Monday, April 16, 2012

Notes on Randall Wilber's "Altitude Training and Athletic Performance"

If you are interested in altitude training, Randall Wilber's book "Altitude Training and Athletic Performance" is a fantastic read. I apologize for the lack of posts recently, but I'm recovering from arthroscopic hip surgery last week and am not up for doing a whole lot of sitting (in fact, I'm typing this out standing at my desk, balanced on my good leg!).  To compensate, I'll post the fairly extensive notes I took while reading Wilber's book a few years ago.  Since these were personal notes, I'll also have to apologize for the many instances of typos and misspellings too! Anyways, hopefully you'll find this helpful and interesting if you ever plan on being at altitude for training or non-training purposes.  Enjoy!
Notes on Randall Wilber's "Altitude Training and Athletic Performance"
John's comments in italic

Part I: Potential physiological benefits

Effects of altitude on training:

Upon exposure to altitude, the body makes several adjustments.
*Cardiac output and stroke volume, both in submaximal and maximal exercise, drops.
*Ventilation, that is, the total flow of air in and out of the lungs, increases.
*This increase in ventilation, coupled with hormonal changes causing an increase in urination, gives rise to a great increase in the body's demand for water.  Fluid intake at altitude, especially during the acclimatization period, should be as much as 4-5 L per day. 
*Lactate buffering is impaired during adaptation to altitude because the increase in CO2 exhalation disrupts the equilibrium of the bicarbonate buffering system, allowing H+ to accumulate faster. 
*In men, carbohydrate utilization increases; that is, a larger proportion of the athlete's daily energy comes from carbohydrates.  Women maintain a similar level of carbohydrate and fat utilization vs. sea level.  Both genders will likely need to increase their caloric intake, but men in particular must consume extra carbohydrates.  High-glycemic food ought to be consumed immediately after exercise at altitude.
*The body demands more iron to produce red blood cells.  Unsuplemented, the ferritin levels of lowland athletes training at altitude often drops significantly.  Furthermore, athletes who are iron-deficient upon ascent (defined in one study as men with <30 ng/ml ferritin and women with <20 ng/ml) respond very poorly to altitude.  Therefore, iron supplements are a necessity.  Not only solid iron sulfate, but so-called "liquid iron" might be a good idea too. One study prescribes 5-45 ml/day of a 9mg/ml liquid iron solution, depending on ferritin levels.
*The increase in oxygen demand, coupled with increased training at altitude, results in an increase of cortisol, a stress hormone.  Cortisol can impair training and depress the immune system, making athletes more susceptible to illness.  It can also cause muscle catabolysis, where the body "eats" its own skeletal muscles.  This muscle catabolysis may be able to be counteracted by consuming protein after exercise and before bed 
*Oxidative stress increases, possibly because of the increased oxygen consumption, increased UV exposure, and free radical production.  Oxidative damage can result in fatigue, DOMS, and injury.  Wilber suggests supplementation with 400mg of vitamin E per day to counteract oxidative damage.  Wilber also mentions sunblock, sunglasses, vitamin A, and vitamin C.  Other (and more natural) antioxidant sources, such as tea, pomegranate juice, blueberries, and grape juice, will likely also go a long ways towards protecting the body.
*Under proper diet and hydration, body composition should not drastically change.

Performance is hampered at altitude for a few reasons:
*Reduced partial pressure of oxygen: Arterial oxygen delivery works by diffusion.  O2 diffuses from the air into the blood, then from the blood into the muscles.  This diffusion is governed by a difference in partial pressure.  Hence, with a lower ambient PO2, diffusion does not happen at as high of a rate, and the arterial blood saturation % drops. 
*Decrease in maximal oxygen consumption:  probably related to the reduction of PO2.  Because blood passes through the lungs too quickly to be fully saturated, the maximum rate of oxygen consumption decreases vs. sea level conditions.  This leads directly to a decrease in VO2 max. 
*Decrease in training capacity: the lack of available oxygen necessitates a slower pace during intense intervals, which can lead to less quality in workouts.

*Sprint work, at least with full recovery, is not impaired at altitude, and is in fact aided by the decrease in air resistance.   

Training response to altitude:
*The decreased arterial saturation of O2 stimulated the kidneys to produce erythropoietin (EPO), a hormone that stimulates red blood cell production. 
*The lack of oxygen also stimulates an increase in the number of mitochondria and the enzymatic activity within them.  Results are inconclusive on whether altitude stimulates changes in capillary beds or muscle composition.

