Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio

Abstract

The purpose of this study was to compare the training intensity distribution (TID) of the undefeated world champion male rowing New Zealand (kiwi) pair over a four-year Olympic cycle, across training phases, training years, and between individuals. Training data, including heart rate and boat speed, were recorded in the athletes rowing in the same boat between March 2013 and August 2016, ending with the Rio Olympics final. Progressive exercise tests assessed first (LT1) and second (LT2) lactate thresholds and associated heart rates, to determine the percentage of training performed below, between and above these demarcation points. Training an average of only 12-15 h/wk throughout the Olympic cycle, the mean percent distribution of time (±SD) at each training intensity was 80.4 ± 5.5% <LT1, 17.9 ± 5.3% LT1-LT2 and 1.8 ± 0.8% >LT2 for Rower A and 67.3 ± 9.0% <LT1, 30.2 ± 9.4% LT1 - LT2, and 2.4 ± 1.4% >LT2 for Rower B. Across the years 2014-2016, Rower A performed most likely more training <LT1, while Rower B performed mostly likely more training between LT1-LT2. Training appeared to become more polarised, with greater amounts of time spent <LT1, with increased training duration (R=0.38-0.43). Two of the world's best rowers, rowing together in the same boat with an undefeated record across an Olympic cycle, travelled markedly different "roads to Rio" within the context of their TID, with one rower displaying a polarised model of TID, and the other pyramidal. However, TID trended towards becoming more polarised in both rowers with increased training duration.

Full text

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Note. This article will be published in a forthcoming issue of the
International Journal of Sports Physiology and Performance. The
article appears here in its accepted, peer-reviewed form, as it was
provided by the submitting author. It has not been copyedited,
proofread, or formatted by the publisher.
Section: Original Investigation
Article Title: Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion
Rowers: Different Roads Lead to Rio
Authors: Daniel J. Plews1, and Paul B. Laursen1
Affiliations: 1Sport Performance Research Institute New Zealand (SPRINZ), Auckland
University of Technology, Auckland, New Zealand
Journal: International Journal of Sports Physiology and Performance
Acceptance Date: September 7, 2017
©2017 Human Kinetics, Inc.
DOI: https://doi.org/10.1123/ijspp.2017-0343

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Article Type: Original report
Title: Training intensity distribution over a four-year cycle in Olympic champion rowers: different roads
lead to Rio
Author: Daniel J. Plews1, and Paul B. Laursen1
Affiliations:
1. Sport Performance Research Institute New Zealand (SPRINZ), Auckland University
of Technology, Auckland, New Zealand
Contact Information:
Daniel Plews
SPRINZ, AUT University
Millennium Institute of Sport & Health,
17 Antares Place,
Mairangi Bay, 0632, New Zealand
Ph: +64 21 900 694
Corresponding Author: Daniel Plews:
plews@plewsandprof.com
Abstract: 250
Main text: 2800
Figures and tables: 3 Figures, 2 Tables
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Abstract
The purpose of this study was to compare the training intensity distribution (TID) of the
undefeated world champion male rowing New Zealand (kiwi) pair over a four-year Olympic
cycle, across training phases, training years, and between individuals. Training data, including
heart rate and boat speed, were recorded in the athletes rowing in the same boat between March
2013 and August 2016, ending with the Rio Olympics final. Progressive exercise tests assessed
first (LT1) and second (LT2) lactate thresholds and associated heart rates, to determine the
percentage of training performed below, between and above these demarcation points. Training
an average of only 12-15 h/wk throughout the Olympic cycle, the mean percent distribution of
time (±SD) at each training intensity was 80.4 ± 5.5% <LT1, 17.9 ± 5.3% LT1-LT2 and 1.8 ±
0.8% >LT2 for Rower A and 67.3 ± 9.0% <LT1, 30.2 ± 9.4% LT1 - LT2, and 2.4 ± 1.4% >LT2
for Rower B. Across the years 2014-2016, Rower A performed most likely more training <LT1,
while Rower B performed mostly likely more training between LT1-LT2. Training appeared to
become more polarised, with greater amounts of time spent <LT1, with increased training
duration (R=0.38-0.43). Two of the world’s best rowers, rowing together in the same boat with
an undefeated record across an Olympic cycle, travelled markedly different “roads to Rio”
within the context of their TID, with one rower displaying a polarised model of TID, and the
other pyramidal. However, TID trended towards becoming more polarised in both rowers with
increased training duration.
