European Journal of Applied Physiology

, Volume 106, Issue 1, pp 149–157 | Cite as

Aerobic power and peak power of elite America’s Cup sailors

  • Vernon Neville
  • Matthew T. G. Pain
  • Jonathan P. Folland
Original Article


Big-boat yacht racing is one of the only able bodied sporting activities where standing arm-cranking (‘grinding’) is the primary physical activity. However, the physiological capabilities of elite sailors for standing arm-cranking have been largely unreported. The purpose of the study was to assess aerobic parameters, VO2peak and onset of blood lactate (OBLA), and anaerobic performance, torque–crank velocity and power–crank velocity relationships and therefore peak power (Pmax) and optimum crank-velocity (ωopt), of America’s Cup sailors during standing arm-cranking. Thirty-three elite professional sailors performed a step test to exhaustion, and a subset of ten grinders performed maximal 7 s isokinetic sprints at different crank velocities, using a standing arm-crank ergometer. VO2peak was 4.7 ± 0.5 L/min (range 3.6–5.5 L/min) at a power output of 332 ± 44 W (range 235–425 W). OBLA occurred at a power output of 202 ± 31 W (61% of Wmax) and VO2 of 3.3 ± 0.4 L/min (71% of VO2peak). The torque–crank velocity relationship was linear for all participants (r = 0.9 ± 0.1). Pmax was 1,420 ± 37 W (range 1,192–1,617 W), and ωopt was 125 ± 6 rpm. These data are among the highest upper-body anaerobic and aerobic power values reported. The unique nature of these athletes, with their high fat-free mass and specific selection and training for standing arm cranking, likely accounts for the high values. The influence of crank velocity on peak power implies that power production during on-board ‘grinding’ may be optimised through the use of appropriate gear-ratios and the development of efficient gear change mechanisms.


VO2max Oxygen uptake Sailing Grinding Arm cranking Hand cycling Torque–velocity Power–velocity OBLA 


Standing arm cranking (‘grinding’) is the main physical activity performed during professional big-boat yacht racing, including the America’s Cup. During America’s Cup yacht racing, all manoeuvres are powered manually without the assistance of stored energy (Neville et al. 2006). There are four grinding pedestals on International America’s Cup Class version 5 yachts, each manned by two athletes. The grinding cranks are manually driven by a cyclic upper body action whilst standing. This provides the mechanical power to turn the winches which in turn controls the sails and mast, and hence the manoeuvres of the boat (Whiting 2007). There are typically five or six designated grinders in the crew of 17, however, all athletes assist with grinding, with the exception of the helmsman. It has been suggested that America’s Cup grinding utilises both anaerobic and aerobic energy systems, depending on the intensity of racing (Bernardi et al. 2007), but the physiological characteristics of America’s Cup sailors are poorly understood and very little research has been published on standing arm cranking.

Arm cranking has received some scientific attention, due to its role in cardiovascular and injury rehabilitation (Carson 1989; Westhoff et al. 2008), as well as being an appropriate mode for assessing upper body athletes (Tesch 1983; Driss et al. 1998; Hubner-Wozniak et al. 2004; Pearson et al. 2007; Zagatto et al. 2008) and individuals with spinal cord injury or lower limb disability (Hicks et al. 2003; Goosey-Tolfrey et al. 2006; Valent et al. 2008). During upper body exercise, athletes trained for this type of work appear to be able to achieve a high proportion of their lower body VO2peak, with seated arm cranking values of ~4.1 L min−1 being reported for elite upper body trained athletes such as wrestlers, kayakers, rowers and swimmers (Secher et al. 1974; Tesch 1983; Horswill et al. 1992). To date the majority of arm cranking research has been performed seated, with restricted lower limb involvement. The physiological responses to standing arm cranking have not been widely documented, with only a few reports on aerobic power (Vokac et al. 1975; Bernardi et al. 2007) and peak power (Vandewalle et al. 1989; Hubner-Wozniak et al. 2004; Bouhlel et al. 2007; Pearson et al. 2007) present in the literature. Standing arm cranking appears to elicit a similar cardiorespiratory response to seated arm cranking, but with a higher peak work load evident (13%) (Vokac et al. 1975). This indicates that cranking is more efficient when standing than when sitting. A recent report of a relatively inexperienced America’s Cup team found an average VO2peak of 4.1 L min−1 (47 ml kg−1 min−1) (Bernardi et al. 2007).

The peak velocity of grinding during America’s Cup racing has been reported to be between 120 and 150 revolutions per minute (Bernardi et al. 2007), but the optimum velocity for power production and the nature of the torque–velocity and power–velocity relationships during standing arm cranking have not been determined. This may have an important bearing on the selection of gear ratios and the optimisation of power production during big-boat sailing. During elite sprint cycling a polynomial power–velocity relationship has been described (Martin et al. 1997; Dorel et al. 2005; Gardner et al. 2007), and contrary to the hyperbolic force–velocity relationship of isolated muscle (Wilkie 1949), the relationship between torque and velocity appears to be linear (Martin et al. 1997; Dorel et al. 2005; Gardner et al. 2007; Sprague et al. 2007). Similar results have been found during seated arm cranking (Vandewalle et al. 1989; Vanderthommen et al. 1997).

