Secondary Logo

Journal Logo

Original Research

Electromyographic and Force Plate Analysis of the Deadlift Performed With and Without Chains

Nijem, Ramsey M.; Coburn, Jared W.; Brown, Lee E.; Lynn, Scott K.; Ciccone, Anthony B.

Author Information
Journal of Strength and Conditioning Research: May 2016 - Volume 30 - Issue 5 - p 1177-1182
doi: 10.1519/JSC.0000000000001351
  • Free



Although traditional free weight resistance is most common, the constant external load does not optimize the muscle's ability to work throughout a full range of motion (ROM). The length tension relationship of muscles, along with the body's lever systems, creates biomechanically disadvantageous positions, known as sticking points (2), for producing maximal force and acceleration. For example, the bottom position of a squat or bench press will serve as the sticking point, whereas positions more toward concentric end ROM allow for greater force to be exerted. Thus, these sticking points limit the external resistance that can be used and effectively do not optimize the ascending strength curves of the squat, bench press, and deadlift. Given the shortcomings of traditional free weight resistance, practitioners have looked to alternative modes of resistance to bypass the sticking point.

Accommodating resistance allows for the resistance to increase in the biomechanical advantageous positions as the muscle is capable of exerting greater force. For example, the use of bands and chains during the bench press, squat, and deadlift provides an increasing load during the concentric phase of the lift to accommodate the ascending strength curve (3,16). This theoretically makes accommodating resistance training superior to traditional free weight training, as more work (force × displacement) is completed using bands or chains. Anecdotally, chains and bands show favorable results for increased strength and power (19–21). Moreover, the empirical evidence on chains and bands is promising, although limited, with studies showing accommodating resistance to affect muscle activation (12,13), power (1,12,25), force (1,12,24,25), velocity (12), impulse (1,24), and rate of force development (RFD) (1,23). Increases in RFD may potentially be the result of a more rapid stretch-shortening cycle, which allows for faster initial concentric velocities (9). In addition, the few training studies available support the use of variable resistance as a means to increase strength and power (2,6,10,18), while also reducing joint stress (15). Moreover, a recent meta-analysis reported variable resistance training to result in greater improvements in strength than traditional weight training (22).

Although practitioners use accommodating resistance to increase performance, a paucity of research exists, with most studies investigating bands. Although elastic bands provide a curvilinear change in force, chains provide a linear change as the chains are lifted vertically (16), making conclusions from band studies difficult to extrapolate to chains. Furthermore, muscle activation literature on the topic is limited to the squat (8,12) and lunge (13), while only 1 study to our knowledge has investigated the deadlift exercise with accommodating resistance (24). This is significant due to the biomechanical differences between a squat and deadlift (11), making conclusions from squat studies difficult to generalize to the deadlift. Therefore, the purpose of this study was to compare the muscle activation (through surface electromyography or EMG), ground reaction force (GRF), and RFD (measured as S-gradient) of the barbell deadlift loaded with free weights and the barbell deadlift loaded with free weights and chains.


Experimental Approach to the Problem

Anecdotally, chains are reported to increase strength, muscle size, and power. However, the literature is limited on chains. Therefore, this study used a randomized (first subject), then balanced, crossover design to investigate the effects of chains on muscle activation, force, and RFD. This study compared muscle activation using surface electromyography (EMG) of the gluteus maximus, vastus lateralis, and erector spinae and RFD and GRF using a force plate during 2 deadlift conditions: (a) the deadlift performed with chains (CH) and (b) the deadlift performed without chains (NC). Participants were instructed to lift with the intent to move as fast as possible during the concentric phase for both conditions, then to lower the barbell under control (Figure 1). Shoes, grip position, and chalk use were determined by the participant but remained consistent between trials. Belts and straps were not allowed.

Figure 1:
Deadlift performed with chain resistance.


Thirteen (mean ± SD age = 24.0 ± 2.1 years, age range = 20–27 years, height = 179.3 ± 4.8 cm, body mass = 87.0 ± 10.6 kg) recreationally resistance-trained men with at least 6 months of deadlift experience (mean ± SD 1RM barbell deadlift = 176.2 ± 20.5 kg) volunteered for this study. Before participating in the study, subjects signed a statement of informed consent approved by the University Institutional Review Board. All participants were free of musculoskeletal injuries or other conditions that would hinder their ability to participate. Subjects were instructed to refrain from resistance training between trials.


