Secondary Logo

Journal Logo

Original Research

Effects of Variable Resistance Using Chains on Bench Throw Performance in Trained Rugby Players

Godwin, Mark S.1; Fernandes, John F.T.2; Twist, Craig2

Author Information
Journal of Strength and Conditioning Research: April 2018 - Volume 32 - Issue 4 - p 950-954
doi: 10.1519/JSC.0000000000002421
  • Free



Power output, the product of force and velocity, is a key element of sporting performance (2,3,11,21). For example, Baker (3) determined that power output during bench throws is a potent discriminator of rugby league playing standard, whereas Cronin and Hansen (11) found that squat jump power was greater in faster players. Consequently, strength and conditioning practitioners often use various resistance training practices to enhance power output (25,30,35,39).

There is dispute as to whether athletes should train with heavy, light, or a combination of loads to induce adaptations in neuromuscular power (15). Although some authors suggest the use of high loads (>60% 1RM; (25,30,35,39)), these loads are performed at low velocities. The use of lighter loads might resolve this issue (≤60% 1 RM; (19,23,27,37)). However, during traditional resistance training exercises (e.g., squat, bench press), there are large periods of deceleration toward the end of the movement accounting for 23 and 52% of the total duration at 1RM and 80% 1RM, respectively (10). Practitioners can therefore use variable resistance to augment lifting performance during power training phases (4,20,22). Resistance exercises using variable resistance attempt alter the load during certain phases of the range of motion by attaching chains to a barbell (34). During lifting (i.e., muscle-shortening phase), the links are sequentially lifted from the floor providing a linear increase in resistance that reflects the ascending strength curve. Conversely, as the bar is lowered (i.e., muscle lengthening), the chain links are also lowered onto the floor decreasing the resistance (26). The mass of the chains added is typically substituted for a percentage of the total constant load lifted (5,7,32,34). Baker and Newton (4) proposed that the addition of chains alters the kinetic profile of the strength exercise by allowing an increased duration of acceleration throughout the range of motion and an ability to maintain higher forces toward the end of the movement. A recent meta-analysis by Soria-Gila et al. (31) reported greater strength gains in programs using chains compared with conventional training, with significant increases observed in both trained and untrained populations where programs lasted >7 weeks.

Despite the use of chains being purported as a popular resistance training method (14,33), few studies have examined their kinetic effect during dynamic resistance exercise. In studies examining the addition of 5% chain resistance to 75 and 80% 1RM during the clean and snatch, no difference in bar displacement, bar velocity, ground reaction forces, and rate of force development was reported compared with a comparable constant load (7,9). However, the relatively low contribution of the chains might have been insufficient to elicit changes in lifting performance throughout the movement range. Swinton et al. (34) observed equivocal results, in that peak force and impulse were improved but velocity, power, and rate of force development were impaired with chains amounting to 20 and 40% 1RM across 3 different loads. Conversely, during bench press exercise at 75% 1RM (15% chains and 60% constant load), peak and mean velocity were improved with the use of variable resistance (5). It would seem that the effect of chains is dependent on both the exercise and the contribution of the chains to the load lifted. Furthermore, to our knowledge, no study has assessed the kinetic profile of ballistic-type movements e.g., bench throws. A study that determines if bench throw performance is improved with the use of chains would help to clarify the aforementioned disparity. Accordingly, the aim of this study was to examine whether the addition of chain resistance improved of the bench press throw.


Experimental Approach to the Problem

The study incorporated a randomized crossover design, with participants visiting the laboratory on 3 occasions over a 15-day period. The first visit consisted of a familiarization to the testing procedures and one repetition maximum (1RM) assessment on bench press. During the familiarization trials, participants undertook several unweighted bench throws to check technique. Participants were then randomly assigned to either the variable resistance (with chains) or control condition (without chains). Participants returned to the laboratory a minimum of 7 says later for trial 1 with or without chains and a further 7 days for trial 2 (opposing condition). Participants were asked to record a food diary for the 48 hours before the testing sessions and asked to repeat this for the second testing session. In addition, the participants were asked not to engage in any activity that was considered unduly fatiguing with regard to upper body impulsive-type movements in the 48-hour period before each testing session.