This study aimed to describe two important components of physical performance in elite America’s Cup sailors during standing arm cranking (‘grinding’). The first objective was to report key indices of aerobic endurance performance [aerobic power and the onset of blood lactate accumulation (OBLA)] and to determine any differences between crew positions. The second was to document the torque–crank velocity and power–crank velocity relationships, and thus determine the peak power and optimum crank velocity, of America’s Cup grinders.



Thirty-three elite professional male America’s Cup sailors participated in the study. Their physical characteristics are shown in Table 1. All athletes sailed for a team ranked in the top three during the 32nd Americas Cup. Their combined sailing experience included 8 Olympic medals and 98 America’s Cup campaigns. Informed consent was obtained from all athletes, and the study was approved by the University’s Ethical Advisory Committee.
Table 1

Anthropometric characteristics of America’s Cup sailors (mean ± SD)



Age (years)

Stature (m)

Body mass (kg)

Σ 7 skinfolds (mm)

Body fat (%)

Fat-free mass (kg)

Body surface area (m2)



36 ± 7

1.88 ± 0.05a

105 ± 6b

77 ± 15

13 ± 4

91 ± 5e

0.23 ± 0.01f



34 ± 6

1.83 ± 0.06

94 ± 9c

99 ± 16

17 ± 3

77 ± 6

0.22 ± 0.01



35 ± 6

1.79 ± 0.02

84 ± 5

67 ± 19d

11 ± 4d

74 ± 4

0.20 ± 0.01



34 ± 6

1.80 ± 0.08

82 ± 5

68 ± 14

12 ± 3

72 ± 6

0.20 ± 0.01



40 ± 7

1.84 ± 0.04

88 ± 4

78 ± 10

14 ± 2

75 ± 4

0.21 ± 0.01



36 ± 6

1.84 ± 0.06

92 ± 11

78 ± 18

14 ± 4

79 ± 9

0.22 ± 0.02










aGrinders taller than Bowmen (P = 0.01) and Trimmers (P = 0.04)

bGrinders heavier than all other positions (P < 0.01)

cUtilities heavier than Trimmers (P = 0.05)

dBowmen less than Utilities (P < 0.05)

eGrinders greater than all other positions (P < 0.001)

fGrinders greater than all other positions (P < 0.03)

Experimental design

All athletes visited the laboratory for an initial test where anthropometric measurements were taken prior to the athletes performing a step test to exhaustion to determine peak oxygen uptake (VO2peak) and lactate threshold. A sub-group of ten athletes (grinders) then returned to the laboratory 1 month later to perform four maximal 7 s isokinetic sprints, at different velocities of arm cranking in a randomised order, for the measurement of peak torque and power at each crank velocity.


All measurements were taken in accordance with the prescribed methods of the International Society for the Advancement of Kinanthropometry (Marfell-Jones et al. 2006). Nude body mass was measured with a digital scale to the nearest 0.1 kg (Tanita BWB-800, Tokyo, Japan) and height was measure with a stadiometer to the nearest 0.005 m. Skinfold thickness was measured in duplicate at seven sites (biceps, triceps, subscapular, supraspinale, abdomen, thigh, calf) using Harpenden skinfold calipers (Baty International, West Sussex, UK). Body fat % was calculated from the sum of seven skinfolds (Siri 1961; Jackson and Pollock 1978). Body surface area (BSA) was calculated using the Mosteller formula (Mosteller 1987):
$$ {\text{BSA}} = \surd \left( {{{\left( {{\text{height}} \times {\text{body mass}}} \right)} \mathord{\left/ {\vphantom {{\left( {{\text{height}} \times {\text{body mass}}} \right)} { 3 , 600}}} \right. \kern-\nulldelimiterspace} { 3 , 600}}} \right). $$


All tests were performed on an adjustable SRM electronically braked scientific ergometer (Schoberer Rad Meßtechnik Scientific, Jülich, Germany), which was specifically modified for standing arm cranking (Fig. 1). The centre of the ergometer handles were 0.44 m apart (medio-lateral displacement), and the crank arm length was kept constant at 0.25 m for all measurements. Torque was recorded continuously at 200 Hz (SRM torque software) and averaged over 360°. Crank velocity was measured every revolution. The SRM Powercrank was calibrated daily according to the manufacturer’s guidelines. Pulmonary gas exchange was measured breath-by-breath, using an automated on-line gas analysis system (Oxycon Pro, Jaeger, Hoechberg, Germany). The athletes wore a nose clip and breathed through a sealed low-resistance mouthpiece and impeller turbine digital sensor (TripleV, Jaeger) that measured inspired and expired gas volumes, and was connected to the analysis system via a capillary line. The gas was analysed for O2 and CO2 concentrations using paramagnetic and infrared analysers, respectively. The analysers were calibrated automatically before each test with gases of known concentration and the volume sensor was calibrated using a 3-L syringe. The on-line values were calculated by the Jaeger computer software (IntelliSupport). Heart rate (HR) was monitored every 5 s with a telemetric HR monitor (Polar S720, Finland).
Fig. 1

