Concentric-Only Versus Touch-and-Go Bench Press One-Repetition Maximum in Men and Women

Abstract

Background:

One-repetition maximum (1RM) tests are time-consuming, and they might not always be logistically possible or warranted due to increased risk of injury when performed incorrectly or by novice athletes. Repetitions-to-failure tests are a widespread method of predicting the 1RM, but its accuracy may be compromised by several factors such as the type of exercise, sex, training history, and the number of repetitions completed in the test.

Hypothesis:

The touch-and-go bench press would provide a higher 1RM than the concentric-only bench press for both genders regardless of whether the 1RM was obtained by the direct or repetitions-to-failure method and the error in the 1RM prediction would be positively correlated with the number of repetitions performed to failure and negatively correlated with the 1RM strength and resistance training experience.

Study Design:

Cross-sectional study.

Level of Evidence:

Level 3.

Methods:

A total of 113 adults (87 men and 26 women) were tested on 2 sessions during the concentric-only and touch-and-go bench press. Each session consisted of an incremental loading test until reaching the 1RM load, followed by a repetitions-to-failure test.

Results:

The 1RM was higher for the touch-and-go bench press using both the direct (men, 7.80%; women, 7.62%) and repetitions-to-failure method (men, 8.29%; women, 7.49%). A significant, although small, correlation was observed between the error in the estimation of the 1RM and the number of repetitions performed (r = 0.222; P < 0.01), 1RM strength (r = −0.169; P = 0.01), and resistance training experience (r = −0.136; P = 0.05).

Conclusion:

The repetitions-to-failure test is a valid method of predicting the 1RM during the concentric-only and touch-and-go bench press variants. However, the accuracy of the prediction could be compromised with weaker and less experienced individuals and if more than 10 repetitions are completed during the repetitions-to-failure test.

Clinical Relevance:

The repetitions-to-failure test does not require any sophisticated equipment and enables a widespread use in different training environments.

Resistance training (RT) is commonly prescribed taking into account the results of a one-repetition maximum (1RM) test. ²⁸ The prescription of the loads as a percentage of the 1RM (%1RM) enables practitioners to target specific neuromuscular adaptations and optimize training adaptation and recovery. ³ Although 1RM tests are reliable and easy to implement,^1,8 they are time-consuming and they might not always be logistically possible or warranted due to increased risk of injury when performed incorrectly or by novice athletes.^9,14 To overcome the limitations of a direct 1RM testing, alternative prediction methods have been proposed to estimate the 1RM from the maximum number of repetitions performed to failure with a given load (ie, repetitions-to-failure [RTF] tests) or through the assessment of the load-velocity relationship.^7,18,26

The load-velocity relationship is now commonly established during the warm-up sets of RT sessions.^31,32 Practitioners need to record the mean concentric velocity against 2 to 5 loads and then the 1RM can be estimated through a linear regression as the load linked to the velocity of the 1RM (V1RM). ⁹ Although some studies have shown that the same V1RM can be used for all individuals to accurately predict the 1RM during upper-body exercises,^7,8,26 other researchers have used an individualized V1RM to estimate the 1RM.^1,13 Therefore, it is important to note that this method of predicting the 1RM could require a direct 1RM assessment to obtain the individual’s V1RM. In addition, the acquisition of devices that are able to measure movement velocity with a high reliability is important to obtain accurate 1RM estimations, ²⁵ and these devices may be cost prohibitive for nonprofessional athletes, especially in team sports where multiple devices would be necessary, thus limiting its use by coaches and athletes. Therefore, there is a need of alternative and more heuristic methods of predicting the 1RM that would allow a widespread use in different training environments.

One method is the RTF test, which consists of previously proposed regression equations that estimate the 1RM from the maximal number of repetitions performed with a submaximal load.^15,18,29 This method does not require sophisticated equipment (such as linear position transducers or inertial sensors) to estimate the 1RM, making it a convenient choice in practical settings. Although useful, the amount of repetitions performed in RTF tests, training history, and sex of an individual can all affect the precision of an estimated 1RM, but the research on these effects is scarce.^19,29 Moreover, other studies indicated that the accuracy of RTF equations could be exercise-dependent. ³⁴ For example, LeSuer et al ¹⁵ highlighted the Mayhew equation to be the most accurate for predicting the bench press 1RM, while the Wathan equation was the most appropriate for predicting the back squat 1RM.

