The Effect of Concurrent Plyometric Training Versus Submaximal Aerobic Cycling on Rowing Economy, Peak Power, and Performance in Male High School Rowers

Participants

Intervention

The endurance group performed 30 min of steady-state cycling at ventilatory threshold at the same time as the experimental group. The “Talk Test” (cycling at an intensity at which a conversation was possible) was used to prescribe cycling intensity and has been shown to be a valid approximation of ventilatory threshold [27]. The athletes were also familiar with this method, as it was used at their club to prescribe low-intensity aerobic workouts. Cycling was used so that one group was not provided with a higher volume of rowing-specific training than the other and is also a common cross-training method for rowers. The endurance group was also supervised. After each group completed the 30 min of training, they proceeded to their on-the-water rowing practice. All subjects completed the same on-the-water workouts, and members from the PLYO and E groups rowed together in mixed boats. As all rowers were on the same team, outside of the training intervention (endurance or plyometric), rowing volume and intensity were identical for each member of the study.

Measurements

All measurements were taken during a single testing session on the Saturdays before and after the 4 weeks of training. Testing occurred over a 90-min period for each participant. All post-testing was performed at least 48 h since the last plyometric or biking session, allowing adequate recovery and avoiding undue influence of the last training session.

Upon arrival, participants had their stature (seca 217, UK), mass (Belfour Inc., WI, USA), and thigh and calf circumferences measured (Gulick tape measure). These measurements were taken by the same member of the research team for pre- and post-testing and were taken half way up the thigh and calf, as has been done in previous plyometric studies [28]. Rowing economy (RE) was assessed with a submaximal 8-min step test on a rowing ergometer (Model D, Concept2, VT, USA), which consisted of four 2-min stages at 90, 120, 150, and 180 W. The Concept2 rowing ergometer is considered to be an accurate rowing ergometer [29] and is the gold standard. Expired gases were collected using a one-way non-rebreathe mouthpiece to determine oxygen consumption (VO₂), a known measure of energy expenditure (TrueOne 2400, Parvo Medics, Sandy, UT, USA). The accuracy of the O₂ analyzer is 0.1%, the flowmeter and analyzers were calibrated to less than 1% variance, and typical repeatability (coefficient of variation) in our lab for VO₂ is ~5%. VO₂ was obtained during the last 15 s of the 2-min stage, and based upon previous testing and pilot work, we determined participants were able to achieve steady state within the 2-min stages. Rating of perceived exertion (RPE) was measured using the Borgs Scale (RPE 0–10) [30]. Participants completed a 2-min warm-up and cooldown before and after the test, respectively.

After the RE measurements, participants rested for 30 min and then performed a maximal 500-m TT on the rowing ergometer. All participants were familiar with performing maximal 500-m trials as these were performed frequently as part of their normal training and/or performance assessments, prior to enrollment in the study, but none were performed during the intervention period. Stroke rate was not capped for the participants. Time, stroke rate, and average watts were measured during the TT. A 2-min warm-up and cooldown was completed before and after the TT, with participants asked to rest for another 30 min following the TT. Peak rowing power (RP) was then measured over three maximal trials of 15 s on the rowing ergometer. The recorded value for peak power was the highest wattage observed during the 15-s period. For both the RE and TT, the participants were instructed that they could self-select the ergometer resistances, which had to be the same for both the pre- and post-testing. However, as each rower was on the same team, they selected the same ergometer resistances (control 4.4 ± 0.2 vs. plyometrics 4.5 ± 0.0, arbitrary units, p > 0.05) which are also in accordance with US National Team testing guidelines. For the RP test, the resistance was set to the highest value (10) on the rowing ergometer.

Statistical Analysis

All statistics were performed using commercially available software (SPSS v. 23, Armonk, NY). Pre-planned t test comparisons were used to determine significance at baseline, in changes within the PLYO and E groups, and at post-test for stature, mass, circumferences, TT performance, and RP. In the case of violating normality, a non-parametric alternative method was employed. In addition, the magnitude of the change from pre-training to post-training was also determined using standardized differences in means (i.e., effect size, ES, Cohen’s d) [31, 32]. Threshold values were established as small (0.2), moderate (0.6), large (1.2), and very large (2.0). A mixed-model analysis of variance (ANOVA) was used to determine differences in RE. Linear regression analysis was used to determine the slope of the line for VO₂ across work rate and the intercept for each individual. The level of significance was established at p ≤ 0.05. To understand the potential relationships between rowing power, rowing economy, and performance, Pearson correlation coefficients were calculated. Data are reported as means ± standard deviation.

