Investigation of bench press muscle activity and kinematic parameters under stable and unstable load conditions

Abstract

Background

Unstable load training (ULT) has become increasingly popular for its potential to enhance muscle activation, strength, and neural development. Unlike unstable surface training (UST), ULT introduces instability through an unsteady load. Studies suggest ULT increases muscle activation in both stabilizing and primary muscles, particularly in multi-joint exercises. This study aims to assess whether ULT in the bench press improves stabilizer muscle activation, identify the most effective ULT methods, and examine their impact on kinematic parameters.

Methods

Sixteen male participants (age: 25 ± 3.14 years, body mass: 85.05 ± 5.39 kg, height: 1.78 ± 0.06 m, lifting experience: 6.75 ± 1.56 years, and bench press 1RM: 127.31 ± 13.45 kg) performed the bench press exercise under four different load conditions: Standard bar with a stable load (75% of 1RM), Standard bar with a stable load (60% of 1RM), Standard bar with an unstable load (60% of 1RM), Flexible barbell (60% of 1RM). Displacement, velocity, movement duration, and time parameters were compared under four different conditions. Surface electromyography was used to detect the activity of the pectoralis major, triceps brachii, anterior deltoid, lateral deltoid, rectus abdominis, and external oblique muscles.

Results

The Stable Load 75% condition exhibited the highest muscle activation across most muscles. It showed greater activation than Unstable Load, Stable Load 60%, and flexible barbell in the Pectoralis Major, Triceps Brachii, Rectus Abdominus, External Oblique, and Lateral Deltoid (p < 0.002). In the Anterior Deltoid, the Standard Unstable Load (97.02%) showed higher activation than Stable Load 60% (76.84%) (p < 0.001), with no significant difference to flexible barbell (89.49%). The flexible barbell showed greater displacement (cm) (eccentric: 39.84 ± 4.5, concentric: 40.17 ± 4.62) and velocity (m/s) (eccentric: 0.62 ± 0.11, concentric: 0.54 ± 0.11) compared to the Standard Unstable Load and Stable Load 75% (p < 0.005). The Standard Unstable Load (eccentric: 1.14 ± 0.61, concentric: 1.07 ± 0.42) and Stable Load 75% (eccentric: 0.87 ± 0.28, concentric: 0.95 ± 0.21) had the longest movement duration (s), while flexible barbell (eccentric: 0.64 ± 0.15, concentric: 0.72 ± 0.13) exhibited the shortest durations (p < 0.000).

Conclusion

This study demonstrates that Stable Load 75% resulted in the highest muscle activation across major muscle groups. Muscle activation depends on both the load and the phase of movement. Unstable Load Training (ULT) using a flexible barbell enhances movement dynamics by increasing displacement and velocity but does not show a significant difference from Stable Load 60% in most kinematic parameters. Therefore, load selection should align with training goals, whether the aim is to maximize muscle activation or improve movement efficiency.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13102-025-01280-6.

Introduction

In recent years, training on unstable surfaces or with unstable loads has gained popularity in many fitness centers. It is thought that instability training leads to greater activation of stabilizing muscles and offers more advantages for sport performance and daily tasks [1, 2]. When a detailed literature review is conducted in this area, it is observed that studies yield different results. Research indicates that resistance training performed under unstable conditions may result in an increase in muscle activation compared to traditional training, may show no difference, or may even result in a decrease [3–8]. The varying results may be related to the type of instability device used, the difference in workload, the function of the muscle, and other relevant factors [9–13]. Additionally, some studies have shown a decrease in peak force output when training on an unstable surface [3, 14]. The deficits in force output observed during training on an unstable surface are one of the main reasons why training on unstable surfaces is not recommended for an athletic population [4, 15].

Unstable load training (ULT) serves as an alternative to unstable surface training (UST). The key distinction between UST and ULT lies in the location and manner in which instability is introduced. In UST, instability is created by positioning the body on an unstable training surface, such as performing a squat on a BOSU ball. In contrast, ULT involves instability through an unsteady load that the body must control [16]. It appears that ULT has become a more common practice in strength and conditioning due to its potential to enhance strength and neural development [1, 2, 17]. The general assumption is that having to control unstable loads will increase the activation of stabilizing muscles during multi-joint exercises and strengthen them more effectively than traditional loads [6].