Keywords: polarised training, threshold training, high intensity, elite, rowing
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Introduction
Those of us fortunate to be immersed in the world of high performance sport will be
familiar with “rumor-mill” conversations had by coaches, athletes and support staff concerning
the types of training performed by the very best athlete’s in the world, and how such training
is perceived to have inevitably lead to their success.
In the context of striving to maximize performance for an endurance event, one area
that often gains much interest is that of training intensity distribution (TID).1,2 Indeed, various
models of TID have been discussed in the literature, including the polarised model3, the
threshold2 model and more recently the pyramidal model4. The polarised model was first
described within the training performed by the East German system from 1970-80, whereby a
high volume of low-intensity training appeared balanced against regular application of high-
intensity training bouts (~90% to >100% VO2max). This was partially confirmed in 2004 by
Fiskerstrand & Seiler 3, who showed a "polarized" pattern of training also when they described
the training and performance characteristics of 28 international Norwegian rowers developing
across the years 1970-2001. This polarised model is said to represent by 75-80% at a low
intensity (<2 mM blood lactate), 5% at threshold intensity (~4 mM blood lactate), and 15-20%
at high intensity (>4 mM blood lactate)4. This training organization contrasts the classic
threshold model (~57% low intensity, 43% threshold, 0% high-intensity5) of endurance
training, whereby large volumes of mid-zone threshold work is thought to be optimal2. This
former study on world class international rowers provided evidence to support its importance
as a model for endurance athletes striving to be the best in the world, and subsequently has
been largely adopted by many endurance athletes around the world 6,7. More recently, a number
of other retrospective studies have offered another model of TID in cycling,8 running,9 and
triathlon.10 Within this “pyramidal” model, most training carried out low intensity, with
decreasing proportions of threshold and high-intensity training performed. Exact percentage
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
definitions have yet to be truly established, however the pyramidal model general includes
~20% and 5% of TID at threshold and high intensity training levels, respectively.1
We had the opportunity to measure and report on the training organization of a highly
successful men’s rowing pair, double Olympic gold medallists (London 2012 and Rio 2016)
and multiple world champions (2009, 2010, 2011, 2013, 2014, 2015), over their final four-year
reign as champions. Analysis includes comparison across training phases, training years, and
between individual rowers. Herein, we define the “polarized model” as a training intensity
distribution (TID) profile whereby 75% of the total training is performed at a low intensity
(below first lactate threshold) with 10-15% performed at high intensities (above second lactate
threshold). We define the “threshold model” as a training profile whereby most training (>40%)
is performed between the first and second lactate thresholds.11 Finally we define the “pyramidal
model” as a training profile whereby <75% of the training is performed at a low intensity, and
>20% is performed at a threshold intensity.4
Methods
Two elite rowers of Rowing New Zealand agreed formally to act as participants for this
case report by providing written informed consent. The human research ethics committee of
AUT University approved the study.