Participant Characteristics

During the training period, one subject from the PLYO and one subject from the E group dropped out due to injuries, which were incurred during their on-the-water rowing training and were unrelated to the intervention. Acceptable adherence to the training program was set at 85%, and all remaining participants achieved this or higher. There were no differences in age, stature, mass, or thigh and calf circumferences between or within the groups at baseline prior to training (p > 0.05) (Table 1). There no were significant changes in any of the anthropometric measures within either group (p > 0.05); as such, at post-testing, there were no differences between the groups for the stature, mass, and thigh and calf circumferences (p > 0.05).

Rowing Economy

Rowing economy (Fig. 1) did not differ significantly between PLYO and E at all of the work rates before training. In response to training, the analysis of variance indicated no significant interaction effects for VO₂ values between the two groups at all of the work rates (90, 120, 150, 180 W, Fig. 1, p > 0.05) but a main effect for time was found to be significant (p < 0.05) indicating a generally lower VO₂ or energy cost from pre- to post-intervention. Additionally, linear regression analysis indicated no group differences in the slope (VO₂/watt relationship) at baseline, in response to training, or at post-intervention in either group (PLYO: p = 0.17, p = 0.18; E: p = 0.2, p = 0.18). RPE was also taken in the last minute of each stage, and there was a significant reduction in RPE following training in the PLYO group after training (p = 0.05), but not for the CON group (Fig. 2).

500-m Time Trial Performance

There was no difference in the 500-m times between the groups before training (p = 0.32). TT performance did not change for E after training (p = 0.83, ES = −0.08); however, the TT performance was significantly improved in PLYO (Fig. 3, p  < 0.05, ES = −0.91). Furthermore, difference in TT performance was statistically different between the groups at post-testing. The average power output during the 500-m TT was not different between groups at baseline (363 ± 54 W vs. 372 ± 75 W, E vs. PLYO, respectively, p = 0.77), but PLYO significantly increased power output, and thus, the groups were significantly different post-training (368 ± 46 W vs. 414 ± 24 W, E vs. PLYO, respectively, p = 0.02, ES = −1.89). Finally, there were no group differences in the average stroke rate during the 500-m TT, at baseline (E 37 ± 4 strokes/min vs. PLYO 38 ± 3 strokes/min, p = 0.52), in response to training, or post-training (E 38 ± 4 strokes/min vs. PLYO 39 ± 5 strokes/min, p = 0.52).

Peak Rowing Power

There was no difference in RP between the groups before training (p = 0.69, ES = −0.07). After training, there was no change in RP within PLYO or E (Fig. 4; p = 0.18, ES = 0.14, p = 0.12, ES = 0.11). Although, there was a trend for a difference in RP between the groups following training (p = 0.08, ES = 0.66), with a tendency for greater power output in the PLYO group. RP was negatively correlated to VO₂ (energy expenditure) during work rates 90–180 W before training (r = −0.549, −0.879, −0.745, −0.736, p = 0.34, p = 0.0, p = 0.001, p = 0.002, for each work rate, respectively). In addition, post-training RP was negatively correlated to the VO₂/watt slope during the RE test (r = −0.576, p = 0.039). Finally, peak rowing power was negatively correlated with 500-m time performance before (r = −0.92, p = 0.0) and after training (r = −0.78, p = 0.0). Finally, the peak power responses were also highly reliable, and exploration of the three baseline trials indicated a significant reliability coefficient (Cronbach’s α = 0.99, p = 0.00).