Image of the modified SRM arm ergometer used for America’s Cup grinding performance analysis. Crank length was 0.25 m and the distance between the handles was 0.44 m

Aerobic power

Each athlete was able to self-select the height of the arm ergometer axis, which was typically ~50% of stature. A step protocol (Washburn and Seals 1983; Smith et al. 2004) was adopted with up to eight, 4 min stages each consisting of 3 min of constant work, followed by a 30 s rest interval and a 30 s ramp up to the next step. The initial power output was 75 W which was increased by 40, 45 or 50 W at each step, based on the athletes previous response to a laboratory aerobic power test, and with the aim of reaching exhaustion after six or seven steps. The athletes were required to maintain a constant velocity of 80 rpm (Smith et al. 2001), and the test was terminated when the athlete was no longer able to maintain a velocity above 75 rpm despite verbal encouragement. Earlobe blood samples (~25 μL) were taken during each 30 s rest interval and immediately analysed for the lactate concentration using an automated blood lactate analyser, YSI 2300 Stat (Yellow Springs Instruments Inc., OH, USA). A further blood sample was analysed 3 min post exercise. The rating of perceived exertion (Borg scale) was also recorded at the end of each step. The cardiorespiratory variables were averaged over the final 30 s of each step, and the highest 30 s average oxygen uptake was taken to be the VO2peak. An absolute blood lactate (BLa) concentration of 4.0 mmol L−1, also referred to as the ‘OBLA’ (Sjodin and Jacobs 1981) was used to determine lactate threshold (Heck et al. 1985). HR, VO2 and work load at OBLA were calculated using linear interpolation between the relevant data points.

Peak power

The height of the arm ergometer’s central axis was set at 0.9 m from the ground. The ergometer’s isokinetic mode (constant velocity) was adopted for the sprints. After an initial 5 min self-paced warm-up, athletes performed four maximum effort isokinetic sprints at different crank velocities in a randomised order, from a range of six crank velocities: 80, 90, 100, 110, 120 and 140 rpm. Each sprint was 7 s in duration with a 10 min rest interval between trials. A 10 s countdown was given to the start of the sprint during which time a velocity of 50 rpm was maintained with no resistance. Verbal encouragement was given throughout the test. Torque and angular velocity were recorded throughout each sprint and analysed off-line. Once the prescribed velocity for each sprint was achieved, the highest torque and power values (over 360°) were calculated.

Power was determined as the product of torque (T) and crank velocity (ω) expressed in radians s−1. Linear regression of the torque–crank velocity relationship was used to determine maximal torque (Tmax) and maximal crank velocity (ωmax) by extrapolation to the respective intercepts. For each individual, maximal power (Pmax) was determined as the apex of the power–crank velocity relationship, and optimal crank velocity (ωopt), the crank rate at which Pmax occurred.

Statistical analysis

Anthropometric, aerobic power and OBLA data for the different crew positions were compared with one-way ANOVA. When significant main effects were found, a Bonferroni post hoc test was used to determine differences between positions. Pearson product–moment correlation coefficients were calculated to assess bivariate relationships. Statistical analyses were performed with SPSS version 14.0 for Windows. Statistical significance was set at P ≤ 0.05, and data are presented as mean ± SD.


The anthropometric characteristics of the sailors are shown in Table 1. Grinders were heavier than all other positions (P < 0.01), while bowmen had a lower body fat percentage than utilities (P = 0.05).

For all the athletes VO2peak was 4.69 ± 0.50 L min−1 (range 3.58–5.48 L min−1), which occurred after 26 min 29 s ± 2 min 27 s of the step test at a power output of 332 ± 44 W. Grinders achieving a higher power output than the afterguard (369 ± 35 vs. 297 ± 50 W, P = 0.01) and bowmen had a higher relative VO2peak than grinders (56 ± 6 vs. 48 ± 4 ml kg−1 min−1, P = 0.04; Table 2). Figure 2 shows the BLa response to increasing VO2 as a percentage of VO2peak.
Table 2

Maximal physiological responses of America’s Cup sailors during a standing arm cranking (‘grinding’) step test (mean ± SD)



HR max (beats min−1)

VO2peak (L min−1)

VO2peak (ml kg−1 min−1)

VE (L min−1)

Peak power (W)

Peak [BLa] (mmol L−1)



