In this article I will try to provide an overview of the specific physiological characteristics of successful rowers. Since data describing elite masters age rowers 35-80 years old really do not exist, (though I now have data on about 1,000 masters rowers that will answer some questions), I will focus on available data from young, elite men and women. I think it is a fair assumption that older successful rowers differ quantitatively but not qualitatively from their young counterparts. I will draw much of the information below from 3 excellent resources. Drs. Fred Hagerman of Ohio University in the United States, Dr. Neils Secher of Copenhagen, Denmark, and Dr. J.M. Steinacker of Ulm, Germany. All three physiologists have tested and examined large numbers of elite level rowers in their respective countries and published excellent reviews of their findings. In addition, I have recently published a 30 year retrospective study of the training and performance characteristics of international medal winning rowers from Norway. This work is based on data collected over 3 decades by national team coach and talent development coordinator Åke Fiskerstrand.
To the uninitiated, a casual view of a rowing race might suggest that rowing is primarily an upper body sport. The error of this impression is well known by rowers. The rowing stroke takes advantage of a freely sliding seat so that the drive of the oar is sequentially aided by forceful extension of the legs, extension of the trunk, and a contribution of the arms that is minor in absolute force contribution but critical in technical importance. Since the boat is accelerated as it moves in reaction to the sweeping arc of the oar, acceleration will be proportional to force times time. Therefore the rower must achieve an optimal combination of high stroke power and long stroke length. This combined high force and long impulse duration requirement tends to select for rowers of a specific size and length.
Compared to athletes in other endurance disciplines, successful rowers are as lean, but heavier and taller, with long arms, and a tall sitting height. Hagerman has collected data on more than 3000 elite U.S. rowers since 1964. Data from female rowers has been collected since the late 70s. Heavyweight oarsman averaged 6 feet 3.5 inches tall, and 194 pounds (1.92 m and 88kg). Their female counterparts averaged 5 feet 11″ and 169 pounds (1.8 m, 77 kg). More recently, the average height and weight of the mens’ and womens’ U.S. 1992 Olympic team is reported in the table below (from Hagerman). As a reminder,1kg = 2.2 pounds. 1inch = 2.54 centimeters.
|N||Age||HT cm||WT kg||Body Fat %|
By comparing the 30 year data with the table from the 1992 Olympic Team, it is appears that the height and weight of elite rowers has changed little in the last 3 decades. However, bodyfat % has decreased over three decades, so that lean body mass is higher today than in 1964.
It is also worth mentioning that the very best oarsmen, winners of international medals, tend to be slightly heavier then this average, generally over 200 pounds (91 kg). The apparent advantage afforded the larger athlete in rowing has led to the inception of a weight restricted “lightweight” class of male and female rowers. Light weight men are restricted to a bodyweight of 72.5 kg while lightweight women must not exceed 59 kg. Data from U.S National team candidate testing suggests that lightweight males average 6 feet tall (1.84m) while lightweight females are about 5’7″ (1.7m). Body fat percentage in the lightweight categories is not surprisingly even lower than that observed in heavyweights. Elite lightweight males average 5-7% body fat while elite females are under 15%.
In the Table below, I have reproduced data presented by Dr. Fred Hagerman on the characteristics of the 1992 U.S: Olympic Team. This will serve as a basis for further discussions on rowing physiology. The data represent mean values for the group. So, there is a distribution above and below this mean.
I think the female values have probably improved in recent years, in parallel to the much improved performance of U.S. women’s teams at Worlds. In fact, the current 2k erg standard for elite level competitiveness is now 6:50 for women, about 10 seconds faster than in 1992. For the men, the power output in the table above corresponds to an average time of about 6:02 for 2000 meters. The current competitive standard in for elite rowers is now set at “under 6:00”. So, the standards have gotten tougher in the last 10 years, particularly on the women’s side. The current world (2003) and Olympic (2004) champion in the single, Norwegian Olaf Tufte, demonstrates perhaps the gold standard for physiological capacity in rowers. His VO2 max is 7.1 liters during running and 6.8 on the rowing ergometer. He has rowed 1:15 for 500 meters, 5:46 for 2,000 meters, and a stunning 15:32 for 5,000 meters (1:32/500m avg.). He weighs about 94kg. This combination of exceptional aerobic and anaerobic capacity seems unique even among his world class peers.
I do not have comparable physiological data for lightweights, but I discuss their performance data in “Ergometer Analysis”.
