A few days ago I got involved in a discussion with a famous ex-athlete. We were trying to decide on what's worth monitoring on athletes, from the point of view of optimizing training, diet, or indeed anything else that might enhance performance. You'll be surprised to hear that the athlete was bullish, while I was bearish. Here's my thoughts, anyway, in the hope of stimulating some discussion that might be useful for everyone.

VO2max is categorically a waste of time. The error of measurement is at best about twice that of an athlete's mean output power in an event (which is the same as the variability in time, if you're a runner), so there's no way it's going to track the small changes that would indicate whether the athlete is getting better or worse. Ditto exercise economy (VO2 per unit of speed or output power).

I don't know about the error of measurement of lactate threshold or similar measure of maximum steady state. Someone's doing a search for me. If it's similar or less than the variability in performance (unlikely), then it might be worthwhile. You'd have to be careful about diet, prior hard training, and anything else that might dissociate the lactate threshold from current competitive performance. Lactate threshold and VO2max would only ever be useful for endurance athletes, of course.

Even if you could measure VO2max, economy, and max steady state precisely, do you know of any training or strategy that SELECTIVELY enhances one or two of them? I don't. It's an important question, because you'd want to taylor your training or whatever strategy to correct a weakness or a developing weakness that you'd detected with your lab tests. But if you can't tell me how to correct the weakness selectively, what's the point of measuring it???

Lactates and heart rates in a standardized submaximal test might help you decide if an athlete--not necessarily an endurance athlete--is getting overtrained. There are one or two papers consistent with this use, but is anyone actually using them to monitor their athletes in a real-life situation? I'd like to hear from you if you do, or if you know of anyone who does.

What about monitoring other things to keep track of overloading and overtraining? Is anyone using an orthostatic test and cutting back training if the heart rate rises too much? Is anyone monitoring pituitary hormones or their releasing factors, or catecholamines? These are all usually very expensive to assay. Please let me know.

In the absence of any worthwhile lab test, aren't field tests, time trials, or competitive performances the only things you should monitor?

So, I'd like to hear from anyone who thinks they have a rational, productive monitoring program based on lab tests, or who knows of such a program. Replies to me or the list. I'll send a summary when replies tail off. Thanks!

Will

I've had a great response to my request about monitoring athletic performance. Here's a summary of the replies sent to me personally. I sent this summary to the senders first, to make sure they were happy with it. I've also edited and added the two replies sent to the list, so they're all in one place for easy reference. The message is over 10 KB, but I hope you will agree that exceeding our size limit is justified in this case.

We need an overview. I might have a go after we've had a chance for more discussion.

Will

Brendon Downey <xtr15042401=AT=xtra.co.nz>

[is a very practical sport scientist-coach who thinks, for example, that it's possible to measure things like stride length in subelite runners and devise workouts that attempt to change these towards the values of Olympians. His message opens up the issue of kinanthropometric/skills monitoring...]

"Dr. Hofmann Peter" <Peter.Hofmann=AT=kfunigraz.ac.at>
Inst. of Sports Sciences, University of Graz, AUSTRIA

...We had the same experience that lab tests are not really useful for the regulation of the training of top athletes...

Some ideas from our group about monitoring training in white water kayakers (race performance about 2 min)... Our training system is based on the heart rate performance curve (better known as the Conconi test)...These tests give

an idea of the developement of the specific aerobic performance (kayak speed at anaerobic threshold and maximum performance in the incremental test) during the various training periods. Additionally, we evaluated the anaerobic threshold by means of lactate steady state tests ( 4 times 5 min at the predeterimined heart rate threshold). We found steady state lactate in almost all cases except in the athletes overreaching the threshold indicating that the heart rate threshold (at least as we determine it) gives a lactate steady state indicating the upper limit for heavy endurance exercise. Additionally, we do (in the same test session after a break of 1-2 hours) a maximal test (about 2 min) including lactate measurements after the run (and 3rd, 6th, 9th, 12th and 15th min of recovery) to get an insight in the specific "anaerobic" abilities of our athletes (maximal lactate, increase and decrease of lactate during recovery, decrease of speed during the 2 min run consiting of a number of laps through 3 gates). Beside these tests we do some tests with the same method (based on the heart rate performance curve and if necessary including lactate) on a flywheel ergometer specially adapted for kayakers (during winter training) or running field tests (if they do some running training during the preparation period). Of course we do standard strenght tests in the weight room (maximal strength, endurance).