One Repetition Maximum Testing and Experimental Conditions

On the first of 2 days of testing (minimum of 48 hours but no more than 120 hours between visits), participants read and signed an informed consent form before performing a standardized dynamic warm-up (4) consisting of 2 repetitions of 20 m: A-skips, high knees, butt kicks, lunges, cariocas, and running backward. Subjects then followed a standard testing protocol (7) to test 1 repetition maximum (1RM) for the deadlift exercise. Participants completed 10 repetitions at 50% of their predicted 1RM, 5 repetitions at 70%, 3 repetitions at 80%, and 1 repetition at 90%. Participants then completed up to 5 single-repetition sets to determine their 1RM. Two minutes were given between warm-up sets and 5 minutes between 1RM attempts. The process of increasing the weight was performed in increments of 10–20 pounds until a load that allowed for completion of the actual 1RM was reached. If the participant was not able to execute the lift successfully, the weight was reduced by 5–10 pounds until a successful 1RM was completed. A successful 1RM required full knee and hip extension as determined by a National Strength and Conditioning Association (NSCA) Certified Strength and Conditioning Specialist.

The 1RM was recorded to determine loads used for the experimental trial conditions on day 2 when subjects performed one set with a load of 85% 1RM of 3 repetitions with chains and one set of 3 repetitions without chains. The order of conditions (CH and NC) was randomly determined. For the CH condition, the chains accounted for approximately 20% (19.9 ± 0.6%) of the 85% 1RM. The load of the chain was quantified by having the subject stand upright on the force plate (AMTI force platform; Advanced Mechanical Technology, Inc., Watertown, MA, USA) with an unloaded barbell and the safety clips used to hold the free weight plates and chains in place before zeroing the force plate. Chains were then added to each side until the desired load of approximately 20% of the 1RM was reached. These chains, plus the remaining load needed from free weights were added to the barbell during the CH condition, with chains being placed outside the free weights. Free weights made up the entire load during the NC condition. We chose to equate loads at the top of the lift based on frequent practitioner use, and because this was believed to positively facilitate acceleration of the barbell from the bottom position of the deadlift. Participants were instructed to pull the barbell explosively with maximal effort on the concentric phase of every repetition for each condition, then to lower the barbell back to the floor slowly and under control.


Surface EMG was used to measure muscle activation during chain (CH) and no chain (NC) conditions, concentric and eccentric phases, and top and bottom ROM of the 1RM test, always using the third repetition of the experimental trials.

To avoid confounding the EMG signal, the participant's skin was prepared by shaving the hair at the placement site and cleaning the site with isopropyl alcohol. Electromyography data were collected and stored on a personal computer (Dell Latitude D610; Dell, Round Rock, TX, USA). Three separate bipolar (3.5 cm center to center) surface electrode (BIOPAC EL500 silver-silver chloride; BIOPAC Systems, Inc., Goleta, CA, USA) arrangements were placed over the longitudinal axes of the gluteus maximus, erector spinae, and vastus lateralis muscles, with the reference electrodes placed over the iliac crest ( Electrodes for the gluteus maximus were placed on the line between the sacral vertebrae and the greater trochanter, corresponding with the greatest prominence of the middle of the buttocks. Electrodes for the erector spinae were placed at a width of 2 fingers lateral from the spinous process of L1. Electrodes for the vastus lateralis were placed 2/3 on the line from the anterior spina iliaca superior to the lateral side of the patella. All measurements were taken on the left side of the participant's body. The EMG signals were preamplified (gain 1,000×) using a differential amplifier (EMG 100C, bandwidth = 1–500 Hz; BIOPAC Systems, Inc., Santa Barbara, CA, USA).

Signal Processing

The EMG signals were band-pass filtered (fourth-order Butterworth) at 10–500 Hz. The amplitudes of the signals were expressed as root mean square (rms) values. All analyses were performed with custom programs written with LabVIEW software (version 7.1; National Instruments, Austin, TX, USA). The EMG rms values for the third repetition for the experimental trials were used for analyses. The EMG rms values were calculated separately for the bottom half and top half of the concentric and eccentric phases, which were determined using the visual display of the data collection software (AcqKnowledge 3.8.1; BIOPAC Systems, Inc.). Data for each of the muscles tested were then normalized to each subject's highest recorded values achieved during the concentric phase of the deadlift 1RM testing. Intraclass reliability values exceeding 0.9 were found for EMG amplitude values.