Training data was recorded in the athletes rowing in the same boat between March 2013
and August 2016, ending with the Rio Olympics final. Upon commencement of the data
collection period, the pair were current Olympic champions and unbeaten in their boat class
since June 2009. Training year (not a calendar year) commenced at around 6 (± 2) weeks post
world championships (after the end of year break). This year then ran to the world rowing
championships event the following year, with the exception 2013, when the athletes took an
extended break after the 2012 Olympics.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Intensity zone determination
At the start of each training macro-cycle (~September), both athletes underwent
progressive exercise tests to assess first (LT1) and second (LT2) lactate thresholds. The
incremental exercise test was performed according to the Australian rowing physiological
assessment guidelines,12 of which both athletes were well accustomed to. Here, starting power
output and step increments were chosen based on each athlete’s personal best time recorded
over 2000m performed on the Concept II ergometer. The drag factor on the rowing ergometer
was adjusted to match the heavyweight male standard (130 units), as per Rowing New Zealand
standards. Subjects performed 7 x 4-min incremental steps, with the last step concluding as an
all-out effort. Rowers were asked to maintain their target power output during each step of the
test. Passive rest of 30 s followed all stages, during which an earlobe capillary blood sample
was collected to determine blood lactate concentration (Lactate Pro, Arkray, Japan). Both LT1
and LT2 were determined via the modified D-max method, as calculated via a software
package.13
These data were then used to identify training zones for the remaining year cycle. Such
one-off testing methods have been previously shown to adequately prescribe and monitor elite
endurance athletes across a full year.14
Training data analysis
All training sessions (rowing and cross-training) were recorded via a heart rate monitor
(Garmin 910xt, Garmin USA), downloaded (Ant Agent) and uploaded to a software platform
for analysis (www.trainingpeaks.com). Training intensity was estimated based on the heart
rates (HR) at LT1 and LT2 determined during the progressive exercise test. Total average time
recorded in each training zone was automatically calculated using the function available in
Trainingpeaks.com, which determined the time (hours: min: sec) and percentage of total time
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
in each zone. Training variables were broken down into the total training time, time spent at a
HR below LT1 (<LT1), time spent in the HR zone between LT1 and LT2 (LT1-LT2) and time
spent at a HR above LT2 (>LT2). Training time spent below 100 beats.min-1 was excluded
from analysis, and each training period was broken down into monthly totals, and averaged
accordingly. The maintenance of all HR monitoring equipment was performed regularly by the
resident Rowing team physiologist (DJP), with all data regularly inspected for erroneous
recordings. Small portions of erroneous data within a session were visually removed using the
Trainingpeaks.com software (“cut” function), while sessions involving larger portions of
corrupt HR data (>75%) were removed from the analysis.
All training load values (training stress score (TSS)) were calculated via the HR training
stress score method using Trainingpeaks.com.
Training
Training was categorised into on-water rowing, cycling, and rowing ergometer training,
which were the main modalities of training conducted by the rowers over the four years.
Different forms of training modalities (e.g. hiking, swimming, running) were classified as
“other”. This rowing pair did not perform strength training as part of their preparation. Each
annual period ended with a world rowing championship event (Chung-ju, Korea 2013;
Amsterdam, Netherlands 2014; and Aiguebellette, France 2015), which was won by the pair
each time. Their final successful event was the Rio Olympics (Brazil) in 2016. Throughout the
period, training was generally split into 4-week mesocycles, consisting of 3 heavy weeks and
1 recovery week.
Statistical analysis
Individual training data are presented as means (monthly or yearly) and 90% confidence
limits (CL) unless otherwise stated. Magnitude-based inferences were used to establish
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
individual between- and within-year (first training mesocycle to last training mesocycle)
differences in TID (i.e. time spent <LT1, time spent between LT1-LT2 and time <LT2).
Between-athlete differences were also analyzed using the same technique.15 The following
threshold values for ES statistics used were ≤0.2 (trivial), >0.2 (small), >0.6 (moderate), >1.2
(large), and >2.0 (very large). The trivial threshold for ES was set at 0.2. Quantitative chances
of spending more or less % of the time at a training intensity were also evaluated qualitatively
as follows: 25-75% possibly, 75-95% likely, 95-99% very likely, >99% almost certain. If the
chance of higher or lower differences was >5%, then the true difference was assessed as
unclear.
