How To Train
This page is my attempt to provide an approach to training based on the currently available research. I've found that most training plans follow established protocols without reference to the latest exercise science. This work is primarily focused on working out how I personally should train, with a focus on ultra-endurance. Exercise performance is incredibly complex, and the science is limited, so I've taken some simplifying steps and used personal experience to compensate.
- 1 Trainable Factors
- 2 Training Techniques
- 3 Fellrnr's Three Component Model
- 4 Training Frequency
- 5 Muscle Fiber Recruitment
- 6 References
1 Trainable Factors
These are the most likely factors that determine performance and are amenable to training. There are other factors, such as genetics that are obviously not trainable.
- V̇O2max. Aerobic capacity appears to be the key component of performance for races from the 5K to 24 hours.
- Running Economy. How fast you run for a given oxygen consumption is another critical aspect of performance. One paper concluded that Running Economy factors included "metabolic adaptations within the muscle such as increased mitochondria and oxidative enzymes, the ability of the muscles to store and release elastic energy by increasing the stiffness of the muscles, and more efficient mechanics leading to less energy wasted on braking forces and excessive vertical oscillation" . There is also evidence that reduced breathing effort is partly responsible for improved RE. See The Science of High Intensity Interval Training.
- Endurance. It's generally accepted that training is required to race longer distances. Being a fast 5K runner doesn't mean you can run the marathon distance. Even trained runners show symptoms of neuromuscular fatigue running marathon or longer races. I've split "endurance" into two parts, mechanical and biochemical.
- Mechanical Endurance. I'm using this term to primarily reflect the ability of the muscles to withstand eccentric exercise, but also to reflect changes in bone and connective tissue. There's good evidence from Delayed Onset Muscle Soreness (DOMS) that this is both trainable and critical for performance. DOMS produces immediate weakness in the muscles, well before any soreness occurs.
- Biochemical Endurance. Beyond the mechanical endurance there appears to be other factors. This is evident from long distance cycling, which includes no eccentric exercise. It's unclear to me the details of this biochemical endurance, but I suspect it's related to the glycogen and fat metabolism.
There are also some other possibly useful training adaptations.
- Fractional utilization. The percentage of V̇O2max that can be maintained for given distances may be both trainable and a determinate of performance. The research for this is a little more limited, and confounded by improvements in V̇O2max resulting in faster race pace, which results in shorter race times, and the fractional utilization for the shorter times is shorter. This concept is sometimes thought of as Lactate Threshold, though that concept is rather more complex than it appears.
- Heat adaptation. This is valuable in races above 40f/4c, which is common.
- Altitude adaptation. There are sea level performance improvements from altitude training, but the cost is prohibitive for many athletes.
- Uphill muscles. Running up hill uses slightly different muscles and biomechanics, so for hilly races some uphill running is likely beneficial.
- Downhill technique. The relative performance benefit of improved downhill running technique might be significant, something that's especially important for trail races. Note that the benefits of downhill running on mechanical endurance are separate.
2 Training Techniques
The options for improving the trainable factors is often are to determine based on the research.
This is the best studied trait as it's easy to evaluate in the lab. The most effective training appears to be High Intensity Interval Training, though any form of exercise is likely to improve V̇O2max in sedentary subjects. A reduction in body fat is another way of indirectly improving V̇O2max (as well as running economy.)
2.2 Running Economy
Improving Running Economy is not as well studied as V̇O2max, but there are a number of well documented approaches.
- There is research supporting the use of HIIT to improve RE
- Explosive (plyometrics) or maximal strength training improves RE, and can even improve the economy of cycling.
- There's some evidence that intermittent hypoxia might help with RE.
- One study suggests that training volume may improve RE, but it's unclear if that's overall mileage covered, or if it's an increase in a specific training type.
It seems likely to me that the underlying mechanism for improving RE via HIIT or plyometrics/max strength training is similar. Given the injury risk of plyometrics/max strength training I'm going to focus on HIIT only.
2.3 Mechanical Endurance
The evidence from Delayed Onset Muscle Soreness (DOMS) is clear that it's the eccentric load that builds mechanical endurance. This occurs in running, but not cycling or swimming, and is greatly multiplied by downhill running.
2.4 Biochemical Endurance
There is very little research available into Biochemical Endurance, and I've reviewed this at The Science of the Long Run.
2.5 Fractional utilization
The vast confusion around Lactate Threshold makes this tricky to study, but the available research indicates that Tempo Runs (and other "threshold" training) are ineffective and should be avoided. Instead, training should be polarized to either be long and slow, or HIIT.
2.6 Heat adaptation
Training in the heat, or passive exposure to heat creates adaptation. See Heat Acclimation Training for details.
2.7 Altitude adaptation
2.8 Uphill muscles
It seems intuitively reasonable to me that uphill running uses different muscles, and therefore there is probably benefit to training on inclines.
2.9 Downhill technique
I have no science for this, but my anecdotal experience is that Downhill Running if far more important than anyone realizes. I've worked with a runner who went from 9 min/miles on a given asphalt downhill to sub 7 min/miles at a similar heart rate and perceived effort on the same hill. Skill on downhill trails can be the difference between a slow walk, picking your way and flying down.