8 well-trained males (± SD 19.4 ± 2.3 years [age range: 18–24 years], 88.75 ± 5.98 kg) from a national league rugby union club provided written informed consent to take part in the study. All participants had a minimum of 2 years experience resistance training, a bench press greater than their body mass, and were asymptomatic of illness or injury. The University of Chester Faculty of Applied Sciences Research Ethics Committee approved the study.


Before testing trials, participants completed a standardized warm-up comprising 5 minutes light cardiovascular activity, dynamic flexibility exercises of the chest and triceps, and submaximal bench press throws up to 30% 1RM. For the testing trials, participants performed 3 maximal bench press throws with either 45% 1RM (no chains [control]) or 30% 1RM plus 15% 1RM chains (chain condition) with 5 minutes between each repetition (28). Participants were instructed to lower the bar under control to the chest, then push maximally and release the bar at full elbow extension. During each repetition, the participant's head, shoulders, gluteals, and feet remained in contact with the bench and floor. A spotter was used to catch the bar and replace onto the Smith machine. Commercially available chains (Reliant Marine, Hoylake, United Kingdom) were 12- and 6-mm diameter steel-linked chain in 1-m lengths, with each meter of chain weighing 2.7 and 1 kg, respectively. Additional mass was supplied by D-rings if needed and attached to the bar using 6-mm steel snap hooks (Reliant Marine, Hoylake, United Kingdom). At the lowest phase of the lift, when the bar touched the chest, all the chain rested on the floor. The chains were doubled looped so that all the chains would leave the floor in the concentric phase at full elbow extension. The mean ± SD loads lifted are shown in Table 1.

Table 1.:
Mean ± SD for 1RM, constant load (45% 1RM), and variable load (30% 1RM constant load + 15% 1RM chain load).*

Measurements of peak and mean power, peak and mean velocity, and acceleration were assessed using a FitroDyne rotary encoder (Fitronic, Bratislava, Slovakia) during the concentric phase of the lift. Instantaneous velocity was also measured at 50-ms intervals during the concentric phase of the lift (50–400 ms). The FitroDyne measures the vertical velocity of a Kevlar cable. Briefly, this apparatus comprises an optical sensor and a light source. The velocity of the movement is determined by the displacement time data with sampling frequency determined by the rotational velocity of the sensor (13). The FitroDyne was attached directly under a Smith machine bar through the Kevlar cable (<2 N resistance). The data collection system of the FitroDyne was connected to the computer through USB allowing for capture and saving of each performance measurement. Fernandes et al. (12) has shown this apparatus to produce reliable measures of muscle power and barbell velocity for bench press exercise (coefficient of variation of 3.1 and 4.7% for 30 and 40% 1RM).

Statistical Analyses

All data were log transformed to reduce bias because of nonuniformity of error and analyzed using the ES statistic with 90% confidence intervals (CIs) and % change to determine the magnitude of effects. Magnitude-based inferential statistics were used to provide information on the size of the differences allowing for a more practical and meaningful explanation of the data (6). Thresholds for the magnitude of the observed change for each variable were determined as the within-participant SD in that variable x 0.2, 0.6, and 1.2 for a small, moderate, and large effect, respectively (8,18). Threshold probabilities for a meaningful effect based on the 90% confidence limits were <0.5% most unlikely, 0.5–5% very unlikely, 5–25% unlikely, 25–75% possibly, 75–95% likely, 95–99.5% very likely, and >99.5% most likely. Effects with confidence limits across a likely small positive or negative change were classified as unclear (19). All calculations were completed using predesigned spreadsheets (16,17). The reliability for each measure was tested using intraclass correlation coefficient and SEM (38), with variability between repetitions reported using the coefficient of variation.


Reliability for each outcome measure was favorable and is reported in Table 2.