Pearson’s correlation was used to establish the relationship between the percentage of
time spent at various training intensities and total training hours. The magnitude of correlation
(r (90% CI)) between variables was assessed with the following thresholds: <0.1, trivial; <0.1-
0.3, small; <0.3-0.5, moderate; <0.5-0.7, large; <0.7-0.9, very large; and <0.9-1.0, almost
perfect. If 90% CI overlapped small positive and negative values, the magnitude of correlation
was deemed ‘unclear.’15
Results
The individual characteristics, including progressive exercise test results of each rower,
are shown in Table 1. Table 2 shows the percentage of total training time for each modality
and annual combined TSS across the four years of monitoring.
Overall between-season comparisons
The year-by-year TID data for each rower is shown in Figure 1. Overall, the average
percent distribution of time (±SD) at each training intensity over the 4-year period was 80.4 ±
5.5% <LT1, 17.9 ± 5.3% LT1-LT2 and 1.8 ± 0.8% >LT2 for Rower A and 67.3 ± 9.0%/30.2 ±
9.4/2.4 ± 1.4% for Rower B, for each respective zone.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Rower A between-season comparisons
For time spent below LT1 there were likely increases in the proportion of time spent
<LT1 from seasons 2015-2016 (ES 90% CI = 0.61 ±055) and likely decreases in time spent
>LT1 for 2014-2015 (-0.48 ±0.58).
For time spent between LT1 and LT2 there were likely decreases (-0.58 ±0.55) from
seasons 2015-2016.
For time spent > LT2, there were likely decreases (-0.48 ±0.58) in time between seasons
2014-2015. The remaining between-season comparisons were deemed unclear.
Rower B between-season comparisons
For time spent below LT1 there were very likely decreases ((-0.94 ±0.65) and increases
(0.63 ±0.41) in the proportion of time spent <LT1 from seasons 2013 to 2014 and 2015 to 2016.
For time spent between LT1-LT2, there were very likely increases (0.98 ±0.65) and
decreases (-0.66 ±0.39) between seasons 2013-2014 and 2015-2016.
Finally, for time spent >LT2 there was likely decreased (-0.61 ±0.86) time spent at this
intensity from seasons 2013-2014. The remaining between season comparisons were deemed
unclear.
Within-season year comparisons
The within-season differences in TID, comparing the first to last training mesocycle for
each year throughout the Olympic cycle are shown in Table 3.
Between-athlete comparisons
Between athlete effect size differences in the proportion of time spent at the specified
training intensity zones are shown in Figure 2.
For LT1, Rower A spent most likely a larger proportion of time at this intensity than
Rower B during 2014, 2015 and 2016 seasons.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
For between LT1 and LT2, Rower A spent most likely less time in this zone than Rower
B during these same years (2014, 2015 and 2016 seasons)
Finally for time >LT2, Rower A spent proportionally likely less time in this zone than
Rower B across years 2013 and 2014 and very likely less training time in this zone during 2016.
Training intensity relationships
Relationships between total training time and the distribution of time spent below LT1
are shown in Figure 3. Moderate relationships between these variables were shown for both
Rower A (R = 0.43 (0.17; 0.63)) and Rower B (R = 0.38 (0.11; 0.6). The relationship between
total training time and all other training intensities (LT1-LT2 and >LT2) was unclear.
Discussion
Within the context of elite sport, the road to Rio followed by world and Olympic
champions is of particular interest to those of us fascinated by the world of elite sport
performance and sports science. The present case report describes the heart rate TID in some
of the most dominant rowers of the modern era rowing together in the same boat. The main
finding was that individually, within the context of their TID, the pair trained markedly
different to one another, with one rower displaying a more polarized profile, and the other
displaying a more pyramidal TID, clearly spending more time training at higher intensities.
Other novel findings in the present study were the low relative weekly training hours completed
by the pair on an annual basis, and the increased quantity of <LT1 zone work that occurred as
training duration increased.
