Secondary Logo

Journal Logo

Original Research

The Effect of Cadence on Cycling Efficiency and Local Tissue Oxygenation

D. Jacobs, Robert; E. Berg, Kris; Slivka, Dustin R.; Noble, John M.

Author Information
Journal of Strength and Conditioning Research: March 2013 - Volume 27 - Issue 3 - p 637-642
doi: 10.1519/JSC.0b013e31825dd224
  • Free



Efficiency is an important aspect of endurance performance. In fact, some studies have found efficiency to be a stronger predictor of performance than V[Combining Dot Above]O2max (14,15,23). Pedaling cadence is one factor that affects cycling performance and efficiency. The evolution of high-cadence cycling in the professional peloton was first seen with the 5-time Tour de France winner Miguel Indurain (1991–95). High-cadence cycling was made popular by the 7-time Tour de France winner Lance Armstrong (1999–2005). Although high-cadence cycling has become popular in recent years by elite and profession cyclists, the scientific evidence supporting high-cadence cycling remains controversial. Consequently, it is unknown if high-cadence cycling is beneficial for the nonelite athlete.

Measurement of cycling efficiency is based on aerobic energy expenditure (19,27). Gross and delta efficiency and economy are commonly used to determine cycling efficiency. However, the current literature as to which type of efficiency/economy is most valid when investigating cycling performance is inconclusive (18,19,23). Also, studies have often used relatively short work bouts to assess efficiency/economy (18,19). Short exercise durations may not include the V[Combining Dot Above]O2 slow component that characterizes exercise intensities at or above ventilation threshold (2), which may alter the calculated efficiency and economy.

Lucia et al. (16) demonstrated that when professional cyclists are riding at 75% of V[Combining Dot Above]O2max, gross efficiency (GE) is significantly higher and blood lactate is significantly lower at 100 vs. 60 rpm. Professional cyclists appear to prefer higher cadences (14,16,18). Data observed in professional cyclists while participating in the 3 biggest stage races in the world (Giro d' Italia, Tour de France, and Vuelta a Espana) suggest that preferred cadence varies with power output and terrain. For instance, the mean cadence used was 90 and 70 rpm for flat roads and mountain passes, respectively (14). Therefore, power output should be considered as a key factor in studying cadence optimization and efficiency of performance (14,16).

It has been suggested that hemodynamics and blood flow play a role in cycling performance and efficiency. However, few studies have examined hemodynamics and blood flow while cycling (13,20,25). Furthermore, to the authors' knowledge, only a study by Takashi et al. (25) investigated how cadence influences blood flow and intramuscular oxygen saturation levels, but efficiency was not addressed.

The primary question we examined in this study was the influence of cycling cadence on efficiency/economy in nonelite competitive athletes. Specifically, the purpose was to investigate the effects of various cadences (60, 80, 100 rpm) on cycling efficiency (GE and economy), selected physiological variables (local tissue oxygen saturation, heart rate, blood lactate), and both global and local rating of perceived exertion (RPE). We also compared data at minute 4 and minute 8 of each work bout to assess the effect of exercise duration on the variables. We hypothesized that the slowest cadence would be the least efficient and economical and would elicit higher heart rate, blood lactate, and RPE values in comparison with the moderate and high cadences.


Experimental Approach to the Problem

We used nonelite competitive athletes to test the influence of 3 cycling cadences on cycling efficiency/economy and various physiological descriptors. High-cadence cycling has become popular in elite and professional cyclists, but research is needed to determine if it is beneficial to athletes of lesser ability.

Two testing days separated by a minimum of 48 hours were required for data collection. On the first day of testing, a graded exercise test was conducted to assess peak V[Combining Dot Above]O2 and peak watts. This information was used to determine the workload for the second testing session that consisted of cycling at a constant submaximal work rate at cadences (independent variable) of 60, 80, and 100 rpm to determine the effect of cadence on cycling efficiency (dependent variable). Cycling was performed in three 8-minute trials at 75% of the maximal power attained during the peak incremental test. Each 8-minute trial was separated by a minimum of 10 minutes of rest. The sequence of testing was randomly assigned. The pedal cadences selected for study were based on the observation of professional cyclists during competition (14). Hence, the cadences selected in this study are similar to the preferred cadences used by professional cyclists.