Table 2.:
Reliability of measures for the study.*

There was a very likely trivial difference in peak power between the chain and no chain conditions (0%, ES 0 ± 0.1) while the difference in mean power was unclear (1.4%, ES 0.07 ± 0.28). Peak (2.8%, 0.22 ± 0.23) and mean bar velocity (4.7%, 0.38 ± 0.23) were possibly and likely greater in the chain condition compared with no chain condition. There were very likely greater bar acceleration (25.2%, ES 0.96 ± 0.74) in the chain condition compared with the no chain condition (Table 3).

Table 3.:
Bench throw performance with and without chains.*

When the initial phase of the bench press was analyzed separately, bar velocity was likely to most likely greater at 50 ms (10.9%, ES 0.39 ± 0.51), 100 ms (14.6%, ES 0.50 ± 0.39), 150 ms (15.8%, ES 0.55 ± 0.33), 200 ms (16.2%, ES 0.64 ± 0.29), 250 ms (15.9%, ES 0.70 ± 0.23), 300 ms (15.6%, ES 0.73 ± 0.17), 350 ms (13.8%, ES 0.66 ± 0.19), and 400 ms (9.4%, ES 0.51 ± 0.28) in the chain condition compared with no chain condition (Figure 1).

Figure 1.:
Mean bar velocity from initiation of the bench press exercise at 50–400 ms for chain and no chain conditions. Values are mean ± SD. *Small change and †moderate change.


This study has demonstrated that certain kinematic variables can be augmented with the use of chains during bench throw exercise in well-trained male rugby union players. In particular, peak and mean bar velocity, and bar acceleration are improved with the addition of variable resistance. Our findings support the use of variable resistance for enhancing dynamic muscle function of collision sport athletes.

This study reported small increases in peak (2.8%) and mean (4.7%) bar velocity during the bench press throw, which were greater than the measurement error. These findings are consistent with those of Baker and Newton (5), albeit the relative improvements in performance with the addition chains are smaller than the 10% reported previously. This is similar to the findings of Berning et al. (7) and Coker et al. (9) and is in contrast to Stevenson et al. (32) who noted impaired peak and mean squat velocities with elastic bands. Improvements in mean velocity, peak acceleration, and time to peak velocity may be explained by the reduction in mass at the start of the lift. This supports McCurdy et al. (24) who suggested that a greater acceleration of a bar would occur at the start of the lift because of the lighter total mass. It is well documented that muscles can contract at higher velocities with less force, as illustrated by the inverse nature of the force-velocity relationship (10). In addition, Baker and Newton (5) suggest that, in the chain condition, there is an increase in preparatory muscle stiffness. When the bar is unracked, neural receptors sense the load and recruit the appropriate number of motor units and rate of firing (5). However, during the concentric phase, the load is lighter because of the chains being furled, as such the disparity between this and the load at the bottom of the lift may result in improved bench throw performance (5). A further explanation proposed by Baker and Newton (5) is an enhanced stretch-shortening cycle (SSC). The SSC of exercise incorporates a passive prestretch (eccentric phase), which serves to enhance performance on subsequent active shortening (concentric phase; (29)). During the prestretch, the muscle and series elastic components store energy, the amount of which is relative to the force acting on the muscle (1,36). Theoretically, for both conditions, the elastic energy storage during the eccentric would be similar as the load is the same. At the bottom range of the lift, the SSC is more compliant during the chain condition compared with the control as the load is lighter (5).

It is unsurprising that peak velocity was not different between conditions, given that peak velocity for bench throw exercise occurs near the release point (28). That is, at this point in the movement, the total load would be similar between conditions because the chains would be completely unfurled. This study observed no significant differences in peak and mean power, which is supported by previous findings (32,34). Swinton et al. (34) noted a loss of peak and mean power, despite increased force, during deadlift exercise with chains, which was explained by a loss of velocity during the exercise. Interestingly, we noted higher mean velocity with chains but no changes in power. As power is a product of force and velocity, it is plausible that the lower force negated the higher velocity, especially given that the relative loadings through the movements are lower with chains. Moreover, it is suggested that untrained participants may show greater improvements with the use of variable resistance training (7). That we found no improvement in peak and mean power may be due to the trained status of our participants. That is, they were already highly impulsive athletes due to the nature of the training with which the engage themselves in physically preparing for their sport.