The pyramidal distribution displayed by Rower B included a high volume of low
intensity training, a medium volume of threshold training, and smaller amounts of high-
intensity training1. However, we cannot overlook the associated problems with using heart rate
as method for describing TID. In our case, HR was used to distinguish time spent at different
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
intensities, which has been often shown to underestimate the training time spent at high-
intensity due to the time lag in HR response.16 By the same notion, the “time in zone” method
may also overestimate time spent <LT1. In the study by Sylta and colleagues16, the authors
showed that a time in zone TID assessment was different to a “session goal” method for
establishing TID demarcation. The TID came out at 96% below LT1, 3% between LT1 and LT2
and 1% above LT2 for the time in zone analysis, and 86%, 11% and 3%, respectively, for the
session goal analysis method. Given that Rower B showed an average TID of roughly
67/30/2%, one could argue that his TID may have been closer to a threshold model than our
analysis suggests. This would certainly be more the case in the years 2014 and 2015, before he
shifted to more of a pyramidal model in 2016. Nevertheless, in working with athletes in the
field, heart rate monitoring is currently the most practical method available for describing the
TID of such high performing athletes. Nevertheless, this comparison shows the importance of
including other measures of assessment, such as session RPE and session goal, to provide a
greater representation of TID.16
Another likely reason for the TID differences shown may be due to differences in cross-
training modality of choice for each rower. As shown in Table 2, the training modalities in
2013 were quite similar in the rowers compared with other years. This may be due to a late
start to the training year after the 2012 Olympic break, alongside the need for greater specificity
in their training (hence more time rowing together). However, in 2014 and 2015, for
motivational reasons, the rowers spent more time training separately, with Rower A spending
more time cycling, and Rower B spending more time on the Concept Rowing Ergometer. Given
that the duration of exercise is typically longer and heart rates lower during cycling as opposed
to rowing,17 this could explain some of the TID differences observed between the two rowers.
Finally, for the Olympic year of 2016, Rower A incurred an injury earlier in the training season,
and spent more time cycling as part of his rehabilitation program (30% vs 11% for Rower A
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
vs. B, respectively). Subsequently, Rower A spent more time rowing on the water in the single
scull (53% vs 67%), and on the Concept Ergometer (9% vs. 23%). These differences may
explain both HR zone percentage and TID differences.
As well as the above suggested reasons for differences in TID, we cannot discount
possible physiological contributions. First, in alignment with other markers routinely measured
by the athlete’s physiologist (time trials and power profiles, Table 4), the individual athletes
may exhibit somewhat different opposing inherent energetic profiles (Table 1), with Rower A
perhaps more aerobically dominant and Rower B perhaps more anaerobically prevailing.
Indeed, this notion is supported by the peak blood lactate values at the end of a 7 x 4-min step
test presented in Table 1, as well as performance data presented in Table 4, with Rower B
showing high blood lactate values, and greater superiority in time trials over shorter distances
on the erg compared to Rower A. In concert with these physical profiles is the observation of
individual athlete tolerance, motivation, and personal satisfaction in completing periods of
work within their area of relative energetic dominance. To be clear, the aerobically-dominant
Rower A was routinely more motivated to complete longer aerobic training, while the more
anaerobically-gifted Rower B more inclined to complete intermittent high-intensity power
intervals. Thus, while the athletes routinely performed on water rowing training together, there
was a good degree of individualization applied by these athletes and their support team to allow
for the different polarization profiles to emerge (i.e., more long aerobic cycle training for
Rower A and higher intensity intervals on the rowing ergometer for Rower B (Table 2). Perhaps
these choices came from an innate perception of what works best for each rower on an
individual basis, and provides some evidence for the individual responses to training programs.