Ground Reaction Force and Rate of Force Development

A force plate was used to measure GRFs and RFD of the experimental trials (CH and NC). The force plate (Advanced Mechanical Technology, Inc.) sampled at 1,000 Hz and interfaced with a personal computer containing an A/D converter and LabVIEW software (LabVIEW, version 7.1; National Instruments Corporation), which provided values for peak force and RFD during the concentric phase of the deadlifts. Rate of force development was calculated using the S-gradient formula (S-gradient = F0.5/T0.5, where F0.5 is one-half of maximal force [Fm] and T0.5 is time to achieve that force) (26). Intraclass reliability values between 0.8 and 0.9 for force plate measures have previously been reported from our laboratory.

Statistical Analyses

A 2 (condition: CH, NC) × 2 (phase of movement: concentric, eccentric) × 2 (ROM: bottom, top) repeated-measures analysis of variance was used to analyze the normalized EMG amplitude data for each muscle. Follow-up tests included 1-way analysis of variances or paired samples t-tests with Bonferroni corrections as appropriate. Paired sample t-tests were run to determine peak GRF and RFD differences between conditions (CH and NC). An alpha level of 0.05 was used to determine statistical significance. IBM SPSS Statistics 21 (IBM Corporation, Somers, NY, USA) was used to perform all statistical analyses.



Table 1 summarizes the mean normalized EMG amplitude values for the conditions (CH and NC) separated by phase (concentric and eccentric) and ROM (bottom and top). The results revealed no significant 3-way or 2-way interactions (p > 0.05). However, for the gluteus maximus, there were significant main effects for condition and phase (p ≤ 0.05). Significantly greater EMG activity (p ≤ 0.05) was present during the NC condition vs. the CH condition. In addition, the concentric phase resulted in significantly greater EMG activity (p ≤ 0.05) than the eccentric phase. For the erector spinae, there were significant main effects for ROM and phase (p ≤ 0.05). Compared with the top ROM, the bottom ROM had significantly greater EMG activity (p ≤ 0.05). In addition, the concentric phase demonstrated significantly greater EMG values than the eccentric phase. Electromyography data for the vastus lateralis revealed there was a main effect for phase, with significantly greater EMG activity (p ≤ 0.05) for the concentric vs. eccentric phase.

Table 1:
Normalized electromyographic (EMG) amplitude values (mean ± SEM).*†

Force Plate

Table 2 summarizes peak GRFs and RFD values between conditions. The results revealed that deadlifting at 85% with an accommodating chain resistance of approximately 20% results in a significant reduction in GRF (p ≤ 0.05) and no change in RFD (p > 0.05).

Table 2:
Ground reaction force (GRF) and rate of force development (RFD; measured as S-gradient) values (mean ± SD).


To our knowledge, this study is the first to compare muscle activation and force-time characteristics of the deadlift performed with and without chains as an accommodating resistance. Muscle activation for the gluteus maximus was significantly greater for the NC condition. Muscle activation levels for the erector spinae and vastus lateralis muscles were unaffected by chain use. The deadlift exercise, independent of chain use, demonstrated differences in the muscle activity of the erector spinae, gluteus maximus, and vastus lateralis during the different points in the ROM (bottom, top) and phases (concentric, eccentric). When match loading at the top, the use of chains as a variable resistance of 20% of the 85% 1RM load during the deadlift exercise resulted in decreased GRF but did not affect RFD.

Normalized EMG amplitude (reflecting muscle activation) of the gluteus maximus was greater during the NC condition. This may have been due to the greater average load when lifting without chains, potentially altering lifting technique compared with the CH condition. This cannot be directly confirmed, as we did not measure the kinematics of the 2 deadlift conditions. However, others have also found differences in the patterns of muscle activation through EMG during variations of the deadlift exercise. For example, muscle activation of the vastus medialis, vastus lateralis, and tibialis anterior were greater during the sumo deadlift than during the conventional deadlift exercise (9). These altered patterns of muscle activation with changes in technique and the use of chains emphasize the need for the strength and conditioning coach to carefully consider the implications of these methods on muscle activation and specificity of training.

