In the last decade a growing number of athletes have shifted to using power to monitor and pace their bike training sessions and races. Training by power has a lot of benefits. However, my professional opinion is that training by power should not be done exclusively and is most effective only when incorporating heart rate (HR) based training and monitoring by rating of perceived exertion (RPE). In this newsletter I will provide the benefits and limiters of each type of training and then provide what I see as a balanced training approach of combining power, HR, and RPE into the training paradigm. The majority of this article will pertain to cycling training. However, I will briefly mention the pros and cons of monitoring run training by use of the HR monitor and run pace.
Benefits of Power-Based Training
There are many benefits of training by power. One of the most practical reasons for training by power is that you can dial in short intervals and immediately insure you are training in the proper training zones. An example is zone 5 intervals where one is aiming at improving VO2max/Peak Power Output (PPO). If an athlete is training by HR alone, one may under-estimate or over-estimate the intensity of the short training interval due to the fact that HR does not instantaneously increase when power is increased (aka HR lag). Oftentimes one is nearly complete with the interval before HR reaches steady state (which can take 3+ minutes) and the interval is complete before the athlete realizes he or she has over or under exerted the interval. This is also particularly apparent when training for sprint power/neuromuscular power where training by HR is impossible due to the short intervals. Furthermore, power-based training can be used to calculate training stress and insure training is progressively adding adequate stress to an athlete and more importantly insuring adequate recovery.Caution is warranted, as training stress via the current TSS score formula that has been proposed (1) may lose validity beyond one hour of training. The reason for the decreased validity of the power-based training stress score is that power output (4) and VO2 max (4,8,9) decrease during prolonged exercise sessions. Another benefit of power-based training is that it is very easy to track fitness improvements. One can compare average power during the interval and compare the power to average HR (and vice versa). With proper training, power output will be increased at a given HR. This is critical with cycling performance as tracking speed at a given HR is highly variable (due to temperature, course variations, wind, etc.). On a related note, training by power allows you to record your efforts. For example, you may race on a particular course and note the average power output for that course. At a later time you race the exact same course and have a different finish time. Did you race better or worse? By evaluating the power data, you can decide how you progressed. HR data alone will not tell you this. With power data you can evaluate your strengths and weaknesses.
Explosive Power and its Benefit to the Endurance Athlete
There is recorded data from athletes where you can compare your power output for a given length of time (i.e. 5 second, 5 minute, 20 minute, etc.) to average power output in a given category of cyclists (i.e. Cat 1/2, Cat 3, etc.) It may seem odd for a triathlete to compare power out at any length of time other than one hour or beyond to other athletes, but it is becoming more evident that sprinting ability is actually quite important for triathletes. A triathlete will likely never sprint in a triathlon; however, the concept of power reserve is very important for a triathlete, as power reserve has been shown to be a better predictor of performance than VO2max in time trial performance (7). One of the deficits I frequently encounter with triathletes is the ability to "hammer" the pedal. Power-based training allows one to track the progression of peak power for different time intervals. An analogy of how increasing power reserve can help a triathlete's time trial performance may help. Two athletes are performing the squat exercise with weights. One athlete has a ten-repetition max of 480 pounds and another athlete has a ten-repetition maximum of 280 pounds. Both athletes can squat 250 pounds 20 repetitions; however, one can speculate that squatting 250 pounds 20 repetitions will be more stressful for the athlete with the one repetition max of 280 pounds compared to the athlete that has a one-repetition maximum of 480 pounds. This can be generalized to bike performance. If two athletes have an identical threshold power output of 315 Watts, but one athlete has a peak 30'' power output of 500 Watts and the other athlete has a peak 30'' power output of 600 Watts, one can speculate the athlete with the greater 30'' peak power output will be less "stressed" while racing at threshold (assuming the same training volume/paradigm).
The general training paradigm for coaching and training endurance athletes has been to focus on volume and maintain training intensity below threshold (anaerobic, ventilatory, or lactate threshold depending on the coaches or athlete's nomenclature). However, research is beginning to show that this may not be the optimal training paradigm (2,7) and that adding explosive strength training and sprinting to the training paradigm can yield better endurance performance, perhaps via increased efficiency and increased threshold power (5), and I would also speculate that this is also due to improved power reserve. A relatively recent research article has found that replacing a portion of endurance training with explosive style weight training (while maintaining overall training volume) can lead to improved time trial performance (7.9% improvement in the explosive training group vs 5.9% in the endurance-only training group). Furthermore, the time trial performance was significantly greater at four weeks compared to baseline in the explosive training group, but not in the endurance-only training group. Finally, explosive training led to a significant improvement in peak power compared to the endurance only group (7). One cannot monitor and track sprint training by HR alone; however, using a power meter can greatly assist sprint training. Finally, a major benefit of training by power vs HR occurs when an athlete is taking certain medications. There are a number of medications (i.e. beta blockers), which make training by HR impossible and training via a power meter is the viable alternative.