As seasoned senior athletes, they intuitively appeared to know what worked best for them, an
accordingly adjusted their training more towards either the polarised or the pyramidal TID,
within the confines of the training program offered by the coaching team.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Another potential factor contributing to the different training distribution profiles may
be the degree of parasympathetic dominance in Rower A versus B. Rower B had a markedly
higher peak heart rate (Table 1), which may be evidence of lower inherent parasympathetic
activity. Indeed, average resting heart rate variability of these two athletes supports such a
notion, with higher resting levels of parasympathetic modulation in Rower A (Rower A rMSSD
= 151± 21 ms; Rower B rMSSD = 119 ± 40 ms). This results in a slower heart rate recovery
between high-intensity efforts and faster cardio-acceleration, which could have influenced time
in zones and resulting outcomes.18 This again likely favours even more the distinctly different
TID profiles shown herein. Within this same notion, the observed relationship between low
intensity training and total training hours may be due to a greater parasympathetic dominance
and slower cardiac acceleration that would naturally arise with heavier training loads,19 as well
as due to the generally higher total weekly training time accumulating as the season progressed.
Another interesting finding within the present study was the generally low amount of
training completed by the athletes according to elite rowing standards (12-15 h/wk; Table 2 3).
We also showed that training time at a low intensity increased as the training hours increased
(Figure 3). As such, they naturally appeared to become more “polarised” with increasing
training hours, possibly so as to inherently avoid excessive fatigue or overreaching. Indeed,
moderate correlations were shown between total training time and the proportion of time spent
<LT1. Thus, within the context of maximising 2000m rowing performance (~6 min all out),
these data, comparing two elite individuals of similar performance standard, could suggest that
it is more the overall training stress that is the more important factor related to maximising
performance, as opposed to a polarised TID per se.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Practical Application
For practitioners who monitor training in elite athletes and pay attention to TID, these
data are of interest. Perhaps for the first time, we have shown two differing TID profiles
(polarised vs. pyramidal), in two very high level athletes achieving optimal results across a
similar training program. However, although this is apparent in these data, over longer time
periods the TID model can change (i.e. from pyramidal to polarised models and vice versa).20
It is also worth noting that, in the case of both rowers, the majority of time spent training was
still at an intensity <LT1.”
From a practical application standpoint, these data beg the question: “which TID
method is superior and which should be strived for in training?” Based on these results, it would
appear as if either can be effective, and the success of each model may be due more to years of
training, individual physiological responses, as well time available for training.
When considering the three TID models presented herein, the threshold as to where one
model ends and the other begins is also influential on the resulting interpretation. As such,
accurate TID identification can be difficult when using the heart rate time in zone method. For
more accurate demarcations, practitioners should also consider other methods of assessment
such as the “session goal” method.
Conclusion
Two of the world’s best rowers, rowing together in the same boat with an undefeated
record across an Olympic cycle, appeared to take markedly different “roads to Rio” within the
context of TID, with one rower polarised and other pyramidal. Training intensity distribution
tended to become more polarised in both rowers as training duration increased.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Conflicts of interest
The authors declare no conflict of interest associated with the publication of this study.
Acknowledgements
The authors would like to sincerely thank the two athletes for consenting to have their data
included in this study, as well as the reviewers of our manuscript for their helpful comments
and suggestions.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
References
1. Stoggl TL, Sperlich B. The training intensity distribution among well-trained and elite
endurance athletes. Front Physiol 2015;6:295.
2. Seiler S. What is best practice for training intensity and duration distribution in
endurance athletes? Int J Sports Physiol Perform 2010;5:276-91.
3. Fiskerstrand A, Seiler KS. Training and performance characteristics among Norwegian
international rowers 1970-2001. Scand J Med Sci Sports 2004;14:303-10.
4. Stoggl TL, Sperlich B. The training intensity distribution among well-trained and elite
endurance athletes. Front Physiol 2015;6:295.
5. Neal CM, Hunter AM, Brennan L, et al. Six weeks of a polarized training-intensity
distribution leads to greater physiological and performance adaptations than a threshold
model in trained cyclists. J Appl Physiol 2013;114:461-71.