Plyometric Training and Rowing Economy

In the current study, rowing economy across work rate was significantly improved in both groups (Fig. 1), possibly indicating that plyometric and aerobic training had a similar effect on rowing economy. Previously, running economy has been shown to improve after plyometric training [21, 24, 28], and the reason for this improvement has been suggested to be increased musculotendinous stiffness (MTS) in the legs of runners [24]. A high MTS increases an individual’s ability to absorb, harness, and recoil force during the stretch-shortening cycle, such as in running [33]. An increase in MTS is thus beneficial for a sport such as running, which involves an eccentric contraction when the runner’s foot makes contact with the ground, where after plyometric training, the increased MTS results in more stored energy in the series elastic elements and reduces eccentric muscle demand. However, the beginning of the rowing stroke (known as the catch), is an anticipated motion in which the athlete is actually pulling themselves towards their feet, followed by a rapid press of the legs to apply power [34]. As there is not so much of a reactive movement, or ground reaction force in rowing when compared with running, increased MTS is likely less beneficial in rowing. Although, plyometric training has also been shown to increase muscle activation [20] and muscle force [35] and decrease ground reaction times [36] suggesting plyometrics might alter the work to rest ratio of rowing, through improved drive speed, allowing more time of the stroke cycle to be spent in recovery for a given stroke rate. Such a change, perhaps independent of economy, might correspond to improved boat speed, allowing for greater maintenance of the forward propulsion or “run” of the rowing shell in the water from each stroke. It is worth noting that these hypotheses warrant further investigation to confirm if this is, in fact, the case.

Plyometric Training and Rowing Power

Plyometric training has been suggested to improve power [37] in adults and in youth [38]. Results from the current study indicate that plyometric training in this athlete cohort (youth rowers), for this length of time (4 weeks), may elicit a moderate increase in rowing power (Fig. 4). This is in agreement with previous research as plyometrics are typically used to improve power [2, 11, 12]. It has been suggested by rowing biomechanics experts that elasticity stored in the tendons of rowers could be beneficial or initiating the drive sequence of the stroke [34]. The fact that the moderate improvement in rowing power did not reach statistical significance may be attributed to a number of potential factors. The training period was relatively short compared to other studies in the field of plyometrics, so with further training, there could have been a greater and/or significant change in rowing power. The sample size was also relatively small, possibly lacking the statistical power necessary to observe an increase in rowing power. Changes in RP also varied within the groups, suggesting that some individuals may have been better at performing the 15-s peak power ergometer test or possess the innate physiological traits or training acumen to achieve high rowing powers. Also, the motor coordination trained using plyometrics (jumping) is vastly different to the patterns used during the maximal peak power test, with a high rowing resistance. Furthermore, muscle coordination could have differed greatly between individuals due to the athletes being at varying stages of athletic maturity. However, it should be noted that with a greater sample size, and/or a longer training period, a significant difference in peak rowing power could be observed.

The peak rowing power test is a valuable test and predictor of rowing performance, as suggested by Ingham et al. [4]. However, the athletes whom Ingham et al. used were all elite rowers (current or former World Championship Finalists), who have a significantly higher athletic maturity and training history [4] when compared with the athletes in the current study. Therefore, the younger age and relatively lower training history (i.e., coordination and technical mastery) of the athletes utilized in the current study could explain the moderate improvement in rowing power. Specifically, these athletes may not have completed the physical and technical development to truly achieve maximum power, thus perhaps underestimating the possible changes with the training. However, we did find a significant inverse correlation between rowing power and rowing economy, that is, a greater rowing power was associated with lower energy expenditure for a given power output. This is suggestive that rowers with greater power may be more technically masterful and thus expend less energy during rowing.

Plyometric Training and Rowing Sprint Performance

Plyometric training has been shown to improve short running sprint performance [39], and results from the current study indicate that plyometric training is able to improve rowing sprint performance (Fig. 3). These findings are in contrast to a study performed by Kramer et al. in [3], in which 20 min of simple double leg plyometric exercises were added at the end of the strength training sessions completed by female collegiate rowers. Their rowing measures were a 2.5-km ergometer test, and the distance in meters achieved in 90 s. As a comparison with the current investigation, the 90-s rowing test is very similar to the 500-m TT. The average time for the plyometric group to perform the 500-m TT after training was ~95 s. Kramer et al. found no changes in the 90-s test after 9 weeks of training, whereas we observed a significant improvement in 500-m performance [3]. There are several potential reasons for this discrepancy.