All cycle tests were performed on each participant's bicycle attached to a Computrainer Pro cycle ergometer (RacerMate, Inc., Seattle, WA, USA), thus assuring of the same fit for each trial and testing session. The peak V[Combining Dot Above]O2 peak test started at 60 W with the workload increased by 35 W every 3 minutes until the subjects were unable to maintain a cadence of 80 rpm for 2 consecutive minutes (19). Participants were given visual feedback of their cadence via the computer monitor connected to the Computrainer Pro. Data from the last minute of testing were used to obtain values for peak V[Combining Dot Above]O2, heart rate, and watts. The Computrainer Pro was calibrated according to manufacturer's recommendations before all tests.


The participants were 14 trained competitive male cyclists and triathletes (N = 14) from local competitive teams. They had a minimum of at least 1 year of competitive cycling experience. They were in a period of base training at the time of the study where most training was of moderate intensity. They were all injury free and completed an informed consent approved by the University of Nebraska Medical Center, University of Nebraska at Omaha Institutional Review Board for Human Subjects before data collection.

Experimental Trials

The submaximal tests consisted of three 8-minute cadence trials performed at 75% of the peak watts recorded during the graded exercise test. Eight-minute stages have been demonstrated to be of adequate duration to assess cycling efficiency (10,11). The second and third cadence trials did not begin until a minimum of 10 minutes after completing the previous trial. Additional time was given if needed until the subject's heart rate dropped to 10 b·min−1 of the resting value attained after 5 minutes of quiet sitting on the first day of testing. Three cadence trials were performed in a random sequence at 60, 80, and 100 rpm to prevent an order effect. A minimum of 48 hours between testing sessions was required to minimize the effects of fatigue. Also, participants were instructed to refrain from strenuous exercise for a period of 24 hours before the testing sessions and to maintain a consistent diet between sessions. All tests were conducted at the same time of day ±2 hours.


Descriptive data were recorded for age, height, mass, and %body fat. Height and weight were recorded with a medical scale (Detecto, Webb City, Missouri, USA). Body density was estimated using the 3 skinfold site (chest, abdomen, and thigh) regression equation by Jackson and Pollock (12). Harpenden skinfold calipers (model 68875; Baty International, United Kingdom) were used to determine the thickness of the sites, and each site was measured a minimum of 2 times with a difference of no more than 1.0 mm. The Siri equation (24) was used to determine percent body fat from body density.

A graded cycle ergometer test was used to assess V[Combining Dot Above]O2 peak. Expired gases were collected and analyzed using a calibrated ParvoMedics metabolic cart (ParvoMedics, Sandy, UT, USA). The metabolic cart was calibrated to the manufacturers' recommendations before any testing sessions. Ventilatory threshold was determined via the ParvoMedics True One 2400 metabolic cart function.

Blood lactate was measured before the warm-up and in minutes 4 and 8 of each submaximal test. Blood lactate was assessed via a Lactate Pro (Nova Biomedical, Waltham, MA, USA) handheld blood lactate analyzer. The blood was drawn via finger stick with the first blood drop wiped away to avoid contamination from alcohol and sweat.

Heart rate was measured using a Polar heart rate monitor and transmitter (Polar Electro, Lake Success, NY, USA). Heart rate was recorded at the end of every stage during the incremental test and throughout the submaximal workload trials.

Intramuscular oxygen saturation was measured during the three submaximal cadence trials using Inspectra StO2 Tissue Oxygenation Monitor (Hutchinson Technology, Hutchinson, MN, USA). The probe for the StO2 Tissue Oxygenation Monitor was placed half way between the anterior superior iliac spine and the superior border of the patella over the vastus lateralis.

The 15-point RPE scale (3) was used during the V[Combining Dot Above]O2 peak test and the constant workload cadence trials. Rating of perceived exception was measured using a black and white RPE rating scale chart at the end of each stage of the peak test and minutes 4 and 8 of each submaximal trial. Participants were asked to report their RPE both locally and globally. Global RPE represents an overall feeling of the difficulty of exercise, whereas local RPE indicates how hard just the legs are working (6). The influence of observation on RPE was controlled for during data collection by not allowing people in the laboratory during collection other than the investigator and an assistant. No verbal encouragement was given during the submaximal trials.