6. Laursen PB, Jenkins DG. The scientific basis for high-intensity interval training:
optimising training programmes and maximising performance in highly trained
endurance athletes. Sports Med 2002;32:53-73.
7. Seiler KS, Kjerland GO. Quantifying training intensity distribution in elite endurance
athletes: is there evidence for an "optimal" distribution? Scand J Med Sci Sports
2006;16:49-56.
8. Lucia A, Hoyos J, Pardo J, Chicharro JL. Metabolic and neuromuscular adaptations to
endurance training in professional cyclists: a longitudinal study. Jpn J Physiol
2000;50:381-8.
9. Esteve-Lanao J, San Juan AF, Earnest CP, Foster C, Lucia A. How do endurance
runners actually train? Relationship with competition performance. Med Sci Sports
Exerc 2005;37:496-504.
10. Neal CM, Hunter AM, Galloway SD. A 6-month analysis of training-intensity
distribution and physiological adaptation in Ironman triathletes. J Sports Sci
2011;29:1515-23.
11. Laursen PB. Training for intense exercise performance: high-intensity or high-volume
training? Scand J Med Sci Sports 2010;20 1-10.
12. Hahn A, Bourdon P, Tanner R, eds. Protocols for the physiological assessment of
rowers: Human Kinetics; 2000.
13. Newell J, Higgins D, Madden N, et al. Software for calculating blood lactate endurance
markers. J Sports Sci 2007;25:1403-9.
14. Lucia A, Hoyos J, Perez M, Chicharro JL. Heart rate and performance parameters in
elite cyclists: a longitudinal study. Med Sci Sports Exerc 2000;32:1777-82.
15. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies
in sports medicine and exercise science. Med Sci Sports Exerc 2009;41:3-13.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
16. Sylta O, Tonnessen E, Seiler S. From heart-rate data to training quantification: a
comparison of 3 methods of training-intensity analysis. Int J Sports Physiol Perform
2014;9:100-7.
17. Hagerman FC, Lawrence RA, Mansfield MC. A comparison of energy expenditure
during rowing and cycling ergometry. Med Sci Sports Exerc 1988;20:479-88.
18. Aubert AE, Seps B, Beckers F. Heart rate variability in athletes. Sports Med
2003;33:889-919.
19. Plews DJ, Laursen PB, Stanley J, Kilding AE, Buchheit M. Training adaptation and
heart rate variability in elite endurance athletes: opening the door to effective
monitoring. Sports Med 2013;43:773-81.
20. Bourgois J, Steyaert A, Boone J. Physiological and anthropometric progression in an
international oarsman: a 15-year case study. Int J Sports Physiol Perform 2014;9:723-
6.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Figure 1. Mean TID across 4 rowing seasons for Rowers A and B. Each grey bar represents
time spent below the first lactate threshold (<LT1), time between the first and second lactate
thresholds (LT1- LT2) and time above the second lactate thresholds (>LT2). Between-year
substantial differences (i.e. 2013 vs. 2014, 2014 vs. 2015 and 2015 vs 2016) for each training
intensity displayed qualitatively as follows: # = likely decrease, ## = Very likely decrease, **
= Very likely increase.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Figure 2. Between-athlete standardised differences and 90% confidence limits for all training
intensities during the 2013, 2014, 2015 and 2016 seasons. Training intensities are distributed
as follows: time spent below the first lactate threshold (<LT1), time spent between the first and
second lactate thresholds (LT1- LT2) and time spent above the second lactate thresholds (>LT2).
The grey bar represents a trivial effect (<0.2); markers above identify more time spent at that
training intensity for Rower A vs Rower B. Substantial between-athlete differences in the time
spent at each training intensity are displayed qualitatively as follows: * = likely, ** = Very
likely, *** = Most likely.