Benefits of Training by HR
The benefit of HR training includes the ability to factor what is taking place in the athlete's whole-body. Two examples will help illustrate this case:
Example 1: Let’s say an athlete normally relies on power for pacing training intervals and the athlete based the power intervals from a testing performance on an ideal day where the temperature was 70-degrees and humidity was 40% and the athlete felt spectacular. However on a different training day it is 95 degrees with 90% humidity. It will be very difficult for the athlete to be able to maintain the same performance on this hot and humid day compared to the ideal conditions day. HR monitoring will help take into account the training performance for the different training days based on the athlete's condition and the environmental conditions.
Example 2: There is increased VO2 at a constant work rate combined with a decreased VO2 max that occurs with an endurance training session (4,9). Without going into the details of exercise physiology, this means relative workload increases with prolonged endurance performance, even when the absolute workload (Watts) remains constant. An example will help illustrate this. A common method of determining functional threshold power (FTP) for power-based training is the 20-minute maximum effort test. The basic protocol for this test includes a 20-minute all-out boutand the average 20-minute power output is multiplied by .93, .95 or .97 and the derived number is considered FTP. The FTP is then used to develop training zones (1). However, there is a dilemma for ultra-endurance athletes. Is the number, which is derived by a sub-hour test valid 4 hours into a ride? It is highly unlikely that an athlete can produce the same twenty-minute max effort power four hours into a training session as the athlete can with non-fatigued legs. A recent study illustrated this (4) where trained cyclists pedaled at a fixed power output for 2.5 hours and then performed a 5-minute max effort test, which was compared to a baseline five-minute effort. There was a decrease in five-minute power output following the 2.5-hour ride. There was also an increase in oxygen consumption and RPE over time, suggesting a relative increased work rate (the athlete was working harder from an aerobic standpoint later in the ride compared to earlier in the ride). Monitoring training and pacing via HR allows the athlete to adjust the exercise intensity to maintain the same relative intensity. Furthermore, repeated threshold testing is not required when training via HR zones as HR at lactate threshold remains stable throughout the racing season (6), unlike power-based and pace-based training which require frequent testing in order to adjust power and pace zones to account for fitness changes.
Calculating Training Stress by Power, Pace. vs HR
There is a method of calculating stress based on HR known as TRIMP, which takes into account maximum HR, resting HR, training HR, and training/racing time (3). I have modified the calculation of TRIMPS to compare to a 1 hr performance at threshold pace, thus the numbers derived by the modified TRIMPS score can be compared to the commonly used training stress score (1), which can be easily calculated using WKO + software. A major benefit of training by HR is that heart rate is an indicator of cardiac stress (5). There are a variety of conditions (i.e. hot environment, altitude, transition run, etc.) where HR may be elevated due to increased stress. If one is training by power alone; maintaining a given power output that would normally lead to positive training adaptations in a mild environment, may lead to overtraining in a hot environment (5).
The number of athletes using software such as Training Peaks WKO+ is increasing. The software uses power and run pace to calculate training stress for cycling and running, respectively. Under perfect conditions, this is an easy and valid method to keep track of training stress. The benefits of WKO + and tracking training stress by WKO+ are numerous. Under "normal conditions" the training stress score developed by WKO+ and the score developed via the scaled TRIMPS score are very similar. However, problems arise, particularly for tracking running stress under less-than-optimal or unique situations (i.e. transition run, heat, recovery from illness, hills, etc.). The run training stress score calculated by software such as WKO + involves comparing the training session pace to the athlete's threshold run pace. If an athlete is running a hilly course, into the wind, with the wind, is training under extreme environmental conditions, is sick, recovering from illness, etc., the formula of tracking training stress via run pace may no longer be valid.
Let’s look at a couple examples of this.
Example 1: I recently had an athlete perform a run workout following a week of illness. According to the training stress score based on pace, the workout yielded a mild training stress score (based on WKO + software) of 88. However, at the given pace, the HR of the athlete was exceptionally high (likely due to blood plasma volume loss that occurred during the illness) and the HR training stress score was 144. Using the HR value for monitoring training stress, I was able to take into account whole-body stress (5) and realize the workout was not a mild workout, but a rather strenuous one (the equivalent of running roughly 1.5 10K races). The subsequent week's workouts were then modified.