The system is working quite well, the problem is to detect overtraining, overreaching. We are now evaluating a heart rate based system (beat to beat measurement - Polar System Vantage NV) using heart rate variability (HRV). We try the same test as (I heard) is used in Finland - 5 min lying - 5 min standing - measure beat to beat HR and determine the HRV. Unfortunately, we have only little experience with this kind of test and I think it will need some time to bring it to a standard.

As you mentioned some of the "usual" tests are rather expensive (catecholamines, ....) and a much greater problem to us is that they are invasive and therefore the use is restricted (we don´t get a MD to the training every day).

Our general idea is to find and develope diagnostic tools which can be used by the coach or even the athletes themselves - a heart rate based system workes quite well with some (well known) limitations (of course limited to sports that are at least in part dependent on aerobic performance).

As a remark I do not believe that lab tests are valid for top athletes - probably for beginners or younger athletes (we could see that the increase of performance could be detected in this group with both a running test in the lab and the specific tests on water - but again not in the highly trained athletes).

Enid Brown <brown-e=AT=s-h.uwinnipeg.ca>
Physical Activity and Sport Studies
University of Winnipeg, Manitoba, Canada

Your comments regarding lab/field tests finally struck a resonant chord in me. For over a decade I have been an advocate of field tests - but I have not always been a vocal advocate. For many of the reasons you describe, plus the expense and inconvenience of lab tests, field tests have appeared to me to be the only reasonble approach to regular monitoring. Unfortunately the apparently "simple" field test does not carry with it the panache and mystery of the sophistocated lab test. Perhaps there is an unexpressed fear that sport scientists would be out of work if just about anyone could perform athlete monitoring! The role I see for sport scientists is in the design and validation of field tests so that they, as closely as possible, mimic the demands of the sport performance while providing measureable, reliable results.

Jay.T.Kearney=AT=usoc.org (Jay T Kearney)

...Some laboratory monitoring has an inherent psychological value because it allows an assessment of status without requiring a maximum performance. For a high performance athlete the concept of attempting a "maximum" performance, especially if it is in or very near their competitive performance, is very threatening. There are also lots of sports where a single criterion performace test may not be indicative of the objective of the training period. I think there is still lots of room for our "inexact" markers of assessment...

kirke001=AT=mc.duke.edu (Don Kirkendall)
Duke University Medical Center

You have touched on a topic that confounds many of us in the business. I have said that vo2max is like height and weight, just a descriptive measure of little importance or use. As for some of your specific questions:

Heart rate monitoring

European soccer (Dutch, some Scandanavians and Germans) have used those audiotape-paced shuttle runs to exhaustion once and take a pulse after some arbitrary sub-max run. they then repeat the test periodically to that submax run and take pulse again saving the athlete from going to exhaustion.

Monitoring of overtraining

Carl Foster champions a method of monitoring overtraining asking the athlete to use the RPE scale to rate the practice session. See his summary in MSSE recently on the method. He works with speed skating in the US and would guess they use the method.

Time trials

US cycling has used off and on a time trial (of varying distances).

Pituitary hormones

In Bjorn Ekblom's IOC book on soccer is a chapter on German monitoring of hormones in their athletes. Don't know just how much it is used in practice. Very expensive as you mention.

Soccer

We have developed a battery of field tests to look at the range of energy systems and have data on elite soccer players from U-12 through national team, men and women. Two tests are coaches favorites because the results seem to lend "objective" numbers to their gut feeling on the fitness of their players (300 meter shuttle and the "beep" tests). Others of the battery are more for general athleticism. For a description of the tests visit http://www.us-soccer.com/ then click on "coaching" and follow the link to "assessment of fitness" for information on the methods.

General

I have come to favor some assesment of on-the-field performance. When properly motivated, the tests seem to be fairly sensitive for changes in fitness. That first phrase ("When properly motivated") is the key, though. In the end, I work with ball sports which I think is mostly about the ability to recover fast rather than out-and-out endurance.