It was hypothesized that normalized EMG amplitude would be increased for the top ROM for both concentric and eccentric phases using chains, as the resistance would be gradually increasing as the barbell reached the top of the lift. However, for the erector spinae, EMG amplitude was greater at the bottom of the concentric and eccentric, regardless of the chain condition. This may have been due to a more active upper back musculature (i.e., latissimus dorsi, rhomboids, and trapezius) during the concentric phase as the bar is being pulled up. During the eccentric phase, as the bar is being lowered, these muscles may become less active which transfers force to the lumbar spine and consequently the erector spinae. This cannot be confirmed, however, as we did not measure upper back musculature EMG. In any case, increased EMG amplitude from the erector spinae likely reflects that longer moment arm of the torso and barbell to the axis of rotation of the lumbar spine in the lowest position of the deadlift as it is lifted off of, and returned to, the floor. This would be true regardless of whether or not chains are placed on the barbell.

It was further hypothesized that EMG activity for erector spinae would be decreased during the CH condition, as the load was lighter at the initiation of the movement than the NC condition. The chains allowed for the load to be lightest during the sticking point of the deadlift, and the variable resistance of the chains increased the load with the biomechanical advantage of the ascending strength curve (16). There was, however, no difference in EMG amplitude between chain conditions for the erector spinae. With equal amounts of low back muscular activation with and without chains during the deadlift with the loading strategy used, the shear stress on the lumbar spine becomes of interest. Although the low back muscles were equally active between conditions, it cannot be assumed that lumbar shear and compressive forces were equal as lumbar kinematics were not measured. This information would indicate the role of chains in injury prevention during the deadlift exercise. It is speculated that chains during the deadlift when match loaded at the top position would allow for a neutral spine to be maintained, resulting in less stress on the soft tissues of the lumbar spine. Further research is needed, however, to confirm or refute this.

Peak GRF was, as hypothesized, greater without chains. This may be a result of the greater initial load at the bottom of the lift, and the greater average load, without chains. Match loading at the top, although practical, serves as a limitation in terms of loading at the bottom of the lift. Under the NC condition, the lift required greater effort at the bottom of the lift, more total work (force × distance) to be completed, and likely led to the increased peak GRF. Considerations, therefore, must be given to GRFs during different match loading comparisons (e.g., average load or at the initiation of the deadlift), as it is currently unknown how different loading strategies would influence peak GRF.

There was no difference in RFD between conditions, which is inconsistent with the results of the only other deadlift study performed with chains as accommodated resistance. Swinton et al. (24) reported a decrease in bar velocity along with decreased RFD when adding chains. The RFD results contribute to the controversy of training for RFD. It has been suggested that RFD is dependent on the intent to move fast, and not the actual velocity (5), whereas others have suggested that increasing bar velocity is important for increasing RFD and high velocity capabilities (14,17). Swinton et al. (24) reported a decrease in bar velocity along with decreased RFD when adding chains to the deadlift exercise, consequently supporting the suggestion that actual bar velocity is an important variable for RFD. Contrastingly, the current investigation found RFD to be similar between CH and NC conditions. Because all subjects were instructed to move the bar as fast as possible, the results support the notion that intent to move fast is more important than actually moving fast. Given that bar velocity was not measured, the effect of bar velocity is unknown. It should also be noted that in this study, the average RFD value for the CH condition (5,155.7 N·s−1) was higher than for the NC condition (4,128.1 N·s−1). However, the large variability between participants likely prevented this difference from reaching statistical significance. Perhaps, testing subjects with more experience training with chains would decrease this variability, providing a clearer picture regarding the influence of deadlift variations on RFD.

Future research should measure kinetics and kinematics, along with force and EMG to better understand how joint angles and forces (e.g., shear stress at the lumbar spine) are altered when using an accommodating resistance, such as chains. Measuring the EMG of different muscles (e.g., the hamstrings, latissimus dorsi, rhomboids, etc) along with different intensities of 1RM (e.g., 65%), and different accommodated percentage loads (e.g., 30%) would also be of interest, as the literature on the deadlift performed with chains is limited to one other study (24). In addition, future research should investigate different load matching comparisons (e.g., average load or matched at the initiation of the deadlift), as this likely influences the variables of interest as well. The differences in muscle activation and force characteristics using chains in this study warrant further research examining the optimal use of chains for increasing deadlift 1RM, and transfer to other activities, such as jumping and sprinting.