A limitation of this study is the small sample size. However, presenting the magnitude and precision of the data using effect sizes with 90% CIs and magnitude-based inferences addresses any such concerns by indicating the likelihood that the effect is meaningful compared with between-individual variations. These statistics are not susceptible to bias from small sample sizes and should be used to interpret these data.

Practical Applications

The results in this study suggest that the use of variable resistance equivalent to 45% 1RM (30% free weights and 15% chains) can be used as a method to increase certain performance characteristics of the bench press throw. The use of a 15% variable resistance enabled athletes to accelerate and produce a higher mean and peak velocity during upper body exercise, which has been shown to be a key element in sporting performance. As such, athletes who require the ability to develop impulse, such as rugby players, may benefit from the inclusion of this training method. Although the current study only demonstrated the acute effects of variable resistance on the bench press throw, it is recommended that athletes who are speed-strength dominated include the use of chains as a method to enhance acceleration and velocity in upper body movements.


1. Arakawa H, Nagano A, Yoshioka S, Fukashiro S. Interaction between elastic energy utilization and active state development within the work enhancing mechanism during countermovement. J Electromyogr Kinesiol 20: 340–347, 2010.
2. Baker D. A series of studies on the training of high-intensity muscle power in rugby league football players. J Strength Cond Res 15: 198–209, 2001.
3. Baker D. The effects of an in-season of concurrent training on the maintenance of maximal strength and power in professional and college-aged rugby league football players. J Strength Cond Res 15: 172–177, 2001.
4. Baker D, Newton R. Acute effect on power output of alternating an agonist and antagonist muscle exercise during complex training. J Strength Cond Res 19: 202–205, 2005.
5. Baker D, Newton R. Effect of kinetically altering a repetition via the use of chain resistance on velocity during the bench press. J Strength Cond Res 23: 1941–1946, 2009.
6. Batterham A, Hopkins W. Making meaningful inferences about magnitudes. Int J Sports Physiol Perform 1: 50–57, 2006.
7. Berning J, Coker C, Briggs D. The biomechanical and perceptual influence of chain resistance on the performance of the olympic clean. J Strength Cond Res 22: 390–395, 2008.
8. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. New York, NY: Lawrence Erlbaum Associates, 1988.
9. Coker C, Berning J, Briggs D. A preliminary investigation of the biomechanical and perceptual influence of chain resistance on the performance of the snatch. J Strength Cond Res 20: 887–891, 2006.
10. Cormie P, McGuigan MR, Newton R. Developing maximal neuromuscular power: Part 1–biological basis of maximal power production. Sports Med 41: 17–38, 2010.
11. Cronin J, Hansen K. Strength and power predictors of sports speed. J Strength Cond Res 19: 349–357, 2005.
12. Fernandes JFT, Lamb KL, Twist C. The intra- and inter-day reproducibility of the FitroDyne as a measure of multi-jointed muscle function. Isokinet Exerc Sci 24: 39–49, 2016.
13. Garnacho-Castaňo MV, López-Lastra S, Maté-Muñoz JL. Reliability and validity assessment of a linear position transducer. J Sports Sci Med 14: 128–136, 2015.
14. Ghigiarelli J, Nagle E, Gross F, Robertson R, Irrgang J, Myslinski T. The effects of a 7-week heavy elastic band and weight chain program on upper-body strength and upper-body power in a sample of division 1-AA football players. J Strength Cond Res 23: 756–764, 2009.
15. Haff G, Nimphius S. Training principles for power. Strength Cond J 34: 2–12, 2012.
16. Hopkins WG. Spreadsheets for analysis of variability and reliability. Sportscience 19: 36–42, 2015.
17. Hopkins WG. Spreadsheets for analysis of controlled trials, with adjustment for a subject characteristic. Sportscience 10: 46–50, 2006.
18. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 41: 3–13, 2009.
19. Häkkinen K, Komi P, Alén M. Effect of explosive type strength training on isometric force- and relaxation-time, electromyographic and muscle fibre characteristics of leg extensor muscles. Acta Physiol Scand 125: 587–600, 1985.
20. Israetel M, McBride J, Nuzzo J, Skinner J, Dayne A. Kinetic and kinematic differences between squats performed with and without elastic bands. J Strength Cond Res 24: 190–194, 2009.
21. Lyttle AD, Wilson GJ, Ostrowski KJ. Enhancing performance: Maximal power versus combined weights and plyometrics training. J Strength Cond Res 10: 173–179, 1996.
22. Mannie K. Strike-up the band training the benefits of variable resistance. Coach Athl Dir 75: 8–12, 2005.
23. McBride JM, Triplett-McBride T, Davie A, Newton RU. A comparison of strength and power characteristics between powerlifters, Olympic lifters and sprinters. J Strength Cond Res 13: 58–66, 1999.
24. McCurdy K, Langford G, Ernest J, Jenkerson D, Doscher M. Comparison of chain- and plate-loaded bench press training on strength, joint pain, and muscle soreness in division II baseball players. J Strength Cond Res 23: 187–195, 2009.
25. Moss B, Refsnes P, Abildgaard A, Nicolaysen K, Jensen J. Effects of maximal effort strength training with different loads on dynamic strength, cross-sectional area, load-power and load-velocity relationships. Eur J Appl Physiol Occup Physiol 75: 193–199, 1997.
26. Neelly K, Terry J, Morris M. A mechanical comparison of linear and double-looped hung supplemental heavy chain resistance to the back squat: A case study. J Strength Cond Res 24: 278–281, 2009.
27. Newton R, Kraemer W, Häkkinen K. Effects of ballistic training on preseason preparation of elite volleyball players. Med Sci Sports Exerc 31: 323–330, 1999.
28. Newton RU, Kraemer WJ, Häkkinen K, Humphries BJ, Murphy AJ. Kinematics, kinetics, and muscle activation during explosive upper body movements. J Appl Biomech 12: 31–43, 1996.
29. Schenau GJ van I, Bobbert MF, de Haan A. Does elastic energy enhance work and efficiency in the stretch-shortening cycle? J Appl Biomech 13: 389–415, 1997.
30. Schmidtbleicher D, Haralambie G. Changes in contractile properties of muscle after strength training in man. Eur J Appl Physiol Occup Physiol 46: 221–228, 1981.
31. Soria-Gila MA, Chirosa IJ, Bautista IJ, Baena S, Chirosa LJ. Effects of variable resistance training on maximal strength: A meta-analysis. J Strength Cond Res 29: 3260–3270, 2015.
32. Stevenson M, Warpeha J, Dietz C, Giveans R, Erdman A. Acute effects of elastic bands during the free-weight barbell back squat exercise on velocity, power, and force production. J Strength Cond Res 24: 2944–2954, 2010.
33. Swinton P, Lloyd R, Agouris I, Stewart A. Contemporary training practices in elite British powerlifters: Survey results from an international competition. J Strength Cond Res 23: 380–384, 2009.
34. Swinton P, Stewart AD, Keogh JWL, Agouris I, Lloyd R. Kinematic and kinetic analysis of maximal velocity deadlifts performed with and without the inclusion of chain resistance. J Strength Cond Res 25: 3163–3174, 2011.
35. Tricoli V, Lamas L, Carnevale R, Ugrinowitsch C. Short-term effects on lower-body functional power development: Weightlifting vs. vertical jump training programs. J Strength Cond Res 19: 433–437, 2005.
36. Turner AN, Jeffreys I. The stretch-shortening cycle: Proposed mechanisms and methods for enhancement. Strength Cond J 32: 87–99, 2010.
37. Wallace B, Winchester J, McGuigan MR. Effects of elastic bands on force and power characteristics during the back squat exercise. J Strength Cond Res 20: 268–272, 2006.
38. Weir JP. Quantifying test-retest reliability using the interclass correlation coefficient and the SEM. J Strength Cond Res 19: 231–240, 2005.
39. Wilson G, Newton R, Murphy A, Humphries B. The optimal training load for the development of dynamic athletic performance. Med Sci Sports Exerc 25: 1279–1286, 1993.

acceleration; power training; rate of force development

© 2018 National Strength and Conditioning Association