Part II: Altitude Training and Athletic Performance
*results from many years of studies are inconclusive on whether altitude training improves sea-level performance in trained athletes.  Problems result from poor methodology, lack of a control group at sea level, different training protocols, and small study numbers.
*In a retrospective investigation into "responders" and "non-responders," over 80% of "responders" (people who improved after altitude training) were in the "Live High, Train Low" training group.  The rest were in the traditional "Live High, Train High" group.
*It is hypothesized that the ability to do maximal and near-maximal training at near sea level allows athletes to retain the training intensity of sea-level groups but reap the physiological benefits of altitude.
*Of studies supporting the benefits of the LHTH approach, the greatest benefits seem to be from groups training and living between 7,000 and 10,000 feet.
*Of studies not supporting the benefits of the LHTH approach, the only groups with statistically significant DECREASES in performance, VO2 max, and Hemoglobin levels lived and trained at over 13,000 feet.  Therefore, we can conclude that, at the very least, you will not get WORSE by training and living at altitude in any city in the United States.  Up to 10,000 feet seems to be a safe altitude.
*Recent well-designed studies by Levine and Stray-Gundersen suggest that 1) LHTL is superior to sea-level training for periods on the order of one month and 2) Base training at altitude and faster training at low altitude (so-called "Hi-Hi-Lo" training) is equally good From this, it seems obvious that the best choice would be HiHiLo training, because it would save on trips down from the mountains. 
*One recent study asserts that athletes can be divided into "responders" and "non-responders" based on their blood EPO levels (% of sea level baseline: responders had a blood EPO increase of ~156% after 30h and nonresponders increased only ~137% of SL base) and total red cell volume (ml/kg) (total red cell volume increased in responders significantly (~3.0 ml/kg) and remained unchanged in nonresponders).  The "responders" should improve greatly from altitude training. Possibility of testing this before and after altitude training? Blood EPO is measured in IU/ml

"After 14 days at altitude, Epo was still elevated in responders but was not significantly different from sea-level values in nonresponders"

"In conclusion, after a 28-day altitude training camp, a significant improvement in 5,000-m run performance is, in part, dependent on 1) living at a high enough altitude to achieve a large acute increase in Epo, sufficient to increase the total red cell volume and V ˙ O2max, and 2) training at a low enough altitude to maintain interval training velocity and O2 flux near sea-level values."

Practical implementation
Most successful coaches and athletes recommend 7-10 days of acclimatization upon ascent to altitude before intense training begins.  Programs differ during the "stress" phase, but it lasts from 2 to 6 weeks.  Before return to sea level for competition, a week or so of backed-off training is recommended to absorb the hard training.  This may not be necessary if hard workouts are done at low altitude.

Friday, April 6, 2012

The Updated Training Wisdom of John Kellogg

John Kellogg is an enigmatic, immensely intelligent personality who is something of a legend in internet circles.  He is best known for coaching Weldon and Robert Johnson, the twin brothers who founded  Weldon Johnson was a 4:28 miler in high school and a 30:14 10k runner in college whom Kellogg coached to a 28:06 10k and two 4th-place finishes at the US Championships.  His brother Robert, who did not run in college because of recurrent injuries, was coached by Kellogg post-collegiately to 2:23 in the marathon.  John Kellogg has also coached high schoolers and some lesser-known sub-elite runners, including Bruce Hyde, who went from a 4:18 mile to a 3:42 1500m, high school sensation and 8:51 two-miler Ryan Deak, and Paul Stoneham, a 29:00 10k runner.

But more so than his coaching, John Kellogg is known for his writing. was initially started as a platform to promote John Kellogg's coaching philosophy, which can be described as Arthur Lydiard meets exercise physiology meets Zen and the Art.  Over the years, John Kellogg posted hundreds or possibly thousands of long, detailed posts detailing his approach to any runner looking for help, both on and other running forums.  He is also very much a product of the internet: in the early 2000s, perhaps in reaction to the kind of anonymous criticism websites like LetsRun are now infamous for, he stopped posting as "John Kellogg" or (JK), and adopted a rapidly rotating set of pseudonyms, under which he posted prolifically for many years.  In the early to mid-2000s, he would occasionally write articles on training for LetsRun's front page.  While he still continues to post occasionally, the volume of long, detailed posts dropped off after about 2008.