When reviewing the literature, there are few studies related to training with unstable loads. In these limited studies, it has been reported that unstable loads produce higher levels of muscle activation in the upper extremity, trunk, and lower extremity muscle groups compared to traditional methods [16–19]. However [5], compared the traditional bench press method with the bench press using weights suspended from elastic bands and reported no difference in muscle activation. In contrast [19], found greater activation in stabilizer muscles during unstable loads in the bench press exercise compared to the traditional method. The differences in the results may be attributed to the various measurement techniques used in the studies (e.g., instructions that affect movement speed or the loading of only a specific part with elastic bands). Additionally, another factor that could influence the results is the use of absolute loads rather than relative loads for all participants [20]. There are very few studies in the literature examining ground reaction forces with unstable loads. When these studies are reviewed, it is seen that the ULT method shows minimal loss or even higher force output compared to traditional stable load methods [16, 18, 21]. In their research [18], found that the unstable squat exercise created with elastic bands resulted in only a 3% loss in force output compared to squats performed with a stable load. Additionally [21], examined the squat exercise using four different loading methods. These methods were: stable surface with stable load, stable load on unstable surface, unstable load on stable surface, and unstable load on unstable surface. Upon analysing the results, they reported minimal force loss in unstable loads compared to the traditional stable load on stable surface method. Conversely [16], indicated in their study that flexible barbell method generated higher ground reaction forces compared to the traditional stable load method.

This study included three main objectives: (a) determine whether ULT in the bench press can cause greater muscle activation in the stabilizing musculature while maintaining the same or greater amount of muscle activation in the primary movers; (b) determine what is the most effective form of ULT based on the most common forms used in current research; and (c) determine which ULT method will yield more effective results in kinematic parameters (displacement, velocity, movement duration and time).

Three different hypotheses have been proposed in this study: (a) ULT will cause greater muscle activation in the stabilizing musculature, but only ULT with the Bandbell bar will create greater muscle activation in the primary movers when working with a lower load; (b) ULT using a flexible barbell with an unstable load will be superior set up to the standard barbell with unstable load; (c) kinematic parameters will be highest in flexible bar with ULT, second highest in standard bar with unstable load and least high in standard bar with stable load.

Methods

Study design

In this research design, we used a within-subject comparison across four different conditions. The design included two different stable load interventions and two different unstable load interventions: (a) Standard bar with a stable load (75% of 1RM), (b) Standard bar with a stable load (60% of 1RM), (c) Standard bar with an unstable load (60% of 1RM), (d) Flexible barbell (60% of 1RM). The independent variables of the study were the four different load types: stable load 75% (standard bar), stable load 60% (standard bar), unstable load (BandBell), and unstable load (flexible barbell). The dependent variables were muscle electromyography (EMG) activity and kinematic parameters. The setup of the four different unstable load methods is shown in Fig. 1 (a, b,c), respectively.

Fig. 1.

a-Standard bar with stable load %75-%60; b-Standard bar with an unstable load; c-Flexible barbell

Participants

Sixteen male individuals with at least five years of experience in resistance training voluntarily participated in this study (age: 25 ± 3.14 years, body mass: 85.05 ± 5.39 kg, height: 1.78 ± 0.06 m, lifting experience: 6.75 ± 1.56 years, and bench press 1RM: 127.31 ± 13.45 kg). It was confirmed that none of the participants had experienced any upper extremity injuries in the past six months or undergone any upper extremity surgery that could affect their ability to exercise. Participants who did not meet these criteria were excluded from the study. Six participants reported having prior experience with unstable load training (ULT) before participating in this study.

Ethics statement

Before data collection for the study, ethical approval was obtained from the Ethics Committee of Sakarya University of Applied Sciences with the approval dated 11.10.2024 and numbered E-26428519-050.99-145548. The study was conducted in accordance with the ethical guidelines of the University College and the treatment standards for human participants outlined in the 5th Helsinki Declaration. All participants were informed about the procedures both in writing and verbally, and they provided written informed consent to participate in the study.