What do you see or hear about in terms of training and assessment of agility? Big interest in ball sport coaches as that seems to be the main discriminator among and between athletes and from other types of sportsmen.

Tom Switzer <tswitz=AT=mail.utexas.edu>

Yep, I agree about the VO2 measurements. They may have some applications with developing athletes, but are WAY too insensitive to measure minute changes, especially during the course of a season. They may be applicable to athletes coming off a long layoff going into a season. Lactate data is interesting. I've used a handheld analyzer both, in the lab and in the field. Lab results are comparable to bench assays. However, you need too much blood from a finger stick to get three usable drops for sampling, which also becomes cost-prohibitive at $2/strip. Additionally, field samples are easliy contaminated with sweat and grime, and again, most athletes don't like having sore fingers for days afterwords. It'd be nice to take an antecubital venous sample, but this poses legal considerations for us non-MD's. Maybe we could consistenly measure hydration status, as pernicious dehydration has detrimental effects on day to day training (at least in hot climes). Additionally, and I think most importantly, all the laboratory data is worthless if the athlete's head is not in the game. Motivational "concerns" are going to override physical performance changes. Can we eliminate potential placebo effects of such goodies as creatine or a shiny new bike?? Perhaps the most powerful ergogenic aid is a good shrink?!!

On a similar topic, has anybody measured lactate, power output, VO2 during motorpacing in cyclists? I have spoken to several cyclists and coaches who swear that motorpacing on the day prior to an event "primes" a rider more than any other pre-race prep. Physiologically speaking, the responses should be similar to that of riding in a big pack. Maybe it's the feeling of speed per unit effort which somehow "tricks" the body (placebo effect or psychosomatic?). The effects do appear to be real, as most of the riders who've prepared this way report, informally, that they feel ready, and their results support this...

"John D Carney PA-S2, R.D., CNSD" <jcarney=AT=chs.edu>

I'm not certain if this is the sort of thing you're looking for, but a while back, while reviewing literature on glutamine, I ran across an article which looked at glutamine levels in regards to overtraining. In addition, there was a glutamine article which addressed the countering effects of glutamine in regards to catecholamines...

David Liow <liowda=AT=mail.cit.ac.nz>

...One reason that I like VO2 max testing is that it scares lazy athletes (particularly in team sports) and makes them work harder in training.

"Ebert, Tammie" <Tammie.Ebert=AT=Central.tased.edu.au>
Tasmanian Institute of Sport

...with our rowing testing we [use] a submaximal testing approach where we have individuals work at an increasing intensity until they reach just over 4 mmol/L of lactate then the test is terminated. We are then able to plot lactate and heart rate versus power and determine the power output and heart rate at 4 mmol/L (also 2 and 3 mmol/L). We have found this test to be quite good at monitoring changes with training and have just completed some preliminary work in devising a similar test for cycling.

This type of testing allows us to do it regularly without putting the athletes under the stress of a maximal test and it is quite sensitive to measuring changes after a month of training.

I am also currently doing an assignment on VO2max and lactate threshold as predictors of endurance performance. A large amount of the literature steers towards LT as a better predictor than VO2max due to the fact that it is close to the intensity that endurance athletes compete at. I would be interested if you had any research papers that you could recommend I source regarding this topic.

p.pfitzinger=AT=auckland.ac.nz (Peter Pfitzinger)
UniSports Centre for Sport Performance
University of Auckland

...[you wanted] information on the reliability of lactate measurement. Just happened to do my masters thesis on that topic. Please see: Pfitzinger, P. and P. Freedson. The reliability of lactate measurements during exercise. Int. J. Sports Med., 19:349-357, 1998. [I'm currently sorting through issues of the calculation of variability (coefficient of variation) in this paper.]

David Rowbottom <d.rowbottom=AT=qut.edu.au>
School of Human Movement Studies, Queensland University of Technology
...There is a reasonable amount of data to suggest that submaximal lactates may in fact be lower in an overtrained athlete. How can you then distinguish this athlete from another athlete whose performance is improved, and should also have lower submaximal lactates? In other words an improved performance and a reduced performance will produce the same result, rather than distinguishing them from each other. There is also very little data to suggest that submaximal heart rates are affected during overtraining.