Example 2: Another example of where calculating run stress by pace may not be valid is the transition run. I recently had an athlete perform a 25-minute transition run following a 2.5 hour bike. Calculating run training stress via pace yielded a value of 13.2, which is a very low stress score; however, using HR (which takes account overall body stress: cumulative fatigue from previous training days and the current bike ride, environmental conditions, blood volume loss due to dehydration, etc.) to develop the training score yielded a run training score of 30.4. This is a 230% difference. Though the actual numerical value is low in terms of a training stress score, the difference will be additive when evaluated over several days, weeks, and months when tracking cumulative training stress.
Benefits of RPE
RPE has been used since before the advent of heart rate monitors and power meters. Granted training by RPE could be viewed as less specific as training by HR and power. Nonetheless, RPE should be used under certain circumstances. For example, during the early stages of an Ironman race, HR may not feel in-line with the effort due to the excitement experienced by the athlete. In this instance, RPE may be a valid method of pacing. For a number of athletes that are taking medications that affect HR, training by heart rate is not possible. For these athletes that are taking these medications and do not have a power meter, training by RPE is the only alternative for monitoring training. Furthermore, for the runners taking certain medications, RPE is a useful alternative when conditions are present that will affect pace (i.e. hills, wind, environmental extremes, etc).
Benefits of Training by a Combination of HR, Power, and RPE
I propose a method of combining power based training, HR training, and rating of perceived exertion. There are many coaches out there that propose power-based training as the "holy grail," however this goes against exercise physiology principles on many fronts. Power based training is truly valuable when combined with HR and RPE training. For short intervals power based training insures the athlete is achieving the proper zones. Furthermore, progress can be assessed by evaluating how long an athlete is maintaining a given power output (i.e. perform repeat sessions at various power output to determine critical power and evaluate fitness improvement) and comparing power output to HR (proper training leads to increased power output at a given HR).
Heart rate training allows one to take into account external and internal training variables. Heart rate training also allows one to take into account cardiac drift, which power based training lacks. There are many times when HR and power-based training do not adequately fit the training and racing situation and the athlete needs to go by feel (rating of perceived exertion). This can be addressed in terms of training stress, but is not as precise as gauging stress by TRIMPS or TSS. However this does not discredit the validity of perceived exertion.
Conclusion
A valid training program should not be based on training by power or heart rate exclusively, but should take into account all three: power, heart rate, and perceived exertion. It is acknowledged that not every athlete has a power device at his or her disposal. Cyclists have been training for over a century without power training devices. What is more important is that one takes full utilization of their training devices, whether it is a power meter or HR monitor. Simply looking down at their power meter or HR monitor will not lead to fitness improvement.
1. Allen H, Coggan, A. Training and Racing with a Power Meter. Boulder (CO):VeloPress; 2006.
2. Bastiaans JJ, van Diemen AB, Veneberg T, Jeukendrup AE. The effects of replacing a portion of endurance training by explosive strength training on performance in trained cyclists. Eur J Appl Physiol. 2001 Nov;86(1):79-84.
3. Foster C, Hoyos J, Earnest C, Lucia A. Regulation of energy expenditure during prolonged athletic competition. Med Sci Sports Exerc. 2005;37(4):670-5.
4. Hansen, EA, Jensen K, Pedersen PK. Performance following prolonged sub- maximal cycling at optimal versus freely chosen pedal rate. Eur J Appl Physiol. 2006;98(3): 227-33.
5. Jeukendrup, A, Van Diemen, A. Heart rate monitoring during training and competition in cyclists. Journal of Sports Sciences. 1998:16;S91-9.
6. Lucía A, Hoyos J, Pérez M, Chicharro JL.Heart rate and performance parameters in elite cyclists: a longitudinal study. Med Sci Sports Exerc. 2000 Oct;32(10):1777-82.
7. Paton CD, Hopkins WG. Combining explosive and high-resistance training improves performance in competitive cyclists. J Strength Cond Res. 2005 Nov;19(4):826-30.
8. Wingo JE, Cureton KJ. Body cooling attenuates the decrease in maximal oxygen uptake associated with cardiovascular drift during heat stress. Eur J Appl Physiol. 2006;98(1):97-104.
9. Wingo JE, Lafrenz AJ, Ganio MS, Edwards GL, Cureton KJ. Cardiovascular drift is related to reduced maximal oxygen uptake during heat stress. Med Sci Sports Exerc. 2005;37(2):248-55.