Procedures

All participants were required to complete two testing sessions. To minimize the effects of fatigue, they were instructed to avoid exercise for 48 h prior to testing. A familiarization session was conducted to introduce participants to the testing protocol and provide them with the opportunity to practice the unstable load methods used in the study (elastic band and flexible barbell). Before the familiarization session, a 10-minute dynamic warm-up was performed. The familiarization protocol mirrored the measurement protocol by having participants complete one set of five repetitions with each method used in the study. The described method has been previously used and replicated in research to ensure accurate data comparison [18, 19].

During the first testing session, participants performed a 1RM bench press using a standard barbell and typical load, following the National Strength and Conditioning Association’s (NSCA) maximum strength testing guidelines [22]. Since different percentages of 1RM were used in each intervention, a single 1-Repetition Maximum (1RM) test was conducted to determine the loads for all four interventions. As relative intensities were used, measuring each participant’s 1RM value was necessary.

The second testing session was conducted seven days after the 1RM testing session. During the application of the four different methods in the study, the order of the interventions was randomized individually for each participant. Moreover, to ensure a balanced representation across the entire study sample consisting of sixteen participants, a counterbalanced order was employed. The methods used were: Standard bar with a stable load (75% of 1RM), Standard bar with a stable load (60% of 1RM), Standard bar with an unstable load (60% of 1RM), Flexible barbell (60% of 1RM). The selected percentages were consistent with previous studies examining unstable load training (ULT) [5, 18, 19, 23]. Each participant performed the following warm-up sets: one set with an empty bar, one set at 25% of 1RM, and one set at 50% of 1RM. A two-minute rest was provided between warm-up sets, and no specific tempo was enforced for repetitions due to issues identified in previous studies [5, 19].

Each participant was required to perform one set of five repetitions for each intervention, and this set was recorded. Due to time constraints, one set was chosen instead of two, and the previously conducted familiarization session provided sufficient preparation for participants to successfully complete each intervention [18, 19, 23]. To maintain consistency with previous studies, each participant rested for five minutes between working sets [19]. Participants were given two specific instructions prior to performing the exercises. First, they were instructed to initiate the upward phase of the movement as soon as the bar made contact with the chest, without pausing at the bottom position and while maintaining control of the bar throughout the movement. Second, they were asked to perform each repetition at the highest speed possible, provided that proper technique and control could be maintained (Table 1).

Table 1.

Structured study procedure

Phase	Contents
Session 1: Familiarization	• 10-minute dynamic warm-up • Introduction to unstable methods (elastic bands, flexible barbell) • 1 set × 5 reps for each method practiced
Session 1: 1RM Testing	• Standard barbell bench press • 1RM test conducted per NSCA guidelines [22] • 1RM used to calculate test loads
Inter-Session Period	• 7-day full recovery rest interval
Session 2: Warm-Up	• 1 set with empty bar • 1 set at 25% 1RM • 1 set at 50% 1RM • 2-minute rest between sets
Session 2: Main Testing	• 4 randomized and counterbalanced conditions:  Stable Load 75%  Stable Load 60%  Unstable Load 60%  Flexible Barbell 60% • 1 set × 5 reps for each condition • 5-minute rest between conditions • Video and EMG recordings