186 ± 8

5.04 ± 0.41

48 ± 4

199 ± 22

369 ± 35b

11.3 ± 1.3



191 ± 8

4.74 ± 0.35

51 ± 4

184 ± 20

332 ± 19

11.7 ± 1.2



187 ± 7

4.63 ± 0.39

56 ± 6a

189 ± 25

328 ± 35

12.0 ± 2.4



190 ± 9

4.50 ± 0.50

55 ± 5

186 ± 15

311 ± 45

13.2 ± 2.0



181 ± 10

4.18 ± 0.66

49 ± 6

179 ± 10

297 ± 50

10.6 ± 2.4



187 ± 9

4.69 ± 0.50

51 ± 6

189 ± 20

332 ± 44

11.7 ± 1.9









aBowmen greater than Grinders (P = 0.04)

bGrinders greater than Afterguard (P = 0.01)

Fig. 2

Blood lactate concentration (BLa) with increasing VO2 as a percentage of VO2peak during each stage of an incremental step testing to exhaustion on a standing arm-crank ergometer (n = 33). Data are mean ± SD

For the whole crew the OBLA occurred at a power output of 202 ± 31 W (61 ± 6% of Wmax) and VO2 of 3.34 ± 0.04 L min−1 (71 ± 5% of VO2peak), with grinders having greater power output than trimmers at OBLA (227 ± 7 vs. 177 ± 28 W, P = 0.01; Table 3).
Table 3

Physiological responses at lactate threshold of America’s Cup athletes, during standing arm cranking (‘grinding’) (mean ± SD)



Power (W)

HR (beats min−1)

VO2 (L min−1)

VO2 (ml kg−1 min−1)

VO2 (% of VO2peak)



229 ± 21a

159 ± 10

3.67 ± 0.33a

34.7 ± 3.6

73 ± 4



197 ± 23

168 ± 9

3.34 ± 0.33

35.9 ± 2.4

71 ± 6



198 ± 22

163 ± 5

3.26 ± 0.33

39.2 ± 2.7

70 ± 6



177 ± 28

155 ± 5

3.10 ± 0.56

37.7 ± 6.2

68 ± 6



192 ± 36

158 ± 10

3.13 ± 0.46

35.3 ± 4.9

72 ± 4



202 ± 30

161 ± 9

3.34 ± 0.43

36.3 ± 4.2

71 ± 5








aGrinders greater than Trimmers (P = 0.01)

For each of the grinders the relationship between torque and crank velocity during the isokinetic sprints was linear (r = 0.94 ± 0.06). The torque–crank velocity relationship for this cohort of grinders was expressed by the following equation: T = − 0.8421ω + 211.68 (r = −0.99, P < 0.001), with predicted Tmax, 212 Nm and ωmax, 251 rpm (Fig. 3).
Fig. 3

Torque and crank velocity relationship during standing arm cranking (‘grinding’) (r = −0.99, P < 0.01, n = 10)

Peak power during the 7 s sprints is displayed in Fig. 4. Peak power at 80 rpm was significantly less (17%) than at 120 rpm (P = 0.03, Fig. 4a), as was relative peak power (P = 0.01, Fig. 4b). No significant differences in peak power were found between 80 and 110 or 140 rpm. Peak power was significantly correlated to body mass (r = 0.58, P = 0.04). The power–crank velocity relationship was a parabola described by the equation: P = − 0.1206ω2 + 29.201ω − 361.73 (r = 0.73). Pmax, the apex of the power–crank velocity relationship, was 1,420 ± 37 W (range 1,192–1,617 W), and when normalised for body mass was 13.7 ± 1.0 W kg−1 (range 12.0–15.4 W kg−1). ωopt occurred at 125 ± 6 rpm (range 114–133 rpm).
Fig. 4

Relationship between a Peak Power and Crank Velocity during standing arm cranking (‘grinding’) (r = 0.73, n = 10). Asterisk 120 significantly greater than 80 rpm (P = 0.03); b Relative peak power and crank velocity (r = 0.81, n = 10). Asterisk 120 significantly greater than 80 rpm (P = 0.01)


The main findings of this study were that America’s Cup sailors are characterised as having high absolute upper body aerobic and anaerobic power. To the best of our knowledge no other cohort of athletes has achieved an average VO2peak for arm cranking of 4.7 L min−1 (51 ml kg−1 min−1) that we have found for a whole crew of America’s Cup sailors, or the Pmax of 1,420 W (13.7 W kg−1) for a sub-group of grinders. America’s Cup grinding is unique in that it is one of the only able bodied sporting activities where arm cranking is the primary physical activity.