Practical Applications

These results suggest that using chains as a 20% accommodating resistance during the deadlift exercise, using loads of 85% 1RM and match loaded at the top position, will decrease gluteus maximus activation and peak GRF but has no effect on RFD when compared with the deadlift exercise performed without chains. If maximizing gluteus maximus activation or increasing peak GRF is the goal, the use of chains in a similar fashion as performed in this study is not advised. The effects on lumbar forces which has injury risk implications are unknown, although it is speculated that chain-accommodated resistance would allow for a more neutral spine to be maintained during the execution of the deadlift as the load in the sticking point will be lightest. Muscle activation throughout the deadlift differs by muscle, phase, and ROM. Strength and conditioning coaches and fitness professionals should determine their implementation of the deadlift exercise with or without chains based on the goals and training status of the population they are working with.


1. Aboodarda S, Yusof A, Abu Osman N, Thompson M, Mokhtar A. Enhanced performance with elastic resistance during the eccentric phase of a countermovement jump. Int J Sports Physiol Perform 8: 181–187, 2013.
2. Anderson C, Sforzo G, Sigg J. The effects of combining elastic and free weight resistance on strength and power in athletes. J Strength Cond Res 22: 567–574, 2008.
3. 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.
4. Bartolini J, Brown L, Coburn J, Judelson D, Spiering B, Aguirre N, Carney K, Harris K. Optimal elastic cord assistance for sprinting in collegiate women soccer players. J Strength Cond Res 25: 1263–1270, 2011.
5. Behm D, Sale D. Intended rather than actual movement velocity determines velocity-specific training response. J Appl Physiol 74: 359–368, 1993.
6. Bellar D, Muller M, Barkley J, Chul-Ho K, Ida K. The effects of combined elastic- and free-weight tension vs. free-weight tension on one-repetition maximum strength in the bench press. J Strength Cond Res 25: 459–463, 2011.
7. Doan B, Newton R, Marsit J, Triplett-McBride N, Koziris L, Fry A, Kraemer W. Effects of increased eccentric loading on bench press 1RM. J Strength Cond Res 16: 9–13, 2002.
8. Ebben W, Jensen R. Electromyographic and kinetic analysis of traditional, chain, and elastic band squats. J Strength Cond Res 16: 547–550, 2002.
9. Escamilla R, Francisco A, Kayes A, Speer K, Moorman C. An electromyographic analysis of sumo and conventional deadlifts. Med Sci Sports Exerc 34: 682–688, 2002.
10. 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.
11. Hales ME, Johnson BF, Johnson JT. Kinematic analysis of the powerlifting style squat and the conventional deadlift during competition: Is there a cross-over effect between lifts?. J Strength Cond Res 23: 2574–2580, 2009.
12. 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, 2010.
13. Jakobsen M, Sundstrup E, Andersen C, Aagaard P, Andersen L. Muscle activity during leg strengthening exercise using free weights and elastic resistance: Effects of ballistic vs. controlled contractions. Hum Mov Sci 32: 65–78, 2013.
14. McBride J, Triplett-McBride T, Davie A, Newton R. The effect of heavy- vs. light-load jump squats on the development of strength, power, and speed. J Strength Cond Res 16: 75–82, 2002.
15. 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.
16. McMaster D, Cronin J, McGuigan M. Forms of variable resistance training. Strength Cond J 31: 50–64, 2009.
17. Morrissey M, Harman E, Frykman P, Han K. Early phase differential effects of slow and fast barbell squat training. Am J Sports Med 26: 221–230, 1998.
18. Rhea M, Kenn J, Dermody B. Alterations in speed of squat movement and the use of accommodated resistance among college athletes training for power. J Strength Cond Res 23: 2645–2650, 2009.
19. Simmons L. Bands and chains. Powerlifting USA 22: 26–27, 1999.
20. Simmons L. The Westside Barbell Book of Methods. Ebook. Columbus, OH: Westside Barbell, 2007. pp. 94–231.
21. Simmons LP. Chain reactions: Accommodating leverages. Powerlifting USA 19: 2–3, 1996.
22. Soria-Gila M, Chirosa I, Bautista I, Chirosa L, Salvador B. Effects of variable resistance training on maximal strength: A meta-analysis. J Strength Cond Res 2015. Epub ahead of print.
23. 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.
24. Swinton P, Stewart A, Keogh J, 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.
25. Wallace B, Winchester J, McGuigan M. Effects of elastic bands on force and power characteristics during the back squat exercise. J Strength Cond Res 20: 268–272, 2006.
26. Zatsiorsky VM, Kraemer WJ. Science and Practice of Strength Training. Champaign, IL: Human Kinetics, 2006.

EMG; muscle activation; ground reaction force; rate of force development; variable resistance; accommodating resistance

Copyright © 2016 by the National Strength & Conditioning Association.