In the unstable load condition, weight plates were attached to the bar by suspending them with elastic resistance bands (Corength, Decathlon, France). The tape is made of 50% synthetic rubber and 50% natural rubber latex material, with a length of 1.04 m, a thickness of 0.005 m, and a width of 0.0285 m. The resistance bands were looped through the weight plates (quadruple loop) and hung from the bar. Each participant was allowed to choose the most comfortable bench press position, including hand placement. To ensure consistency across all trials, hand placement was marked. As a precautionary measure, the load on each resistance band was limited to a maximum of 50 lbs. When additional weight was needed to reach the designated load level, extra bands were used to achieve the correct load. The flexible barbell is a relatively new training device currently used in various collegiate-level strength and conditioning programs [16]. Unlike traditional Olympic barbells, it is constructed from specialized composite materials and features a flexible and dynamic structure. The bar is approximately 90 inches (228.6 cm) in length, with a diameter ranging from 2 to 3 inches (5.08–7.62 cm). It weighs around 6.8 kg and is designed to accommodate standard Olympic weight plates on both ends. The defining characteristic of the flexible barbell is its ability to produce oscillatory movement during both the eccentric and concentric phases of an exercise. This oscillation introduces an element of instability that promotes increased activation of stabilizing muscle groups and demands greater neuromuscular control throughout the movement. In the stable load condition with a standard bar, a conventional bench press setup was used, in which Olympic weights were loaded onto the bar and secured with safety collars. No verbal encouragement was given to participants.

Measurements

Before placing the surface electrodes (Bagnoli™ Surface EMG Sensor), the participants’ skin was cleaned using cotton and skin cleaning alcohol [24]. The electrodes were attached and secured to the skin with surgical tape (Roll Fix 10 cm). Fifteen of the 16 participants reported that their right arm was their dominant arm, so the electrodes were placed on the right side of the body. The electrodes were positioned on the pectoralis major, triceps brachii, anterior deltoid, lateral deltoid, rectus abdominis, and external oblique muscles according to previous studies [8, 25, 26] and the SENIAM recommendations [24]. The surface EMG data obtained from these muscle groups were recorded at a sampling frequency of 1500 Hz using EMGworks Acquisition (Delsys Bagnoli™). The recorded data were analyzed using Matlab analysis software. To eliminate noise from the data, filtering was applied according to the SENIAM recommendations. Fourth order band-pass filter was applied to the EMG data in the 10–250 Hz range. The filtered signal (mV) mean kinematic data was used to determine the concentric and eccentric phases of the five repetitions. To avoid potential errors when starting the exercises, the first and last repetitions were excluded [8]. The mean values of the filtered EMG data for each muscle were calculated. The EMG data has been normalized using the reference muscle method (Eq. 1). The stable load at 75% was considered the reference at 100%, and the stable load at 60%, the standard bar with an unstable load, and the flexible barbell raw EMG data were normalized relative to the stable load at 75% [6].

1

For obtaining kinematic data, markers were placed on three points of the bar (left end, center, and right end) and on the weight plates. Cameras were positioned with a 90° angle to the optical axis, perpendicular to the motion plane, ensuring they fit within the motion frame. Video recordings of each movement were captured with a camera (1080 × 1920 resolution at 120 fps). For the simultaneous collection of video and EMG data, a 5 V signal sent to one channel of the analog/digital converter card, controlled by a button-operated light (S-link SL-L10, LED light), was used. Records for each participant were digitized using the Tracker video analysis and modeling tool [27]. Interpolation (interp1 with option in Matlab) was used to synchronize the EMG and kinematic data.

Statistical analyses

All statistical analyses were conducted using SPSS software (25.0, SPSS Inc., Chicago, IL, USA). First, the Shapiro-Wilk test was applied to determine the normality distribution of the variables. Homogeneity of variance was assessed using Levene’s test. A One-Way Analysis of Variance (ANOVA) was used to compare the four different bench press variations. If statistically significant differences were detected in the ANOVA results, Tukey HSD and Bonferroni-corrected post-hoc tests were applied to determine the differences between groups. For non-parametric data, the test was used, and if a significant difference was found, Dunn’s test was performed as a post-hoc analysis. The level of statistical significance was set at p < 0.05 for all analyses.

Results

Fig. 5.

Mean EMG% values ​​of the anterior deltoid, pectoralis major, triceps brachii, rectus abdominis, external oblique, lateral deltoid muscles during work sets of 4 different bench press variation

Displacement

In the eccentric phase, displacement values of four different bench press variations, flexible barbell (39.84 ± 4.5) exhibited higher displacement compared to the standard unstable load (34.51 ± 3.53) and stable load 75% (34.65 ± 5.35) (p < 0.005). No significant difference was observed between stable load 60% (37.14 ± 4.63) and other variations (p > 0.005). In the concentric phase, flexible barbell (40.17 ± 4.62) also showed higher displacement than the standard unstable load (34.55 ± 3.47) and stable Load 75% (34.88 ± 5.3) (p < 0.003). No significant difference was found between stable load 60% (37.22 ± 4.62) and other variations (p > 0.005) (Fig. 1).