Although the load-velocity relationship has been found to differ between variants of the same exercise,^10,23 no study has explored whether the prediction accuracy of RTF equations could differ between variants of the same exercise (eg, concentric-only bench press vs touch-and-go bench press). Exercises using the stretch shortening cycle generally enable the production of greater values of concentric force, velocity, and power than concentric-only exercises.^2,22 The 1RM has also been found to be greater for the touch-and-go bench press in comparison with the concentric-only variant.^10,33 However, both variants of the bench press exercise are commonly used in training because each of them allow a greater force production at different points of the lift (touch-and-go bench press—higher force at the beginning of the lift; concentric-only bench press—higher force at the end of the lift). ²² Therefore, it would be important to examine whether the validity of RFT tests to predict the bench press 1RM is affected by the execution mode.

The aims of the present study were (1) to compare the 1RM between the concentric-only and touch-and-go bench press variants assessed by the direct method and the RTF method using the Mayhew equation in men and women and (2) to determine the effect of the maximum number of repetitions performed to failure, maximum strength values, and RT experience on the accuracy of the 1RM prediction. We hypothesized that (1) the touch-and-go bench press would provide a higher 1RM than the concentric-only bench press for both men and women regardless of whether the 1RM was obtained by the direct or RTF method and (2) the error in the 1RM prediction would be positively correlated with the number of repetitions performed to failure and negatively correlated with the 1RM strength and RT experience.

Methods

Participants

A total of 113 healthy participants (87 men [age = 20.8 ± 4.2 years; body height = 1.70 ± 0.20 m; body mass = 74.3 ± 15.4 kg; concentric-only bench press 1RM = 61.3 ± 17.6 kg; touch-and-go bench press 1RM = 66.1 ± 18.4 kg] and 26 women [age = 19.5 ± 1.9 years; body height = 1.60 ± 0.06 m; body mass = 63.8 ± 10.9 kg; concentric-only bench press 1RM = 30.0 ± 5.9 kg; touch-and-go bench press 1RM = 32.3 ± 6.7 kg]) participated in this study. We recruited a heterogeneous sample with regard to the maximal strength capacity and RT experience (1.3 ± 2.4 years [range = 0-10 years]) to elucidate the effect of both variables on the accuracy of the bench press 1RM prediction from a RTF test. Prior to testing and after detailed explanation of the procedures, participants gave their written consent to participate in the study. Participants were instructed to avoid any strenuous exercise for the duration of the study. The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the institutional review board of the University of Granada.

Study Design

A randomized crossover design was used to compare the 1RM performance between the concentric-only and touch-and-go bench press variants and to identify factors that influence the accuracy of the Mayhew 1RM prediction equation. For these purposes, 113 young adults (87 men and 26 women) were tested at the faculty research laboratory on 2 occasions separated by 72 to 96 hours. A single bench press variant (concentric-only bench press or touch-and-go bench press) was evaluated on each session in a randomized order. Each session consisted of an incremental loading test until reaching the 1RM, followed by an RTF test. The loads of the RTF test were randomly assigned and ranged from the 75% to 90% of the previously determined 1RM. Different relative loads were used during the RTF test to explore the effect of the number of completed repetitions on the accuracy of the 1RM prediction. Therefore, 4 1RM values were obtained in the present study for each participant (2 bench press variants [concentric-only bench press and touch-and-go bench press] × 2 methods [direct and RTF]). The 2 sessions for the same participants were held at the same time of the day (±1 hour) to minimize the influence of the circadian rhythm on physical performance.