Statistical Analyses

Descriptive statistics were calculated for each variable. Two-way (cadence × time) repeated-measures analysis of variance was conducted for each dependent variable: efficiency, economy, intramuscular oxygen saturation, blood lactate, and local and global RPE. Data for minutes 4 and 8 were compared on these variables. In the event of a significant F-value, Fisher's protected least significant difference method was used to identify where the differences occurred. A probability of type I error less than 5% was considered significant (p < 0.05). All data are reported as mean ± SD. All statistics were performed on SPSS 17.0 for Windows (Microsoft, Redmond, WA, USA). Sample size to achieve statistical power of 0.80 was estimated from data (e.g., mean, SD, t ratios) of other studies on cycling efficiency and was found to require approximately 10–12 subjects. Other studies of cycling efficiency have reported significant differences with even smaller sample sizes (16,18).



Fourteen locally competitive male cyclists and triathletes completed testing, and their descriptive statistics are provided in Table 1.

Table 1:
Descriptive data of participants (N = 14).

Gross Efficiency and Economy

As cadence increased, efficiency decreased. Cycling at 60 rpm was more efficient than at 80 rpm (p = 0.031) and 100 rpm (p < 0.001), and cycling at 80 rpm was more efficient than 100 rpm (p < 0.001). Efficiency was also influenced by duration of cycling time because it decreased from minute 4 to minute 8 (p < 0.001, Figure 1A). As cadence increased, economy also tended to decrease. Economy was higher at 60 rpm than 100 rpm (p < 0.001) and showed a trend at 80 rpm (p = 0.071). Additionally, economy decreased from minute 4 to minute 8 (p < 0.001, Figure 1B).

Figure 1:
Gross mechanical efficiency (A) and cycling economy (B) for 60-, 80-, and 100-rpm cycling cadences after 4 and 8 minutes of exercise. Data are mean ± SD. a p < 0.05 from all other means for cadences (main effect). b p < 0.05 from minute 8 (main effect). c p < 0.05 from 60 rpm (main effect).

Local Tissue Oxygen Saturation

Tissue oxygen saturation was higher at 80 rpm (59 ± 10%) than at 60 (52 ± 12%, p = 0.018) and 100 rpm (55 ± 10%, p = 0.008) at minute 4, but at minute 8, tissue oxygen saturation at 80 rpm (57 ± 9%) was higher than at 100 rpm (54 ± 9%, p = 0.017) but not different than at 60 rpm (55 ± 11%, p = 0.209).

Heart Rate and Lactate

Heart rate was lower at 60 rpm than at 80 rpm (p = 0.026) and 100 rpm (p < 0.001), and HR at 80 rpm was less than at 100 rpm (p < 0.001) regardless of time. Additionally, HR was higher at minute 8 than at minute 4 (p < 0.001) regardless of cadence (Table 2). Lactate at 100 rpm was greater than at 60 (p = 0.039) and 80 rpm (p = 0.048) regardless of time. Furthermore, lactate at minute 8 was higher than at minute 4 (p = 0.001) regardless of cadence (Table 2).

Table 2:
Heart rate and lactate for 60-, 80-, and 100-rpm cycling cadences after 4 and 8 minutes of exercise (mean, SD, and range).

Rating of Perceived Exception

Global RPE was higher at 100 rpm (13.3 ± 1.5) than at 60 rpm (12.2 ± 1.5, p = 0.010) and 80 rpm (12.1 ± 1.4, p = 0.003) at minute 8. but no differences were observed at minute 4 at 60 rpm (11.6 ± 1.9), 80 rpm (12.1 ± 1.7), or 100 rpm (12.3 ± 1.6). Global RPE was also higher at minute 8 than at minute 4 for the 100-rpm cadence (p < 0.001), but not at 60 or 80 rpm (p = 0.104 and 0.498, respectively). Local RPE was lower at 80 rpm (12.6 ± 1.4) than at 100 rpm (13.7 ±1.5, p < 0.001) during minute 8 with no differences found during minute 4. In addition, local RPE was lower during minute 4 than minute 8 at 60 rpm (12.4 ± 1.7, p < 0.001) and 80 rpm (12.5 ± 1.4, p = 0.027), but only a trend toward significance at 100 rpm (12.7 ± 1.5, p = 0.058).