I would support the notion of a time trial or competitive performance as the best option. Have a look at a recent paper from Urhausen et al. (1998) - MSSE 30, 407-414. They reported an endurance test at 110% anaerobic threshold in cyclists during normal training and after a period of intensified training. Despite a 27% reduction in endurance time to exhaustion (about 15 minutes instead of 22 minutes), there was no difference in submaximal blood lactate or heart rate after 10 minutes of the test, no difference in peak power output during a separate incremental test, and no difference in 10s or 30s anaerobic power output tests. If none of the usual physiological measures can account for the reduction in performance, should we be measuring them to check for overtraining?

Mark Sutherland <perform=AT=pro-sport.co.nz>

...It may be worth checking this page out:
http://www-rohan.sdsu.edu/dept/coachsci/vol15/table.html

[It sure is. This is one of the pages at the valuable Coaching Science Abstracts site http://www-rohan.sdsu.edu/dept/coachsci/intro.html, maintained by Brent Rushall and Robert Carlson.]

On 2 October I posted a summary of responses on the topic of whether sport scientists and coaches can keep track of their athletes by monitoring anything better than time trials or real performances. I received a couple more responses after that, and I have now sorted out calculation of reliabilities of lactate measures in a recent paper. Here's an update.

First, something from me. It turns out that lactate thresholds can be VERY reliable. Pete Pfitzinger and Patty Freedson (Int J Spors Med, 19, 349-357, 1998) looked at six different measures of lactate threshold. The coefficients of variation (CVs) were 1.2 to 2.9%. If we take the median measure of 1.5% (which was the value for the classic 4 mM speed), it means that an athlete's typical variation in speed on a treadmill from test to test was 1.4% of their typical speed. The likely range of the true CV (not in their paper) is 1.1% to 2.0%, so this low CV is unlikely to be a statistical fluke. A CV of 1.5% is the same as a world-class 5000-m runnner's variation from competition to competition (my own as yet unpublished observations). So you really can track endurance performance with this test.

Whether Kim Coach or even Sue Scientist can expect to get this sort of reliability routinely is another question. In two other studies, the coefficient of variation of the lactate measure was 2.5% and 3.0%. Maybe in real life you can't control things well enough to get the kind of reliability you need for the test to be useful. We must also keep in mind the following remark from David Rowbottom <d.rowbottom=AT=qut.edu.au>, in my previous summary. Can anyone tell us how to get around this problem? "...There is a reasonable amount of data to suggest that submaximal lactates may in fact be lower in an overtrained athlete. How can you then distinguish this athlete from another athlete whose performance is improved, and should also have lower submaximal lactates? In other words an improved performance and a reduced performance will produce the same result, rather than distinguishing them from each other."

While I'm on the subject of reliability, I have been way behind in my reading and only recently found the paper by Pereira and Freedson in Int J Sports Med 18, 118-124, 1997, on the reliability of measures of running economy. It was only 1.9%, so again, it's getting small enough to track changes, if you think changes in economy are an issue. But you've got to be able to calibrate your metabolic cart reliably, and you've got to believe the gas concentration on that next new cylinder of calgas...

Here are the other replies. I'll try to find time to give an overview of all this in the Ferret column in the next Sportscience News.

Will

Carl Foster <KarlHeintz=AT=aol.com>

About 6-8 years ago, I wrote a piece that was published in Physician and Sportsmedicine. In it we argued that to monitor athletes you needed a test procedure that: a) correlated with performance, b) changed to reflect changes in fitness and c) was ameniable to goal setting. Since then, about the only things we have come up with for the speed skaters are time trials on a windload braked cycle (maximal exercise) and lactate profiles on the cycle or on well prepared indoor ice. We tried to do a systematic evaluation of responses during warmup, thinking that if HR, lactate or RPE were displaced during a standard cycle ergometer ride at the end of the warmup period we might be able to identify athletes who were becoming seriously overreached and intervene. Mostly for logistic reasons, we have so far failed to get enough data to test this concept fairly, although I think that it remains a viable strategy.