The Pmax produced by grinders in this study was substantially more than that previously reported during arm cranking. To our knowledge, no other study has reported a cohort of athletes to have Pmax greater than 1,000 W during arm cranking. A recent study also on America’s Cup grinders performing standing arm cranking, found Pmax of 929 ± 100 W (Pearson et al. 2007). Other studies of standing arm cranking have reported Pmax values of 720–732 W (8.5–9.6 W kg−1) in javelin throwers and elite wrestlers (Hubner-Wozniak et al. 2004; Bouhlel et al. 2007) and 700 W (10.6 W kg−1) during seated arm cranking in elite gymnasts (Jemni et al. 2006). In addition, the peak aerobic power values observed during this study are the highest reported during arm cranking exercise. Slightly lower results have been reported during seated arm cranking by elite rowers (4.4 L min−1, Secher et al. 1974) and elite sprint kayakers (4.3 L min−1, Tesch 1983), both activities requiring substantial upper body aerobic power. The impressive peak power and aerobic power values in this study are characteristic of the unique requirements of this cohort of athletes that are selected and trained for the specific activity of standing arm cranking. The large body size (and fat-free mass) of these elite sailors would certainly be expected to contribute to the high absolute values recorded. In fact the grinders, who recorded the highest absolute aerobic power (5.0 L min−1) and produced Pmax values more than 40% above any previously documented, had a significantly greater fat free mass (91 kg) than all other crew positions. However, when normalising for body mass, bowmen had the greatest relative VO2peak (56 ml kg−1 min−1). Similar results have been reported previously with grinders, 4.7 L min−1 and bowmen, 52 ml kg−1 min−1 (Bernardi et al. 2007). This disparity between positions is likely due to differences in body mass between roles, with grinders ~20 kg heavier than bowmen as a result of their high power requirements. In addition, the activities performed by bowmen are typically more prolonged with short recovery periods, and thus more continuous and aerobic in nature than the activities of grinders. The high aerobic power seen in all crew positions, indicates the importance of upper body aerobic power for this sport. Although the activity profile of most crew roles is considered to be intermittent with the most intense periods during manoeuvres, the prolonged nature of racing (~100 min), the high number of work bouts and the relatively short rest intervals suggests a heavy reliance on aerobic energy metabolism (Gaitanos et al. 1993) that may explain the high aerobic power values of these athletes.

Compared to seated arm cranking, where the involvement of the lower limbs is restricted, standing arm cranking has a greater contribution from the proximal kinetic chain in force production, hence, one might expect a higher performance during standing than seated arm cranking. Although it has previously been reported that the maximal cardiorespiratory response is unaffected by the type of arm cranking (Vokac et al. 1975). Other factors likely to influence performance during standing arm cranking are crank length, ergometer height and the distance between the crank handles, which are seldom reported in the literature. In this study, these were all similar to that typically employed on America’s Cup racing yachts, but the greater crank length and handle separation compared to standard arm ergometers, likely facilitates a greater range of movement and involvement of a larger muscle mass.

A discontinuous step test protocol was selected for the current study in order to determine both aerobic power and the OBLA. Although it has been suggested that maximal arm crank tests of aerobic power should be short in duration (<14 min) in order to avoid local fatigue (Goosey-Tolfrey et al. 2006), this has not been confirmed, and no significant differences have been observed in any measured variables during arm cranking between step and ramp protocols despite a threefold difference in test duration (Washburn and Seals 1983; Smith et al. 2004).

The OBLA has repeatedly been found to be an important determinant of endurance performance (Sjodin and Jacobs 1981; Mujika and Padilla 2001). The mean VO2 at OBLA of the sailors in the present study was 71% of VO2peak which is less than that typically reported in elite cyclists (~86%, Mujika and Padilla 2001) and well trained kayakers (81%, van Someren and Oliver 2002). The nature of grinding during America’s Cup yacht racing is highly intermittent and characterised by short bouts of maximal effort interspersed by longer rest intervals (Bernardi et al. 2007; Neville 2008). This intermittent activity profile may explain the lower OBLA of grinders compared to the continuous activities of cycling and flat water kayaking. Unfortunately there is almost no comparable data for lactate threshold during arm cranking. One observation of elite paraplegic wheelchair athletes reported as high as 75% of maximum power and 80% of HRmax at OBLA during seated arm cranking (Schmid et al. 1998), however the numerous physiological differences between able bodied and disabled athletes, makes it difficult to compare these findings.

It is generally acknowledged that substantial differences exist between arm cranking and lower body exercise, such as cycling. For example, the maximal aerobic capacity of the upper body in untrained subjects seems to be limited by peripheral factors, including a small involved muscle mass, a greater proportion of type II fibres (Johnson et al. 1973), a low density of capillaries (Turner et al. 1997; Calbet et al. 2005) and mitochondria (Turner et al. 1997), and a greater peripheral resistance (Stenberg et al. 1967). In addition, arm cranking exercise results in greater physiological stress (RPE and HR) at the same VO2 (Leicht et al. 2008) when compared with cycling in untrained subjects. Despite these physiologic and energetic disadvantages the current study demonstrates the exercise capacity of trained elite upper body athletes.