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Figure 3. The magnitude of correlation and 90% confidence intervals for time spent at a
training intensity below the first lactate threshold (LT1) and total training hours for Rowers A
and B. R = 0.43 (0.17; 0.63) and 0.38 (0.11; 0.6).
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Table 1. Physiological characteristics of Rower A and Rower B. Data are the best recorded values over the 4 years period. Peak blood lactate and
hear rate recorded at the end of a 7 x 4 minute step test
Rower
Body
Weight
(KG)
Height
(cm)
VO2 max
(L/min)
Peak
HR
Average
4 min
max (w)
LT1
power
(w)
LT1
HR
(bpm)
LT2
power
(w)
LT2
HR
(bpm)
Peak
Blood
Lactate
(mM)
Max
HR
(bpm)
Rower
A
89.5
188
6.4
182
531
320
144
408
165
6.1
185
Rower
B
99.4
195
6.7
195
552
330
160
410
175
10.8
200
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Table 2. Training modality data for Rowers A and B for season 2013 to 2014 as a percentage
of other modalities. Data also includes training stress scores (TSS).
Year
Weekly
Training
(hh:min)
Rowing on
water (%)
Concept
Ergometer
(%)
Cycling
(%)
Other
(%)
TSS
2013
13:24
72.9
11.4
13.6
2.1
690.7
2014
14:11
63.1
12.9
22.2
1.8
702.5
2015
14:14
64.3
11.7
20.1
3.9
782.5
2016
15:14
53.2
8.9
30.4
7.5
797.4
2013
12:41
76.2
14.2
6.3
3.4
715.6
2014
13:36
65.2
18.0
16.1
0.7
590.7
2015
13:26
61.4
23.3
14.2
1.1
545.2
2016
13:52
66.6
22.6
10.4
0.5
739.0
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Table 3. Training intensity distribution comparisons from the first to the last training macro-cycle of competitive season. Standardised differences
and % chances are shown for time spent below the first lactate threshold (<LT1), between first and second lactate threshold (LT1- LT2) and above
the second lactate threshold (>LT2) for Rowers A and B.
Rower
Year
<LT1
% Chances
LT1-LT2
% Chances
>LT2
% Chances
A
2013
0.29
55/18/27
-0.37
24/17/59
0.87
71/9/20
A
2014
0.09
41/32/27
-0.26
18/27/55
1.02
87/7/6
A
2015
2.12
99/0/0
↑Most likely
-1.96
0/0/99
↓Very likely
0.13
44/32/24
A
2016
0.31
65/28/7
-0.54
3/12/86
↓Likely
1.12
88/6/6
B
2013
1.16
95/3/3
↑Likely
-1.62
0/0/99
↓Very likely
1.96
86/4/10
B
2014
1.12
98/1/1
↑Very likely
-1.23
0/1/99
↓Very likely
-0.08
16/51/33
B
2015
1.36
97/2/1
↑Very likely
-1.38
1/1/98
↓Very likely
0.19
49/30/21
B
2016
2.43
100/0/0
↑Most likely
-2.55
0/0/100
↓Very likely
-0.71
↓Likely
4/9/87
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0

“Training Intensity Distribution Over a Four-Year Cycle in Olympic Champion Rowers: Different Roads Lead to Rio”
by Plews DJ, Laursen PB
International Journal of Sports Physiology and Performance
© 2017 Human Kinetics, Inc.
Table 4. Between-athlete percentage difference in average power between Rower A and Rower
B over various rowing distances recorded in the Concept 2 rowing ergometer.
%
difference
100 m
power (w)
% difference
500 m power
(w)
% difference
2000 m
power (w)
% difference
5000 m
power (w)
%
difference
60 min
power (w)
Rower A
vs Rower
B
-7.3%
-5.1%
-5.3%
-2.6%
-2.5%
Downloaded by Auckland Univ of Tech Library on 09/28/17, Volume 0, Article Number 0