Testing Procedures

Both testing sessions started with the same standardized warm-up: 5 minutes of running at self-selected pace, dynamic stretching, arm and shoulder mobilization, and 1 set of 10 repetitions (for men) or 5 repetitions (for women) with an external load of 20 kg (mass of the unloaded Smith machine barbell) in the tested bench press variant. Thereafter, an incremental loading test adapted from Pestaña-Melero et al ²⁷ was performed to determine the bench press 1RM. A linear velocity transducer (T-Force System; Ergotech) was attached to the barbell of the Smith machine (Ffittech) and the increment of the load during the testing protocol was based on the recorded mean concentric velocity (MCV).

The initial external load was 20 kg for all participants. The load was progressively increased in 10 kg for men and in 5 kg for women until the MCV of the barbell was <0.50 m·s⁻¹. From that moment, the load was increased in increments of 1 to 5 kg for men and 1 to 2.5 kg for women until the 1RM was achieved. Two repetitions and 3 minutes of interset rest were used when the MCV was > 0.50 m·s⁻¹, and 1 repetition and 5 minutes of interset rest were used when the MCV was ≤0.50 m·s⁻¹. Spotters were present during the test for safety reasons and to encourage participants to lift the maximum possible load. The recording of MCV was not used in the present study to predict the 1RM, but it was useful to decide the magnitude of the increment in the load and the duration of the recovery periods.

Ten minutes after the direct assessment of the 1RM, participants performed a set of repetitions to failure (ie, RTF test) with a load ranging between the 75% 1RM and 90% 1RM. Participants performed the concentric phase at the maximum intended velocity and the eccentric phase under control. ³⁴ Spotters were also present for safety reasons and to encourage participants to perform the maximum possible number of repetitions. The Mayhew et al ¹⁷ (1RM = [submaximal load / 52.2 + 41.9·e^{−0.055 . repetitions}] / 100) equation was used to predict the 1RM from the load (kg) and the number of repetitions completed. The Mayhew et al ¹⁸ equation was used over other lifts-to-failure equations because a recent study has shown that it presents the highest accuracy for estimating the 1RM during the bench press exercise performed in a Smith machine. ²⁴

Regardless of the bench press variant, the 5-point body contact position technique (head, upper back, and buttocks firmly on the bench with both feet firmly on the floor) was used. All participants were allowed to self-select the grip width, but it was kept constant for both testing sessions. ²⁴ Participants always initiated the task holding the barbell with their arms fully extended. During the concentric-only bench press, participants lowered the barbell at a self-selected velocity until the barbell made contact with their chest, waited with the barbell on the chest for 2 seconds, and on the word “Go!” performed the concentric phase until their arms were fully extended. During the touch-and-go bench press, participants were instructed to lower the barbell until it touched the chest and then immediately perform the concentric phase.

Statistical Analysis

Descriptive data are presented as means and standard deviation. The normal distribution of the data was confirmed by the Shapiro-Wilk test (P > 0.05). A mixed analysis of variance (ANOVA) with Bonferroni post hoc corrections was applied to the 1RM values considering the “exercise” (concentric-only bench press and touch-and-go bench press) and “method” (direct and RTF) as within-participant factor, and “sex” (men and women) as between-participant factor. The magnitude of the differences was reported by the partial eta square (h2p) for the ANOVAs and Cohen d for pairwise comparisons. Bland-Altman plots were also used to quantify the systematic bias and 95% limits of agreement between the actual and predicted 1RM. Pearson correlation coefficient (r) was used to quantify the association between the error in the 1RM prediction and the maximum number of repetitions performed, 1RM strength and RT experience. The criteria for interpreting the magnitude of the r coefficients were as follows: trivial (0.00-0.09), small (0.10-0.29), moderate (0.30-0.49), large (0.50-0.69), very large (0.70-0.89), nearly perfect (0.90-0.99), and perfect (1.00). ¹² Statistical significance was set at an alpha level of 0.05. All statistical analyses were performed using the software package SPSS (Version 22.0; IBM Corp) and the figures were created in Microsoft Excel for Mac (Version 15.13.3).