The main finding of this study was that cycling efficiency and economy were higher at 60 rpm than 80 and 100 rpm. Heart rate and lactate were lowest during the 60 rpm trial and likely reflect the lower V[Combining Dot Above]O2 and higher efficiency associated with this cadence. Our findings on efficiency and economy are in accordance with those of Neilsen et al. (21) but in opposition with those of Lucia et al. (16) and Mora-Rodriguez and Aguado-Jimenez (18). The difference in results between this study and the Lucia et al. (16) study may be because of the training status of the subjects. Our study used trained local cyclists, whereas Lucia et al. studied 8 world-class professional cyclists. Protocol difference might explain the lack of agreement between this study and Mora-Rodriguez and Aguado-Jimenez (18). They used ramping protocols performed at cadences of 80, 100, and 120 rpm rather than constant work rate trials performed in this study. Based on the results of this study, the 4-minute stages of the ramping protocol used by Mora-Rodriguez and Aguado-Jimenez (18) might have been too short to accurately determine efficiency. In our study, there was a continued decline in efficiency with time and cadence, and use of a 4-minute ramping protocol would likely underestimate steady-state V[Combining Dot Above]O2, thereby increasing the calculation of efficiency. Mora-Rodriguez and Aguado-Jimenez (18) observed that efficiency decreased as cadence increased, although not significantly. Furthermore, their study with a sample size of 9 may have been underpowered thus explaining the lack of statistical significance reported. Consequently, it is difficult to compare results across studies because of variance in level of training and protocol used to assess efficiency and economy.

Our data support the use of longer stages to assess efficiency and economy than 4 minutes or less as used by others (18,19). When work rate exceeds ventilation threshold (VT) or lactate threshold (LT), V[Combining Dot Above]O2 continues to rise as work continues at this rate. This phenomenon is known as the V[Combining Dot Above]O2 slow component (2). It would be helpful to assess VT or LT in studies on cycling efficiency to select work rates either below this level to preclude the effects of the V[Combining Dot Above]O2 slow component or to use longer stages where much of the slow component rise may be captured. The VT of our subjects averaged 70% of peak work rate, which was below the 75% of peak work rate used in the 3 cadence trials. However, in 11 of our subjects, the 70% peak work rate exceeded their VT. Consequently, we assessed efficiency and economy using data from minutes 4 and 8. Both efficiency and economy decreased over time, thus supporting this analysis. Consequently, we recommend that efficiency be calculated from data with stages lasting longer than 4 minutes as recommended by others (9,10) when exercise intensity may approach or be above VT.

The effect of the slow component V[Combining Dot Above]O2 is seen in most of the physiological data analyzed over time. Higher responses occurred for HR and lactate at minute 8 than minute 4 for each cadence that corresponds to the fact that 11 of our 14 participants exercised above their VT. Local RPE was higher in minute 8 than minute 4 at all 3 cadences, whereas global RPE was higher at 2 of the cadences (60 and 100 rpm).

Local tissue oxygen saturation levels have not been frequently investigated during cycling. Our findings indicate that cycling at 80 rpm produces greater tissue oxygen saturation levels than cycling at 60 or 100 rpm. This finding is interesting because as subjects increased cadence from 60 to 80 rpm, their V[Combining Dot Above]O2 also increased. These findings seem contradictory since O2 dissociation typically increases in parallel with V[Combining Dot Above]O2 to facilitate peripheral oxygen uptake. This mechanism for the lower saturation levels during the 60-rpm trial is not likely because of higher intramuscular pressure caused by increased force of muscle contraction (25). Increased muscular force in theory would reduce blood flow and increase peripheral resistance by occluding vessels, which would seemingly increase tissue oxygen saturation levels (25). Our results showed that both local and global RPE increased with time during the 60- and 100-rpm cadence trials. These observations agree with those of Garcin et al. (6). Local RPE in our study, however, remained constant at 80 rpm. This latter finding was unexpected because RPE closely parallels physiological variables such as heart rate, lactate, and respiration (7,22). All these variables increased with time during the 80-rpm cadence, and it would logically follow that local RPE would have increased as well.