Fig. 2.

Displacement graph of four different bench press variation

Velocity

In the eccentric phase, flexible barbell (0.59 ± 0.23) demonstrated higher velocity compared to the standard unstable load (0.29 ± 0.12) and stable load 75% (0.49 ± 0.10) (p < 0.002). No significant difference was found between stable load 60% (0.60 ± 0.11) and other variations (p > 0.005). In the concentric phase, flexible bar (0.54 ± 0.22) and stable load 60% (0.56 ± 0.09) showed higher velocity than stable load 75% (0.40 ± 0.07) (p < 0.001). No significant difference was observed between the standard unstable load (0.36 ± 0.09) and other variations (p > 0.005) (Fig. 2).

Fig. 3.

Velocity graph of four different bench press variations

Movement duration and time

Regarding total movement duration, the standard unstable load (11.23 ± 3.38) and stable load 75% (10.05 ± 2.14) exhibited longer execution times compared to stable load 60% (7.57 ± 1.86) and flexible barbell (7.39 ± 1.29) (p < 0.000) (Fig. 3a). In the eccentric phase duration per repetition, the standard unstable load (1.14 ± 0.61) had a longer duration compared to stable load 60% (0.68 ± 0.24) and flexible barbell (0.64 ± 0.15) (p < 0.001). No significant difference was found between stable load 75% (0.87 ± 0.28) and other variations (p > 0.005). In the concentric phase, the unstable load (1.07 ± 0.42) and stable load 75% (0.95 ± 0.21) had longer durations compared to stable load 60% (0.70 ± 0.16) (p < 0.000). The unstable load (1.07 ± 0.42) also showed longer duration than flexible barbell (0.72 ± 0.13) (p < 0.000) (Fig. 3b).

Fig. 4.

Movement and phase durations of four different bench press variations

Muscle activation

Anterior deltoid

In the eccentric phase, the standard unstable load (97.02%) showed higher muscle activation compared to the stable load 60% (76.84%). The standard unstable load (97.02%) also exhibited higher activation than the reference stable load 75% (100%) and stable load 60% (76.84%) (p < 0.001). In the concentric phase, reference stable load 75% (100%) had higher activation than stable load 60% (74.24%) and flexible barbell (80.73%) (p < 0.001).

Pectoralis major

Reference stable load 75% (100%) exhibited higher activation than the unstable load (82.84%), stable load 60% (76.13%), and flexible barbell (85.55%) (p < 0.000). In the concentric phase, reference stable load 75% (100%) showed higher activation than the unstable load (86.20%), stable load 60% (83.45%), and flexible barbell (87.44%) (p < 0.000) (Fig. 4b).

Triceps brachii

In the eccentric phase, reference stable load 75% (100%) had higher activation than the unstable load (81.70%) and stable load 60% (77.21%) (p < 0.004). In the concentric phase, reference stable load 75% (100%) showed higher activation than stable load 60% (p < 0.003). No significant difference was found between the unstable load (88.45%), flexible bar (87.10%), and other variations (p > 0.005) (Fig. 4c).

Rectus abdominus

In the eccentric phase, reference stable load 75% (100%) had higher activation than the unstable load (78.10%) and stable load 60% (77.18%) (p < 0.002). In the concentric phase, reference stable load 75% (100%) exhibited higher activation than the unstable load (83.19%) and stable load 60% (82.44%) (p < 0.002).

External oblique

In the eccentric phase, reference stable load 75% (100%) exhibited higher activation than stable load 60% (75.82%) (p < 0.002). In the concentric phase, the unstable load (96.91%) and reference stable load 75% (100%) showed higher activation than stable load 60% (71.44%) (p < 0.000).