Results

The ANOVA conducted on the 1RM values revealed a significant main effect of exercise (F = 89.5; P < 0.01; h2p = 0.447; touch-and-go bench press > concentric-only bench press), method (F = 7.4; P < 0.01; h2p = 0.062; RTF method > direct method), and sex (F = 81.9; P < 0.01; h2p = 0.425; men > women). The interaction “exercise × sex” was significant (F = 12.1; P < 0.01; h2p = 0.099) since the increment in the 1RM performance for the touch-and-go bench press compared with the concentric-only bench press was more accentuated in men (≈8.0%) than in women (≈7.6%). The remaining interactions did not reach statistical significance (F ≤ 1.6; P ≥ 0.21; h2p ≤ 0.014). The correlations for the 1RM between both bench press variants were very high (r ≥ 0.94) (Figure 1).

Figure 1.

Comparison of the one-repetition maximum (1RM) between the concentric-only and touch-and-go bench press variants in (a, b) men and (c, d) women obtained by the (a, c) direct method and the (b, d) indirect method based on the number of repetitions performed to failure. Each point represents the data of an individual patient. ES, effect size; r, Pearson correlation coefficient.

Very high correlations were observed between the 1RM obtained by the direct and RTF methods for both the concentric-only and touch-and-go bench press (r > 0.99) (Figure 2). Bland-Altman plots revealed low systematic bias between the direct and RTF method for both the concentric-only bench press (0.56 kg) and touch-and-go bench press (0.82 kg), while the random errors (2.08 kg vs 2.74 kg) and the heteroscedasticity of the errors (r² < 0.001 vs r² = 0.095) were higher for the touch-and-go bench press (Figure 3). A significant, although small, correlation was observed between the error in the estimation of the 1RM and the number of repetitions performed to failure (r = 0.222; P < 0.01), 1RM strength (r = −0.169; P = 0.01), and RT experience (r = −0.136; P = 0.05) (Figure 4). The absolute error in the 1RM prediction was only higher than 5% when more than 10 repetitions were performed.

Figure 2.

Relationship between the one-repetition maximum (1RM) obtained by the direct method and the 1RM obtained by the indirect method based on the number of repetitions performed to failure during the (a) concentric-only bench press and (b) touch-and-go bench press. The regression equations and the Pearson multivariate coefficient of determination (R²) are depicted.

Figure 3.

Bland-Altman plots showing the differences between the actual one-repetition maximum (1RM) and the 1RM predicted from the lift-to-failure method during the (a) concentric-only bench press and the (b) touch-and-go bench press. Each plot depicts the systematic bias and 95% limits of agreement (±1.96 SD; dashed lines), along with the regression line (solid line). The systematic bias ± random error together with the strength of the relationship (r²) are depicted in each plot.

Figure 4.

Relationship between the (a) error in the prediction of the 1RM and the number of repetitions performed to failure (numbers represent the average error and SE), (b) 1RM value, and (c) resistance training experience. 1RM, 1-repetition maximum. The regression equation and Pearson correlation coefficient (r) are depicted. *Significant correlation (P < 0.05).

Discussion

This study was designed to compare the 1RM between the concentric-only and touch-and-go bench press variants assessed by the direct and RTF methods in men and women. The effects of the maximum number of repetitions performed to failure, strength levels, and RT experience on the validity of the RTF method was also examined. The main findings revealed (1) a significantly greater 1RM for the touch-and-go bench press compared with the concentric-only bench press, (2) high validity of the RTF method to predict 1RM in both bench press variants as evidenced by the trivial systematic differences and very high correlations, and (3) significant but small relationships between the error in 1RM estimation and the number of repetitions performed to failure, strength levels, and RT experience.

Supporting our first hypothesis, 1RM was greater for the touch-and-go bench press than for the concentric-only bench press variant. This finding is in line with previous research that showed significantly higher 1RM values in the touch-and-go bench press variant in comparison with the concentric-only bench press.^6,10 These results could be supported by earlier work suggesting that exercises using the stretch shortening cycle produce greater amounts of force, velocity, and power at the early portion of the concentric phase than exercises performed with a pause technique or concentric-only exercises.^2,22 This could be especially important for the bench press exercise because the sticking region is observed at the early portion of the concentric phase. ⁶ Although a considerable amount of research has compared mechanical variables between the traditional and ballistic bench press variants,^4,5,16,21 less studies have been conducted to compare the 1RM between the concentric-only and touch-and-go bench press variants, especially when women are included in the study sample. Our results evidence that women also achieve higher 1RM using the touch-and-go bench press in comparison with the concentric-only bench press. However, the increment in 1RM performance for the touch-and-go bench press compared with the concentric-only bench press was slightly more accentuated in men than in women.