We found efficiency and economy to be higher at 60 rpm than cycling at higher cadences in contrast to studies on elite cyclists, where no differences in gross economy were observed among 80, 100, and 120 rpm (18). The discord in our results on efficiency in comparison with studies using elite cyclists may be the result of various physiological adaptations made over years of high volume training in elite athletes. For example, Coyle et al. (5) found that the percentage of type 1 fibers was significantly correlated to GE (r = 0.75, p < 0.001). The conversion of type 2 to type 1 fibers has been demonstrated in endurance athletes using high volume training (5). World-class cyclists possess greater oxidative enzyme capacity and mitochondrial volume than moderately trained cyclists (4,26). Elite cyclists also possess greater capillary density because of training intensity or volume (8). Increased capillary density would aid in the transport of oxygen and removal of lactate and hydrogen ions and increase the delivery of HCO−3 to aid in muscle buffering capacity. Furthermore, the greater volume of training by elite athletes may enhance the efficiency of neuromotor recruitment (23). Thus, the unique physiological profile of elite cyclists may explain why they are more efficient at higher cycling cadences than moderately trained persons.

It is also possible that the cyclists in our study were unfamiliar with the 60- and 100-rpm cadences. The uniquely slow and fast rpm of 60 and 100, respectively, may require specific training at those cadences to optimally recruit motor units. Lack of familiarization with specific cadences is common in research on cycling efficiency (1,17), which may skew the results favoring cadences commonly used. Thus, most studies on cycling efficiency do not account for the possible motor learning that may occur when using various cadences. It may be helpful to incorporate in the design of studies the opportunity to train at specific cadences for a period before assessing efficiency to allow assessment of a possible effect of motor learning on cycling efficiency.

It should be noted that the cyclists in our study were in a base period of training during the time of data collection when cycling tends to occur mostly at moderate intensity with little or no interval training. Interval training enhances efficiency, peak watts, and V[Combining Dot Above]O2 peak (10), and our results may have been modified had testing been conducted during the competitive season.

The following conclusions are warranted from the results of this study:

  1. Trained local cyclists and triathletes are most efficient and economical when cycling at 60 rpm vs. 80 or 100 rpm.
  2. At a constant work rate, local tissue oxygen saturation levels are higher at 80 rpm than 60 or 100 rpm.
  3. Heart rate and lactate rose with increased cadence and duration of cycling.
  4. Global RPE is lower when cycling at 60 or 80 rpm vs. 100 rpm, but local RPE is lowest at 80 rpm.

Practical Applications

The findings of this study may be helpful to coaches and athletes in determining optimal efficiency to enhance performance. Efficiency is an important determinant of cycling performance, and this study suggests that 60 rpm is more efficient than 80 and 100 rpm for moderately well-trained cyclists and triathletes. Pedaling at 60 rpm was 4.5 and 11.5% more efficient and 4.2 and 10.7% more economical than at 80 and 100 rpm, respectively. Differences of this magnitude in competition are likely to yield meaningful improvement in performance because in competition improved economy likely translates to faster cycling speed and greater endurance.

This study used a work rate of 75% of peak power, and therefore our results may not be applicable to higher and lower work rates. Individual athletes might experiment with various cadences and gear ratios to personally optimize efficiency. Road cycling is not performed at a constant work rate, which also suggests value in individualizing cadence depending on terrain, wind, and other environmental conditions. Use of 60 rpm may be an effective cadence with which to compare other pedaling speeds.


Funding for this study was provided by a Graduate Thesis Scholarship from the University of Nebraska at Omaha.