Even without adjustment for the developments over the last deade, the data in the table above show values that are among the highest reported among endurance athletes. These values represent the average of 25 and 35 athletes, not all of whom won international medals. The VERY best males (in the lab) have achieved values of 7.0 liter/min at max. Let me tell you, that is an extraordinary absolute V02 value! The very best females are at 5 liters/min, also extraordinary. This is not terribly surprising since rowers are very large for endurance athletes, and oxygen consumption increases with body size. However, when maximal oxygen consumption of rowers is scaled linearly with bodyweight, the values are less impressive. While 71 ml/min/kg is quite a “respectable” value (Average males of the same age are at 45 ml/min/kg), it is far from the 80-87 ml/min/kg values that currently typify the world elite cross country skiers and runners. The best female cross country skiers are over 70 ml/min/kg compared to about 60 for the female rowers. Are rowers undertrained, or undertalented? One problem with this comparison is a matter of scaling. Maximal oxygen consumption does not increase linearly with increased body mass (Click here for more on this). So, dividing VO2 by bodyweight is not really appropriate. Without going into details here, it is more appropriate to scale VO2 max to bodyweight^2/3. In the table below, I do this and contrast the data with 1) untrained males of normal weight, as well as 2) at the weight of elite rowers , and ) elite cross country skiers. This will give you some idea of where rowers stand relative to the two extremes. Well I suppose being normal is not really an extreme, but you know what I mean.
|Average untrained U.S. male||72||3.25||45||187.5|
|Big, untrained male||93||3.91||42||190|
|92 Olympic Rowers males, 35||88.1||6.25||70.9||315|
|Top5 5 male rowers, U.S. (est.)||95||6.8||71.6||326|
|Top 5 male Rowers world (est.)||95||7.0||73.7||335|
|Top 5 World Cross Country Skiers||75||6.5||86.7||365|
Elite Rowers have a maximal aerobic capacity about 1.75 times higher than same-age untrained males. However, compared to world class skiers, the best rowers are about 8-10% lower in maximal aerobic capacity, even after accounting for bodyweight differences with allometric scaling. (See Far right Column in table above). This is based on available physiological data from around the world. The reasons for this difference are unclear. From a strict probability standpoint, we could argue that the subset of candidates for elite rowing that meet the “size requirements” for success from which performers are ultimately pulled is smaller than the pool from which skiers (and runners) are drawn. These sports have less restrictive size demands, biomechanically. So, maybe the ultimate rowing athlete has not yet been discovered! Considering how good elite rowers are right now, I don’t think I want to face him when and if he is!
Chapter 3: Skeletal Muscle Characteristics
As observed repeatedly in runners, cross country skiers and cyclists, successful rowers are characterized by an above average percentage of type I (slow) fibers in their leg musculature. Several studies suggest that the Type I percentage is about 70% compared to 40-50% in the population at large. Furthermore, even among rowers, fiber type composition seems to be a discriminating variable. The more successful rowers have an even higher Type I composition. In internationally successful rowers, the percentage has been measured as high as 85% . The remainder are almost all type IIa fibers with almost no IIb fibers present. In general, it is concluded that the presence of substantial percentages of IIb (fast, low mitochondria) fibers in endurance athletes is indicative of either insufficient training years, or inadequate intensity of training. Some older studies on national team females indicated more IIb fibers compared to elite males. However, the intensification of women’s training programs in the last 5-10 years has probably narrowed or eliminated this discrepancy. With intense endurance training over a period of years, the fast IIb fiber subtype appears to convert to the fast IIa subtype that is characterized by greater fatigue resistance.
In trained rowers, the density of mitochondria is high, expressed as the ratio of mitochondria to fiber areas. This adaptation is evident in both ST and FT fibers. Measurements of oxidative enzyme activity reveal that successful rowers demonstrate the expected high levels of these enzymes in their rowing muscles. In contrast, glycolytic activity (a factor in anaerobic capacity) evaluated as the activity of the enzyme lactate dehydrogenase, is not different among groups of oarsmen. However, the better oarsmen possess a greater percentage of the “heart” subtype (LDH isoforms LDH1-3) which has a lower affinity for pyruvic acid. In addition, muscle capillary density is twice as high in successful rowers as untrained. All of these characteristics contribute to a high work capacity and a reduced rate of lactate production at high workloads. The increased capillary density enhances the rate of lactate removal from the active muscle.
During my “rowing career”, I have heard comments from presumably expert coaches that suggest that fast twitch fibers are important for “explosive leg drive” or “fast hands”. I have tried to keep a straight face during these conversations, but this is simply wrong!. Even at high stroke rates, the contraction time of the rowing muscles is sufficiently long to allow slow twitch fibers to generate maximum force. Therefore, there is no advantage to possessing a high percentage of fast twitch fibers. To the contrary, they are conspicuously absent in the most successful rowers. I frequently blame my failure to row faster on my substantial endowment of fast fibers (verified by biopsy).
The above characteristics are completely consistent with the expected metabolic profile of endurance trained muscle. However, rowers show abnormally large cross-sectional areas of individual muscle fibers, both fast and slow, when compared with the same fiber types in other endurance athletes. This is at odds with the general pattern of endurance adaptation (small muscle cell diameters mean reduced oxygen diffusion distances). Closer analysis of the task demands of rowing may help explain this difference. The stroke frequency of competitive rowing is quite low when compared to the contraction frequency employed in cycling or running. In contrast, the peak muscular force is substantially higher. The rower must adopt a pattern of work output that relies on relatively few periods of high force production with longer “rest” intervals between contractions. This pattern of activity is consistent with the development of larger muscle fibers, in appropriate response to the task demands. The extremes of aerobic capacity and muscle power necessary to ensure success in rowing are probably influenced by both genetic inheritance, and intense and specialized training.