The plyometric exercises used in the earlier study of Kramer et al. [3] were more simplistic than in the current investigation (e.g., no rotation or multiple jumps or upper body involvement). Indeed, the authors did indicate that the exercises they selected may not have been the best for improving rowing performance [3] and were performed following strength training, likely in a semi-fatigued state, suboptimal conditions for plyometric training. Additionally, in comparison to the study by Kramer et al., the current study contained more foot contacts (Table 2.) and focused on plyometric training versus steady-state training, matched for training volume (time) between groups. Kramer et al. did not match training volume [3]. To avoid this, training times were matched for volume of the dry-land training in the present investigation. All on the water rowing training was also matched as all participants were members of the same team. Finally, Kramer et al. [3] recruited collegiate female rowers who were older than the high school male participants in the current study (mean age of 21 compared with a mean age of 16 years). As such, when prescribing plyometric training, it is important to consider factors such as performing these exercises in a non-fatigued state, foot contacts, and load in order to optimize any potential performance improvements.

Nonetheless, the current investigation was able to demonstrate that plyometric training improved 500-m rowing performance (Fig. 3). However, it remains to be seen if plyometric training is capable of improving rowing performance of a more gold standard distance, such as 2- or 6-km TT. As peak power was only moderately increased for the plyometric group, but rowing economy did not differ between pre- and post-training, the improvement in rowing performance could be due to increased rowing power but also to other mechanisms not measured in the current study (e.g., muscle coordination, activation, and rate of force development). Alternatively, the rating of perceived exertion, while not different prior to the intervention, was significantly reduced post-training in the PLYO group only (Fig. 2), which suggests a possible shift in maximal work rate, or the plyometric training may have increased the tolerance for work, thereby reducing perceived effort, and either phenomena might have contributed to the increased performance.

Experimental Considerations

The current study is not provided without limitation or further consideration. While the 500-m TT was shown to improve as a result of the plyometric training, the training program (three 30-min sessions/week) might be considered time consuming for coaches that are usually working under stringent time constraints. Although, the results from the current study indicate that plyometric training does not impair rowing economy; thus, if on-the-water or ergometer training is not available (e.g., weather or time constraints), plyometric training might serve as an effective substitute or adjunct.

As the purpose of the study was to highlight the effects of plyometric training, we cannot ascertain whether inclusion of strength training or strength training alone would provide additional or superior benefit. Plyometrics are often performed in conjunction with weight training (also known as complex training) and have been shown to cause greater improvements in power than either modality alone [40]. Whether or not complex training is beneficial for rowing performance has yet to be evaluated.

The 500-m TT measure used here is not the gold standard rowing performance test used by rowing coaches. The 2-km ergometer time trial is the most common rowing performance measure [1]. This test not only measures rowing ability, fitness, and technique but also mental fortitude. For this last reason, the 2-km test was not used to evaluate rowing performance of the young athletes in the current study. In young rowers, 2-km performances can vary greatly due to it being mentally challenging so the much shorter 500-m test was used to provide a more objective measure of whether the plyometric program improved rowing performance. Although it is worth mentioning that the reliability of the 500-m TT was not tested, the athletes were very accustomed to the test distance as it is included in training and assessment. Additionally, the athletes were not periodized for the 2-km test, which typically occurs in the spring season. As the 2-km test was not performed in the current study, further research is needed to definitively demonstrate whether improvements in 500-m performance due to plyometrics can be translated into the full 2-km test. The results of the current study can only be applied to youth males, and it remains to be seen if such training responses would be seen in older and/or more developed athletes. Although the 500-m test is not a gold standard measure for rowing, the fact that the athletes were able to significantly improve performance with plyometric training is important. First, because the athletes were able to improve their rowing sprint performance during a time in which no rowing sprint training was being undertaken, this might be a way to preserve anaerobic performance without detriment to aerobic performance (i.e., rowing economy). Second, coaches can also use the 500-m performance to estimate the 2000-m performance, as some coaches consider the wattage produced over 500 m to be 138% of 2000-m watts. Finally, more detailed exploration of the possible changes in muscle will provide greater insight into the training responses to plyometrics in rowers (e.g., hop test and vertical jump).