1. Bernard T, Vercruyssen F, Grego F, Hausswirth C, Lepers R, Vallier JM, Brisswalter J. Effect of cycling cadence on subsequent 3 km running performance in well trained triathletes. Br J Sports Med 37: 154–158; discussion 159, 2003.
2. Billat VL, Mille-Hamard L, Petit B, Koralsztein JP. The role of cadence on the VO2 slow component in cycling and running in triathletes. Int J Sports Med 20: 429–437, 1999.
3. Borg GA. Perceived exertion: A note on "history" and methods. Med Sci Sports 5: 90–93, 1973.
4. Coyle EF, Feltner ME, Kautz SA, Hamilton MT, Montain SJ, Baylor AM, Abraham LD, Petrek GW. Physiological and biomechanical factors associated with elite endurance cycling performance. Med Sci Sports Exerc 23: 93–107, 1991.
5. Coyle EF, Sidossis LS, Horowitz JF, Beltz JD. Cycling efficiency is related to the percentage of type I muscle fibers. Med Sci Sports Exerc 24: 782–788, 1992.
6. Garcin M, Vautier JF, Vandewalle H, Wolff M, Monod H. Ratings of perceived exertion (RPE) during cycling exercises at constant power output. Ergonomics 41: 1500–1509, 1998.
7. Green JM, McLester JR, Crews TR, Wickwire PJ, Pritchett RC, Lomax RG. RPE association with lactate and heart rate during high-intensity interval cycling. Med Sci Sports Exerc 38: 167–172, 2006.
8. Hawley JA. Adaptations of skeletal muscle to prolonged, intense endurance training. Clin Exp Pharmacol Physiol 29: 218–222, 2002.
9. Hopker J, Jobson S, Carter H, Passfield L. Cycling efficiency in trained male and female cyclists. J Sports Sci Med 9: 332–337, 2010.
10. Hopker J, Coleman D, Passfield L. Changes in cycling efficiency during a competitive season. Med Sci Sports Exerc 41: 912–919, 2009.
11. Hopker J, Coleman D, Passfield L, Wiles J. The effect of training volume and intensity on competitive cyclists' efficiency. Appl Physiol Nutr Metab 35: 17–22, 2010.
12. Jackson AS, Pollock ML. Generalized equations for predicting body density of men. Br J Nutr 40: 497–504, 1978.
13. Kawaguchi K, Hayashi Y, Sekikawa K, Tabusadani M, Inamizu T, Onari K, Bhambhani Y. Vastus lateralis oxygenation during prolonged cycling in healthy males. Appl Physiol Nutr Metab 31: 48–55, 2006.
14. Lucia A, Hoyos J, Chicharro JL. Preferred pedalling cadence in professional cycling. Med Sci Sports Exerc 33: 1361–1366, 2001.
15. Lucia A, Hoyos J, Perez M, Santalla A, Chicharro JL. Inverse relationship between VO2max and economy/efficiency in world-class cyclists. Med Sci Sports Exerc 34: 2079–2084, 2002.
16. Lucia A, San Juan AF, Montilla M, CaNete S, Santalla A, Earnest C, Perez M. In professional road cyclists, low pedaling cadences are less efficient. Med Sci Sports Exerc 36: 1048–1054, 2004.
17. Marsh AP, Martin PE, Foley KO. Effect of cadence, cycling experience, and aerobic power on delta efficiency during cycling. Med Sci Sports Exerc 32: 1630–1634, 2000.
18. Mora-Rodriguez R, Aguado-Jimenez R. Performance at high pedaling cadences in well-trained cyclists. Med Sci Sports Exerc 38: 953–957, 2006.
19. Moseley L, Jeukendrup AE. The reliability of cycling efficiency. Med Sci Sports Exerc 33: 621–627, 2001.
20. Neary JP, McKenzie DC, Bhambhani YN. Effects of short-term endurance training on muscle deoxygenation trends using NIRS. Med Sci Sports Exerc 34: 1725–1732, 2002.
21. Nielsen JS, Hansen EA, Sjogaard G. Pedalling rate affects endurance performance during high-intensity cycling. Eur J Appl Physiol 92: 114–120, 2004.
22. Noble B, Robertson R. Percieved Exertion. Champaign, IL: Human Kinetics, 1996.
23. Santalla A, Naranjo J, Terrados N. Muscle efficiency improves over time in world-class cyclists. Med Sci Sports Exerc 41: 1096–1101, 2009.
24. Siri WE. The gross composition of the body. Adv Biol Med Phys 4: 239–280, 1956.
25. Takaishi T, Ishida K, Katayama K, Yamazaki K, Yamamoto T, Moritani T. Effect of cycling experience and pedal cadence on the near-infrared spectroscopy parameters. Med Sci Sports Exerc 34: 2062–2071, 2002.
26. Westgarth-Taylor C, Hawley JA, Rickard S, Myburgh KH, Noakes TD, Dennis SC. Metabolic and performance adaptations to interval training in endurance-trained cyclists. Eur J Appl Physiol Occup Physiol 75: 298–304, 1997.
27. Widrick JJ, Freedson PS, Hamill J. Effect of internal work on the calculation of optimal pedaling rates. Med Sci Sports Exerc 24: 376–382, 1992.

sports performance; pedaling rate; oxygen dissociation; economy; tissue oxygen saturation

© 2013 National Strength and Conditioning Association