Rowers tend to be stronger than other endurance athletes based on typical strength tests such as leg extensions. However, this greater strength often is associated with their inherently greater size and muscle mass. Quoting Dr. Fred Hagerman, “This increased strength should, in no way, be interpreted as translating into greater rowing power.” Dr. Secher says the same thing, “Oarsmen are strong, reflecting their large body dimensions, but their muscle strength is not correlated in any simple way to their rowing strength.” This premise is supported by several studies that indicate that strength data do not correlate well with rowing ergometer performance.
Only when a simulated rowing position is used does the strength of the best oarsmen distinguish itself from less qualified oarsmen. This supports the concept that even simple strength measurements are significantly dependent on skill. Secher has performed studies which suggest that oarsmen are unique in their ability to develop force with both legs simultaneously. This is a unique movement pattern in endurance sport. In untrained subjects and subjects trained in other disciplines, 2-legged strength is approximately 80% of the sum of the strength of the left and right legs measured seperately. This gap decreases in rowers due to their specific training.
I have gone back to the data I have on 500m, 2000m and 6000m ergometer performances from US national team candidates. The 500 meter is the closest performance measure we have that reflects anaerobic capacity. Even at this short distance, aerobic metabolism contributes significantly, but the 500m is a reasonable test length at any rate. Now what would the absolute best measure of “maximum functional strength” be for the rower? I think it would be the maximum power developed in the first five strokes from a dead stop in the boat. But, since that’s not practical, we go to the erg. When I did just this type of test on collegiate rowers and compared it with power output maintained for 45 seconds, the correlation was about .90, which is very high. So, one maximal stroke is enough to predict performance reasonably well in a 250 meter sprint. This makes sense. Strength and anaerobic capacity are both dependent on muscle mass. I accept that 250 meters and 500 meters on the erg are both reasonable measures of anaerobic capacity, so I hope you do. Now here is the important question. Muscle strength is strongly related to anaerobic capacity (500 meter time). But, is 500 meter performance time strongly correlated to 2k performance time? The answer is YES and NO. Yes, they are related if you take a range of people from untrained to elite oarsmen, or combine lightweight and heavy weight men and women into one very heterogeneous group. But, NO they are not related when you look within a specific group of well trained rowers. When I determined the relationship between power output/kg for 500 meters and 2000 meters among 25 heavyweight men, the correlation was a weak 0.50. In the top 10 heavyweight women it was 0.07 or basically zero! Among the men, 500 meter power varied by 30 percent, while 2k power only varied by 10 percent.
Recently (2005), I applied the same approach to the online CII ergometer rankings. These data have become available since I first wrote this article. I took the top 10% of the performers age 20-40 in the 500m sprint. I then found those who had also performed 2k and 5k races. I then converted performance times to power output in watts using the same equation used by Concept II. The correlation between 500meter power output and 2000meter output was about 0,4. This means that 500meter power output only explained about 16% of the variation in 2k performance. However, the correlation between 2k and 5k power output was about 0.9, meaning that 80% of 2k performance was explained by variation in 5k performance. These data actually support nicely the known physiology of the 2k race. About 85% of the energy requirement during the race is supplied aerobically, while the remainder is supplied via anaerobic pathways. For those who are interested, these data also suggest that 5000meter power should average about 80-85% of 2,000m power. Among the good rowers I analyzed, this ratio ranged from 77% to about 90%. It is interesting to note in this context that the US National team no longer performs strength or sprint tests, only the 2k and 6k ergos.
To conclude this section, I will mention two studies performed by Hagerman and colleagues. The first, published in 1983, compared off-season and in season physical characteristics in 9 members of the men’s US Olympic team. Ergometer performance power increased 14% from OFF season to IN season. Conversely, leg strength was significantly higher during the OFF season. This dissociation between maximal strength and peak rowing performance was more recently supported by a study completed in 1993, but as yet unpublished. This study compared physiological and performance variables in a group of rowers who performed weight training plus aerobic conditioning during the off-season to a group that did only aerobic conditioning. Dr. Hagerman concluded from the results that “not only does supplemental weight training fail to improve physiological and competitive performance, but more importantly it appears that weight training may actually detract from these performances.”
In elite level rowers, both men and women, it appears that there is an optimal level of muscular strength associated with success. Strength training is probably important for many athletes to achieve this optimum. However, there is no evidence that greater and greater strength and muscle mass gains result in faster performances on the erg or on the water, at least not over the 2k distance or longer. For very short sprints, the impact of increased muscle mass on anaerobic capacity may contribute to better performances.
Ventilation and Rowing performance (examined specifically within the larger general article on ventilation)