Lateral deltoid

In the eccentric phase, reference stable load 75% (100%), unstable load (93.46%), and flexible barbell (89.07%) exhibited higher activation than stable load 60% (70.26%) (p < 0.000). In the concentric phase, reference stable load 75% (100%) and flexible barbell (95.33%) showed higher activation than the unstable load (78.76%) and stable load 60% (77.14%) (p < 0.001).

Dıscussıon

This study aimed to determine the differences in muscle activation and kinematic parameters that occur when using an unstable load during the bench press exercise compared to using a stable load. The present findings indicate that the flexible barbell method led to significant differences in kinematic parameters when compared to the stable 60% load, stable 75% load, and standard unstable load conditions. However, no significant difference was found between flexible barbell, stable load 60%, and unstable load methods in the activation levels of the AD, EO, and LD muscle groups, but it was observed that these three methods produced higher muscle activity than the stable load 60% method. It was determined that the stable load 75% method produced higher muscle activity than the other methods in the activation levels of the PM, TB, and RA muscle groups.

According to this study’s kinematic data, flexible barbell method was faster than the other methods in both phases (concentric and eccentric) of the bench press. Also, the displacement of flexible barbell was higher than those of the other methods. Durations for performing a single repetition or all 5 repetitions of the bench press were shorter than all other methods except the stable 60% method. These kinematic findings differ from [19], who reported that the pressing time with an unstable load was longer than the stable load both in the concentric and eccentric phases in the bench press. The unstable load method suspended by elastic bands also took longer than the other methods in our research. Although the Tsunami bar has a structure similar to flexible barbell used in previous studies, weight plates can be attached to it as with standard bars. Therefore, there is no need to suspend the weights with elastic bands to create an unstable load. This can reduce the difficulty in controlling the movement caused by the swing of the suspended weights and shorten the bar path [23] reported that the unstable method applied with suspended weights increased the bar path, and unstable conditions created by mini bands were less predictable than stable conditions in the mediolateral and anteroposterior directions. The ability to perform exercises faster in unstable conditions with the Tsunami Bar can help with the correct timing of counter-contractions resulting from oscillations. This can help the athlete minimize the explosive power deficit and achieve optimum power. For example [28], compared two different power-training regimens over five weeks: a combined weight training program (CT) using speed lifts and plyometrics and flexible barbell (FB) training. Their results suggest that both FB and CT training improved power over five weeks, but that FB training may be more effective than CT in developing lower-body power. When examining the EMG data in our study, anterior deltoid (AD) muscle activation did not differ between loading methods; only the stable 60% load method showed lower activation compared to the other loading methods. This finding is consistent with the study by Dunnick et al. [5]. Researchers reported that the AD muscle did not show a significant difference in stable and unstable loads, but muscle activation was greater at higher loads (80% 1RM). This suggests that unstable loads do not create a significant difference in the activation of the AD muscle. However, it should not be forgotten that Dunnick and his colleagues applied a 2-second cadence in eccentric and concentric actions, which is different in the method section. In the pectoralis major (PM) and triceps brachii (TB) muscles, it was found that the stable 75% load method provided higher activation. These findings are consistent with the findings reported by Costello (2022) and Lawrence (2017). For example, in Costello’s study, PM and TB muscle activation were found to be higher in the stable 75% load method. This suggests that higher stable loads are a priority for the activation of the primary movement muscles.

In the stabilizer muscles, rectus abdominis (RA), external oblique (EO), and lateral deltoid (LD), it was found that the stable 75% load method provided the highest activation for RA and EO, while the stable 60% load method showed the lowest activation for LD. This finding may contradict the findings of [17], who reported that unstable loads significantly increased the activation of these muscles. However, considering that Costello emphasized that unstable loading methods (such as the bandbell bar) require more control and stabilization, it is thought that this difference may be due to different loading protocols. The findings of [23] that the biceps brachii showed higher activation under unstable loads may also be consistent with these results. The biceps brachii plays a critical role in maintaining the balance of unstable loads and therefore may be more activated during unstable loads.

