact confirmation. The absence of real danger eliminates a key real-world constraint that influences swing decisions. In live at-bats, inside pitches create an inherent self-preservation instinct that can lead to mechanical changes, even when unintended. This factor was entirely absent in the VR environment, meaning the approach to inside pitches was not influenced by the threat of getting hit. Instead, hitters simply solved the problem of the inside pitch in a consequence-free setting, which may not translate to real-world adaptability.
5. TAU alone does not fully explain hitting precision—proprioceptive confirmation must be accounted for. Because the study was limited to a visually dominant model, the full range of movement solutions available to hitters was not accounted for. Proprioceptive feedback is necessary for stabilizing both timing and mechanics. Without it, hitters are forced to rely exclusively on visual cues, which do not provide the full sensory reinforcement needed for precise swing calibration. Movement adjustments that would naturally emerge in a real-world environment may have been suppressed or ignored because hitters were unable to receive real bat-ball impact confirmation. By failing to integrate proprioceptive inputs, the study may have artificially constrained and/or afforded the types of mechanical adaptations observed that may not have been viable under real hitting conditions, leading to an incomplete understanding of how hitters refine their swings within an ecologically valid environment.

Implication: Without a properly structured control group, the study does not prove what it claims—it may have only measured how hitters reacted to unstable constraints rather than true movement adaptability. This is why Bayesian learning supports a timing-first approach—without stabilizing the temporal reference, skill acquisition becomes inefficient. A true evidence-based training approach should align with how athletes refine predictive accuracy rather than relying on movement exploration without a stable timing anchor.