Previous studies already established the ability of the RTF method to predict the 1RM during various exercises.^15,18,29 However, to our knowledge, this is the first study comparing the validity of the RTF method to predict 1RM between 2 commonly used bench press variants. For both bench press variants, the differences between the 1RM obtained by the direct and RTF methods were negligible with almost perfect agreement between the methods (ie, low systematic bias and high correlations) and no heteroscedasticity of the errors. In addition, RTF method was equally valid for both men and women, suggesting that sex does not affect the accuracy of the 1RM prediction through RTF. These results are not in accordance with Pérez-Castilla et al ²⁶ who showed an underestimation of the actual 1RM by the Mayhew equation in 2 upper-body pulling exercises (range −6.65 to −2.14 kg). García-Ramos et al ⁷ also discouraged the use of Mayhew equation to predict 1RM of another pulling exercise through the RTF method. The discrepancies between the findings could likely be explained by the exercises being investigated. The Mayhew equation, also used in the present study to predict 1RM, was proposed to be highly valid only for the bench press and the squat exercises, ¹⁵ while predicting 1RM of various pulling exercises using the same equation could be misleading. Collectively, these results support the validity of Mayhew RTF equation to predict the 1RM during the concentric-only and touch-and-go bench press variants for both men and women.

To implement RTF method in practice, several confounding factors should be considered, such as strength levels, training experience, and the load being used.^19,29 For instance, RTF prediction equations have been shown to overestimate the 1RM in bench press when the repetitions range is large.^18,19,29 However, the vast majority of 1RM prediction studies from RTF did not include both sexes, did not include people with different strength levels and training experience, and did not use different loads providing a wide range of repetitions during RTF tests. In this regard, we recruited men and women with varying levels of strength and training backgrounds and prescribed different loads to assess the influence of the number of repetitions performed in the prediction of the 1RM. Supporting our second hypothesis, a positive relationship was observed between the errors in the 1RM prediction and the number of repetitions performed with errors surpassing the 5% threshold after performing more than 10 repetitions. This finding is in line with previous research and suggests that to ensure the accuracy of 1RM prediction through RTF equations, 10 or less repetitions should be performed.^18,19,29 Furthermore, negative relationships were observed between the error in the 1RM prediction and both the 1RM performance and the RT experience. This was expected since the effectiveness of the neural drive (ie, recruitment, rate of onset, firing frequency) as well as intermuscular coordination is thought to be far superior in stronger and more experienced individuals.^11,20,30 Consequently, this probably affected the stability of the 1RM assessed through both the direct and the RTF methods among weaker individuals. Considering all the above, it seems that the Mayhew equation can be used to predict 1RM in both the concentric-only and the touch-and-go bench press variants in both men and women if the load being used does not enable more than 10 repetitions to be performed. However, it is important to keep in mind that the accuracy of 1RM predictions could be compromised with weaker and less experienced individuals.

The present study is not without limitations. For instance, we were able to recruit a considerably lower number of women compared with men. In addition, these findings do not necessarily transfer to other exercises or even modalities of the same exercise (ie, free-weight versions of bench press variants).

In conclusion, the touch-and-go technique allowed a greater 1RM to be lifted in the bench press exercise compared with the concentric-only technique in both men and women. Furthermore, using the RTF method and Mayhew equation to predict the 1RM in both bench press variants is justified since almost identical results were obtained in comparison with the direct 1RM testing. However, the accuracy of 1RM predictions in both bench press variants could be slightly compromised with weaker and less experienced individuals and if the load selected for the RTF test allows for more than 10 repetitions to be completed. Therefore, RTF test can serve as a valid method of predicting the Smith machine bench press 1RM, which does not require any sophisticated equipment and enables a widespread use in different training environments.