Although muscle activation did not show a statistical increase with the Tsunami Bar, it did not show a decrease when the loads were equal. However, there is also research reporting that the Tsunami bar affects muscle activation more [16] found that core muscle and prime mover muscles used during the back squat were significantly more activated when using the FB than those of an equally weighted steel bar when both were lifted at the same speed. The positive difference in kinematic data compared to other methods makes the Tsunami Bar stand out compared to other methods in the unstable load method. It is suggested that its use in speed-based strength training exercises may offer combined benefits of movement speed and stabilization. In addition, the fact that the weight plates can be directly attached like a standard bar instead of hanging them with a rubber band in unstable load exercises is an important usage advantage of the Tsunami Bar. Also, the lack of experience should not be ignored for this equipment, which is newly included in strength training programs. Future studies with more experienced subjects with this equipment may provide more effective results.

Conclusion

The results of this study demonstrate that unstable load training (ULT) with flexible barbell led to higher displacement and velocity in both eccentric and concentric phases compared to other variations, such as standard unstable load and stable load 75%. However, no significant differences were observed between flexible barbell and stable load 60% for most kinematic parameters. In terms of muscle activation, stable load 75% consistently resulted in the highest activation across various muscle groups, including the Pectoralis Major, Triceps Brachii, Rectus Abdominus, External Oblique, and Lateral Deltoid. The standard unstable load showed higher activation in the Anterior Deltoid compared to stable load 60%, but no significant differences were found between flexible barbell and other variations in most muscles. These findings suggest that while ULT may improve displacement and velocity, the effect on muscle activation depends on the load and phase of the movement, with stable load 75% generally yielding the highest activation. Therefore, the choice of load and exercise variation should be considered carefully based on the desired training outcome, whether it is enhancing movement dynamics or optimizing muscle activation.

Limitations

Despite this study’s valuable findings, several limitations should be acknowledged.

• The sample group of this study is limited to 16 athletes, consisting only of male participants with resistance training experience. This limits the generalizability of the results to broader populations, including female athletes or individuals with different training backgrounds.

• While the study utilized surface electromyography (sEMG) to measure muscle activation, this method has inherent limitations. sEMG data can be influenced by electrode placement, skin impedance, and crosstalk from adjacent muscles.

• The study focused solely on acute effects rather than long-term adaptations. While the findings highlight immediate differences in muscle activation and kinematic parameters across different bench press variations, they do not indicate how these variations impact strength development, hypertrophy, or neuromuscular adaptations over time.

Practical applications

The use of both stable and unstable loads together can provide different advantages in training programs. For example, strength training with stable loads provides strength to the main muscle groups, while training with unstable loads can improve balance and stabilization skills. Tsunami Bar can be an alternative option for unstable load methods. Training with Tsunami Bar offers significant advantages in speed and balance-oriented exercises while allowing the bar path to be shorter and the movement to be performed faster compared to stable loads. Tsunami Bar offers a practical alternative to provide instability without the need to use hanging loads with traditional elastic bands. This can improve both control of movement and bar path, making the athlete’s power development process more efficient.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

1RM

One Repetition Maximum

EMG

Electromyography

AD

Anterior Deltoid

EO

External Oblique

LD

Lateral Deltoid

PM

Pectoralis Major

RA

Rectus Abdominis

TB

Triceps Brachii

FB

Flexible Barbell

ULT

Unstable Load Training

UST

Unstable Surface Training

NSCA

National Strength and Conditioning Association

Author contributions

All authors have contributed sufficiently to the manuscript and have approved the final version. B.D. conceptualization, methodology, supervision, writing—original draft, writing—review, and editing. O.C. conceptualization, data curation, formal analysis, writing—original draft, writing—review, and editing. O.I. data curation, formal analysis, writing—original draft, writing—review, and editing. O.C. resources, writing—original draft, writing—review, and editing. M.C. resources, writing—original draft, writing—review, and editing. M.B. supervision, project administration, and critical revision of the manuscript.

Funding

This study did not receive any external funding.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Sakarya Applied Sciences University Faculty of Sports Sciences non-invasive research ethics committee (No:49 Date: 11.10.2024). All study participants signed informed consent forms to participate in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.