3 Fellrnr's Three Component Model
I'm going to simplify the training model into three primary components; V̇O2max, Mechanical Endurance (ME) and Biochemical Endurance (BE). I'm building on the concept of TRIMP (Training Impulse]] to create a three-component model, TRIMPv, TRIMPme, TRIMPbe. (Note I'm still at the stage of creating a theoretical framework, with implementation some way off.)
TRIMPv is the training impulse primarily from HIIT.
- The nature of HIIT means that Heart Rate can't be used due to the lag. Doing a 20 second "all out" interval might not get the heart rate above an easy run level.
- The training impulse for TRIMPv is going to be based on the time above the power output at V̇O2max. Another approach would be a scaling factor that limits the contribution below V̇O2max without ignoring it completely.
- Scaling needs to be non-linear with power, probably exponential.
- Scaling also needs to include time, as 30 seconds is far more than twice as hard as 15 seconds at high intensity.
- Recovery from HIIT is fairly fast, with evidence that subjects can do 4 days/week. If HIIT is similar to resistance training, then 4 days/week might be optimal. (Big "if" of course.)
TRIMPme is the training impulse from eccentric exercise.
- This only applies to running, not cycling, swimming, etc.
- There is no scaling for intensity, as running faster doesn't create much more impact.
- Scaling needs to be based on slope, with downhill creating a lot more impact, uphill much less.
- Recovery from eccentric exercise is the longest time frame. While the soreness peaks around 24-72 hours after training, the weakness can last 4-7 days.
- The performance degradation from eccentric can be dramatic.
TRIMPbe is the training impulse from endurance exercise.
- While this is complex, I'm going to use the idea of Glycogen depletion as the proxy for the overall training impulse.
- HIIT rapidly depletes Glycogen and will change the TRIMPbe of endurance training that follows rapidly afterward.
- This will scale intensity in the same way that the original TRIMP scales.
4 Training Frequency
The different modalities probably have different training frequencies. HIIT has a faster recovery time than endurance training.
5 Muscle Fiber Recruitment
The depletion of glycogen within slow twitch fibers results in the recruitment of fast twitch fibers. Higher intensities also recruit more muscle fibers, with slow twitch being recruited first, then fast twitch being recruited as intensity increases. This suggests to me that very high intensity and very long duration both have training effects on a greater number of muscle fibers, including the "last responder" fast twitch fibers. I suspect that's why long duration and high intensity can have paradoxically similar improvements in V̇O2max.
- G. Y. Millet, J. C. Banfi, H. Kerherve, J. B. Morin, L. Vincent, C. Estrade, A. Geyssant, L. Feasson, Physiological and biological factors associated with a 24 h treadmill ultra-marathon performance, Scandinavian Journal of Medicine & Science in Sports, volume 21, issue 1, 2011, pages 54–61, ISSN 09057188, doi 10.1111/j.1600-0838.2009.01001.x
- A. G. Scrimgeour, T. D. Noakes, B. Adams, K. Myburgh, The influence of weekly training distance on fractional utilization of maximum aerobic capacity in marathon and ultramarathon runners, European Journal of Applied Physiology and Occupational Physiology, volume 55, issue 2, 1986, pages 202–209, ISSN 0301-5548, doi 10.1007/BF00715006
- Leena Paavolainen, Keijo Häkkinen, Ismo Hämäläinen, Ari Nummela, Heikki Rusko, Explosive-strength training improves 5-km running time by improving running economy and muscle power, Journal of Applied Physiology, volume 86, issue 5, 1999, pages 1527–1533, ISSN 8750-7587, doi 10.1152/jappl.19188.8.131.527
- Mark Tarnopolsky, Guillaume Y. Millet, Katja Tomazin, Samuel Verges, Christopher Vincent, Régis Bonnefoy, Renée-Claude Boisson, Laurent Gergelé, Léonard Féasson, Vincent Martin, Neuromuscular Consequences of an Extreme Mountain Ultra-Marathon, PLoS ONE, volume 6, issue 2, 2011, pages e17059, ISSN 1932-6203, doi 10.1371/journal.pone.0017059
- C. Nicol, P. V. Komi, P. Marconnet, Fatigue effects of marathon running on neuromuscular performance, Scandinavian Journal of Medicine & Science in Sports, volume 1, issue 1, 2007, pages 10–17, ISSN 09057188, doi 10.1111/j.1600-0838.1991.tb00265.x
- B. R. Rønnestad, I. Mujika, Optimizing strength training for running and cycling endurance performance: A review, Scandinavian Journal of Medicine & Science in Sports, volume 24, issue 4, 2014, pages 603–612, ISSN 09057188, doi 10.1111/sms.12104
- M. Burtscher, H. Gatterer, M. Faulhaber, W. Gerstgrasser, K. Schenk, Effects of Intermittent Hypoxia on Running Economy, International Journal of Sports Medicine, volume 31, issue 09, 2010, pages 644–650, ISSN 0172-4622, doi 10.1055/s-0030-1255067
- Peter Krustrup, Karin S??Derlund, Magni Mohr, Jens Bangsbo, Slow-Twitch Fiber Glycogen Depletion Elevates Moderate-Exercise Fast-Twitch Fiber Activity and O2 Uptake, Medicine & Science in Sports & Exercise, volume 36, issue 6, 2004, pages 973–982, ISSN 0195-9131, doi 10.1249/01.MSS.0000128246.20242.8B