The torque–crank velocity and power–crank velocity relationships of elite upper body trained athletes have received little attention, with the few reports in the literature indicating a linear relationship between torque and velocity (Vandewalle et al. 1989; Vanderthommen et al. 1997; Driss et al. 1998), similar to that seen in cycling. The parabolic power–crank velocity relationship seen in this study emphasises the influence of crank rate on peak power. The optimum velocity for maximum power output was 125 rpm, which is surprisingly similar to the ωopt during elite track cycling (~129 rpm, Dorel et al. 2005; Gardner et al. 2007). The relatively short upper limbs and greater crank length of grinding compared to cycling (250 vs. 170 mm, Dorel et al. 2005) would be expected to lead to a greater joint excursion during each revolution with grinding. Hence for a given crank velocity, considerably greater joint angular velocities would be expected with grinding compared to cycling. It is surprising therefore that ωopt of the two activities appears so similar, and may be indicative of a greater proportion of type II fibres in the upper body musculature of elite grinders compared to the leg musculature of cyclists. There is some evidence that the upper body muscles have a greater proportion of type II fibres compared to the lower body (Johnson et al. 1973).

The exercise testing mode in this study was isokinetic. To our knowledge the magnitude and nature of the grinding resistance encountered on-board has not been documented, although it seems likely to be isoinertial. Nevertheless, isokinetic ergometry has been shown to provide valid measurement of instantaneous and mean power (Sargeant et al. 1981) and is regarded as an appropriate method for determining optimal crank velocity. In contrast an isoinertial modality involves both velocity and power changing simultaneously and thus a more complex experimental situation. An important recommendation of this study is that future work measure the power produced during on-board grinding to define the mechanical parameters involved.

The only documented evidence of crank velocity during America’s Cup racing reported peak velocity of between 120 and 150 rpm (Bernardi et al. 2007). From the results of the current study it seems that a narrow range of crank velocities between 115 and 135 rpm would be beneficial for optimising power production. This has considerable implications for the design of winch gear ratios and gear selection during big-boat sailing. The on-board winch systems typically have up to eight gears, which are usually changed by either stopping the crank rotation for the newly selected gear to engage, or by changing the direction of crank rotation (i.e. grinding backwards). Both of these gear changing techniques result in a loss of momentum and a velocity substantially outside of the optimum range whilst the grinders are striving to exert maximum power during a manoeuvre. In addition, grinding backwards elicits substantially less power (~17%) than grinding forwards (Pearson et al. 2007). Therefore, it would be highly beneficial if it were mechanically possible to maintain momentum (within the optimum crank velocity range) in the forward direction during gear change; i.e. to change gear while grinding forwards without stopping, such as the use of a ‘crash-box’ gear change system.

Taken together the results of this study underscore the unique nature of this cohort of elite athletes who have high levels of both anaerobic and aerobic upper body power. This poses a challenge to the conditioning of these athletes as both explosive power and endurance are required. Future research may look to investigate the influence of specific training interventions on upper body power and endurance. In addition, research should aim to investigate the physiological demands during competition, particularly the determination of the actual power output of grinders during racing.


The high Pmax with concomitant high VO2peak suggests that America’s Cup grinders require substantial upper body anaerobic power in addition to high aerobic power. The elite nature of these athletes, their high fat-free mass, training and selection for standing arm cranking, as well as the mechanics of the ‘grinding’ ergometer used, may have contributed to their high values. In addition, the influence of crank velocity on peak power implies that power production during on-board ‘grinding’ could be optimised through the use of appropriate gear-ratios and development of efficient gear change mechanisms.



We are grateful to Dr. Cristina Pérez Encinas and Dr. Luis del Moral Garcia for their assistance with the data collection, and to Dr. Massimo Massarini for his contributions to the study protocol.