The tail end of John Kellogg's most frequent involvement in the online running community coincided with when I personally started to become interested in serious training.  Through, his writings profoundly influenced my own training, and helped bring me from a middle-of-the-pack scrub to one of the better runners in my conference.  In the ensuing years, I realized how much more of John Kellogg's writings were out there, preserved in obscure corners of LetsRun's message boards, front page articles, and more.  Inspired by a short collection of John Kellogg posts compiled by Tim Galebach in 2003, which was one of the first places I read about John Kellogg's philosophy, I set out to make a compilation of all of Kellogg's writings that I could find.  It became an on-and-off project for me since 2009, and it’s about as done now as it will ever be, so it is about time this project sees the light of day.  John Kellogg's only published work is a chapter in Kevin Beck's 2005 book, Run Strong.

The result of my work is a near-300-page .pdf file, organized by source.  The bulk of it consists of all of John Kellogg’s message board posts I was able to find, made on from 2003 to 2012.  To preserve his anonymity, the pseudonyms he used when making the posts have been removed.  As I say in the preface to the collection, the writings here are likely only a fraction of John Kellogg’s total contributions to the running community.  I have also included all of the front-page articles he has written for LetsRun (the “JK Speaks” series) and the contents of his short-lived website, (coauthored with Kyle Heffner, who placed 3rd in the 1980 Olympic Trials Marathon), retrieved using the Internet Archive’s Wayback Machine, as well as Tim Galebach's original collection.

I hope you enjoy reading through this collection at your leisure, and I hope you find it a helpful resource in becoming a “student of the sport.” 

So, without further ado:

A collection of John Kellogg’s writings spanning several years and many topics 
Compiled by John Davis between May 2009 and April 2012
on Google Docs:
I recommend downloading it (click File -> Download on the Google Docs menu) to read it, since Google Docs doesn't do the greatest job rendering fonts in a non-clunky way.

Thursday, April 5, 2012

Fun with statistics: A tutorial on using linear regression to develop race time conversions

As you can see, I've been playing around with numbers recently.  "Conversion charts" intended to predict future race performances (or equivalent race times) based on a previous race abound on the internet and in running books.  But they often don't have the race distance you are looking for.  And the various conversion tools runners use have very different meanings behind them—two charts may give wildly different "equivalents" for a given time (say, a 4:30 mile).  Why is that?

Well, any conversion chart has a set of assumptions behind it.  When converting between two distances (say, a 5k runner who wants to know what he can run for the 10k), there is usually an implicit assumption that you will have trained for the 10k as well as the 5k.  Charts and conversion factors are often based on runners who ran both events, and thus probably trained for both events too.  Greg McMillan's calculator and Jack Daniels' tables are examples of conversions based on runners who actually prepared for and raced in a wide variety of events.  While McMillan's chart predicts a 4:20 miler can run a 2:26 marathon, that math is based on marathon runners who actually raced a mile during marathon training.  There are a lot of 2:26 marathoners who can't run 4:20 in the mile and a lot of 4:20 milers who can't run a 2:26 marathon, but the chart isn't valid for them because they didn't prepare for both events!

 Another set of "conversions" exists to compare equivalent performances, which is a very different thing than predicting an equivalent race.  Equivalent performances are based on a ranked performance list, so they allow you to compare two completely unrelated events, like the shotput and 1500m.  The IAAF publishes a book of "points tables" that allow for comparing how "good" a performance is in a particular event.  By necessity, however, these kinds of tables will be biased towards distances and events that are not contested as often—as an example, the 100th best steeplechaser in the world will probably be a lot further back in a 1500m race from the the 100th best 1500m runner, simply because there aren't as many competitors in the steeplechase, even though their performances are equivalent on the world stage.  The same holds true for events like the 15k or 10 miles, since they just aren't as many people who compete in them.  This is the weakness of equivalent-performance charts.

Anyways, getting back to the numbers: it will be useful for any coach to be able to develop a formula to give a rough estimate of what to expect when heading into a race.  It's easy to find 3k-to-5k conversions, but what if your conference runs the two-mile? Or the 600 yards? Or if you want to compare two cross country courses to find out how much faster one is than the other?  What follows is a tutorial on how to go about comparing times among two similar distances or courses using marks from runners who competed in both events.  I'll be using Microsoft Excel to do this analysis, but it is also easy to do in OpenOffice or LibreOffice, two free and open-source spreadsheet programs. You could probably even do this in Google Docs if you'd like.