Conflict of interest statement



  1. Bernardi M, Quattrini FM, Rodio A et al (2007) Physiological characteristics of America’s Cup sailors. J Sports Sci 25:1141–1152. doi:10.1080/02640410701287172 PubMedCrossRefGoogle Scholar
  2. Bouhlel E, Chelly MS, Tabka Z et al (2007) Relationships between maximal anaerobic power of the arms and legs and javelin performance. J Sports Med Phys Fitness 47:141–146PubMedGoogle Scholar
  3. Calbet JA, Holmberg HC, Rosdahl H et al (2005) Why do arms extract less oxygen than legs during exercise? Am J Physiol 289:R1448–R1458. doi:10.1152/ajpcell.00313.2005 CrossRefGoogle Scholar
  4. Carson WG Jr (1989) Rehabilitation of the throwing shoulder. Clin Sports Med 8:657–689PubMedGoogle Scholar
  5. Dorel S, Hautier CA, Rambaud O et al (2005) Torque and power–velocity relationships in cycling: relevance to track sprint performance in world-class cyclists. Int J Sports Med 26:739–746. doi:10.1055/s-2004-830493 PubMedCrossRefGoogle Scholar
  6. Driss T, Vandewalle H, Monod H (1998) Maximal power and force-velocity relationships during cycling and cranking exercises in volleyball players. Correlation with the vertical jump test. J Sports Med Phys Fitness 38:286–293PubMedGoogle Scholar
  7. Gaitanos GC, Williams C, Boobis LH et al (1993) Human muscle metabolism during intermittent maximal exercise. J Appl Physiol 75:712–719PubMedGoogle Scholar
  8. Gardner AS, Martin JC, Martin DT et al (2007) Maximal torque- and power–pedaling rate relationships for elite sprint cyclists in laboratory and field tests. Eur J Appl Physiol 101:287–292. doi:10.1007/s00421-007-0498-4 PubMedCrossRefGoogle Scholar
  9. Goosey-Tolfrey V, Castle P, Webborn N et al (2006) Aerobic capacity and peak power output of elite quadriplegic games players. Br J Sports Med 40:684–687. doi:10.1136/bjsm.2006.026815 PubMedCrossRefGoogle Scholar
  10. Heck H, Mader A, Hess G et al (1985) Justification of the 4-mmol/l lactate threshold. Int J Sports Med 6:117–130. doi:10.1055/s-2008-1025824 PubMedCrossRefGoogle Scholar
  11. Hicks AL, Martin KA, Ditor DS et al (2003) Long-term exercise training in persons with spinal cord injury: effects on strength, arm ergometry performance and psychological well-being. Spinal Cord 41:34–43. doi:10.1038/ PubMedCrossRefGoogle Scholar
  12. Horswill CA, Miller JE, Scott JR et al (1992) Anaerobic and aerobic power in arms and legs of elite senior wrestlers. Int J Sports Med 13:558–561. doi:10.1055/s-2007-1024564 PubMedCrossRefGoogle Scholar
  13. Hubner-Wozniak E, Kosmol A, Lutoslawska G et al (2004) Anaerobic performance of arms and legs in male and female free style wrestlers. J Sci Med Sport/Sports Med Australia 7:473–480Google Scholar
  14. Jackson AS, Pollock ML (1978) Generalized equations for predicting body density of men. Br J Nutr 40:497–504. doi:10.1079/BJN19780152 PubMedCrossRefGoogle Scholar
  15. Jemni M, Sands WA, Friemel F et al (2006) Any effect of gymnastics training on upper-body and lower-body aerobic and power components in national and international male gymnasts? J Strength Cond Res/Natl Strength Cond Assoc 20:899–907Google Scholar
  16. Johnson MA, Polgar J, Weightman D et al (1973) Data on the distribution of fibre types in thirty-six human muscles. An autopsy study. J Neurol Sci 18:111–129. doi:10.1016/0022-510X(73)90023-3 PubMedCrossRefGoogle Scholar
  17. Leicht AS, Sinclair WH, Spinks WL (2008) Effect of exercise mode on heart rate variability during steady state exercise. Eur J Appl Physiol 102:195–204PubMedCrossRefGoogle Scholar
  18. Marfell-Jones M, Olds T, Stewart A et al (2006) International standards for anthropometric assessment, vol 1. Potchefstroom, South AfricaGoogle Scholar
  19. Martin JC, Wagner BM, Coyle EF (1997) Inertial-load method determines maximal cycling power in a single exercise bout. Med Sci Sports Exerc 29:1505–1512. doi:10.1097/00005768-199711000-00018 PubMedGoogle Scholar
  20. Mosteller RD (1987) Simplified calculation of body-surface area. N Engl J Med 317:1098PubMedGoogle Scholar
  21. Mujika I, Padilla S (2001) Physiological and performance characteristics of male professional road cyclists. Sports medicine (Auckland, NZ) 31:479–487Google Scholar
  22. Neville V (2008) America’s Cup Yacht racing is not just about the boat. Sport Exerc Sci March:26–27Google Scholar
  23. Neville VJ, Molloy J, Brooks JH et al. (2006) Epidemiology of injuries and illnesses in America’s Cup yacht racing. Br J Sports Med 40:304–311 (discussion 311–302). doi:10.1136/bjsm.2005.021477 Google Scholar
  24. Pearson S, Hume P, Slyfield D et al (2007) External work and peak power are reliable measures of ergometer grinding performance when tested under load, deck heel, and grinding direction conditions. Sports Biomech/Int Soc Biomech Sports 6:71–80CrossRefGoogle Scholar
  25. Sargeant AJ, Hoinville E, Young A (1981) Maximum leg force and power output during short-term dynamic exercise. J Appl Physiol 51:1175–1182PubMedGoogle Scholar
  26. Schmid A, Huonker M, Aramendi JF et al (1998) Heart rate deflection compared to 4 mmol x l(-1) lactate threshold during incremental exercise and to lactate during steady-state exercise on an arm-cranking ergometer in paraplegic athletes. Eur J Appl Physiol Occup Physiol 78:177–182. doi:10.1007/s004210050404 PubMedCrossRefGoogle Scholar
  27. Secher NH, Ruberg-Larsen N, Binkhorst RA et al (1974) Maximal oxygen uptake during arm cranking and combined arm plus leg exercise. J Appl Physiol 36:515–518PubMedGoogle Scholar
  28. Siri WE (1961) Body composition from fluid space and density. In: Brozek J, Hanschel A (eds) Techniques for measuring body composition. National Academy of Science, Washington, DC, pp 223–244Google Scholar
  29. Sjodin B, Jacobs I (1981) Onset of blood lactate accumulation and marathon running performance. Int J Sports Med 2:23–26. doi:10.1055/s-2008-1034579 PubMedCrossRefGoogle Scholar
  30. Smith PM, Price MJ, Doherty M (2001) The influence of crank rate on peak oxygen consumption during arm crank ergometry. J Sports Sci 19:955–960. doi:10.1080/026404101317108453 PubMedCrossRefGoogle Scholar
  31. Smith PM, Doherty M, Drake D et al (2004) The influence of step and ramp type protocols on the attainment of peak physiological responses during arm crank ergometry. Int J Sports Med 25:616–621. doi:10.1055/s-2004-817880 PubMedCrossRefGoogle Scholar
  32. Sprague RCt, Martin JC, Davidson CJ et al (2007) Force-velocity and power–velocity relationships during maximal short-term rowing ergometry. Med Sci Sports Exerc 39:358–364. doi:10.1249/01.mss.0000241653.37876.73 PubMedCrossRefGoogle Scholar
  33. Stenberg J, Astrand PO, Ekblom B et al (1967) Hemodynamic response to work with different muscle groups, sitting and supine. J Appl Physiol 22:61–70PubMedGoogle Scholar
  34. Tesch PA (1983) Physiological characteristics of elite kayak paddlers. Can J Appl Sport Sci 8:87–91PubMedGoogle Scholar
  35. Turner DL, Hoppeler H, Claassen H et al (1997) Effects of endurance training on oxidative capacity and structural composition of human arm and leg muscles. Acta Physiol Scand 161:459–464. doi:10.1046/j.1365-201X.1997.00246.x PubMedCrossRefGoogle Scholar
  36. Valent LJ, Dallmeijer AJ, Houdijk H et al (2008) Influence of hand cycling on physical capacity in the rehabilitation of persons with a spinal cord injury: a longitudinal cohort study. Arch Phys Med Rehabil 89:1016–1022. doi:10.1016/j.apmr.2007.10.034 PubMedCrossRefGoogle Scholar
  37. van Someren KA, Oliver JE (2002) The efficacy of ergometry determined heart rates for flatwater kayak training. Int J Sports Med 23:28–32. doi:10.1055/s-2002-19268 PubMedCrossRefGoogle Scholar
  38. Vanderthommen M, Francaux M, Johnson D et al (1997) Measurement of the power output during the acceleration phase of all-out arm cranking exercise. Int J Sports Med 18:600–606. doi:10.1055/s-2007-972688 PubMedCrossRefGoogle Scholar
  39. Vandewalle H, Peres G, Sourabie B et al (1989) Force-velocity relationship and maximal anaerobic power during cranking exercise in young swimmers. Int J Sports Med 10:439–445. doi:10.1055/s-2007-1024940 PubMedCrossRefGoogle Scholar
  40. Vokac Z, Bell H, Bautz-Holter E et al (1975) Oxygen uptake/heart rate relationship in leg and arm exercise, sitting and standing. J Appl Physiol 39:54–59PubMedGoogle Scholar
  41. Washburn RA, Seals DR (1983) Comparison of peak oxygen uptake in arm cranking. Eur J Appl Physiol Occup Physiol 51:3–6. doi:10.1007/BF00952531 PubMedCrossRefGoogle Scholar
  42. Westhoff TH, Schmidt S, Gross V et al (2008) The cardiovascular effects of upper-limb aerobic exercise in hypertensive patients. J Hypertens 26:1336–1342. doi:10.1097/HJH.0b013e3282ffac13 PubMedCrossRefGoogle Scholar
  43. Whiting P (2007) The 32nd America’s Cup: a simple guide. Wiley, ChichesterGoogle Scholar
  44. Wilkie DR (1949) The relation between force and velocity in human muscle. J Physiol 110:249–280PubMedGoogle Scholar
  45. Zagatto AM, Papoti M, Gobatto CA (2008) Anaerobic capacity may not be determined by critical power model in elite table tennis players. J Sports Sci Med 7:54–59Google Scholar

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  • Vernon Neville
    • 1
  • Matthew T. G. Pain
    • 1
  • Jonathan P. Folland
    • 1
  1. 1.School of Sport and Exercise SciencesLoughborough UniversityLoughboroughUK

Personalised recommendations