We have done some work with very simple monitoring of the training load and pattern based on the perceived exertion to the whole training session (recently reported in MSSE). Over long periods of time, characteristic patterns do tend to emerge for each athlete who uses it, such that they can usually identify a pattern that is associated with impending illness. However, also for logistic reasons, we haven't enough data with the elite skaters to know if it works outside of a laboratory like situation.

What no one has mentioned, is the subjective value to the coach of seeing the athlete perform in the lab. Whether a time trial or a standard training session, we have found it useful to have the coach stand by the cycle, giving the athlete splits just as they might in competition. However, insead of the athlete whizzing by a 50+ kph, the coach can "look in their eyes" and get a sense of how they respond in competitive simulations. Several of the speed skating coaches have remarked to me that this has a value independent of any numbers that I might generate, although they seem to like those also. Thus, it may be that involving the coach in the laboratory monitoring situation, to give them a "God's eye view" of an athlete who they would never ordinarily see up close during high intensity effort is coequal in value to objective outcomes.

"S. Kelly Harris" <skharris=AT=julian.uwo.ca>

You are correct about the error and that small changes are difficult to interpret. I think it depends upon the quality of the measure (gas chromatography vs mass spec) and the experience of the technician in controlling the lab environment, ensuring the subject is controlled as much as possible (diet, time of day, point in training, etc.). Even with tight controls you will still see the amount of error you mention.

What I mean in saying "as a guide" is that many people interpret test results, black vs white, yes vs no, and should be careful in the way they interpret and act upon test results. All things should be considered. The coach and exercise physiologist should share as much information about what the results could, do, or do not, mean considering all conditions of the athletes status (psychological, physiological, etc.). Performances may be down and VO2max may be up and the coach should be able to interpret why from working and communicating closely with the athlete.

I also see that you are speaking more along the lines of the world's elite and that there are smaller changes at that level and obviously harder to interpret if they can be interpreted at all. Once again, this is why I use the term "guide". One shouldn't panic if their test scores are down 3% on a one-off test. As you asked, how often have the measures been repeated. I think you have made some very valid points in that with athletes at the highest level when you want that 2% improvement that means the difference between 5th and 1st, how the hell can we monitor the athletes more precisely and reliably. I was coming at this from my present coaching situation where I have athletes in a local club which can see huge performance gains in a year if they choose to do the work.

You are also correct about the interpretation of ventillatory thresholds being difficult to interpret reliably but I've had lots of experience with it in the last year and am a whole lot better at the art or should I say witchcraft than I was a year ago. It really depends upon how many times you repeat the measure per condition, your experience interpreting the data, and how great the change is from condition (early season) to condition (late season).

..I hope to make coaching more science than art but leave enough art in it keep it interesting...

This is a comment regarding one of the questions raised in the "What's Worth Monitoring Update (Oct 21 1998). Paragraph 3 read...

...We must also keep in mind the following remark from David Rowbottom <d.rowbottom=AT=qut.edu.au>, in my previous summary. Can anyone tell us how to get around this problem?

"There is a reasonable amount of data to suggest that submaximal lactates may in fact be lower in an overtrained athlete. How can you then distinguish this athlete from another athlete whose performance is improved, and should also have lower submaximal lactates? In other words an improved performance and a reduced performance will produce the same result, rather than distinguishing them from each other."

In answer to this...

1. If the incremental procedure which generates the submaximal data culminates in a maximal, therefore performance, test then this lactate 'paradox' can be resolved.

2. The same paradox may also occur with changes in dietary status. Inclusion of a maximal test may also contribute to this being resolved.

3. I believe that the use of maximal performance tests to terminate incremental tests is critical to the validity of this type of procedure especially where data is being related to athletic performance.

4. When an element of performance is included within an incremental test, the whole procedure, when repeated over the training year, logically allows an "analysis of changes in performance". That is the ability to track changes in performance and directly relate this to changes in the physiological properties which underpin the performance.

In isolation, any submaximal analysis using measures of blood lactate is subject to the concerns raised by David Rowbottom.

David H. Saunders
FASIC, Centre for Sport and Exercise
University of Edinburgh
46 Pleasance
EDINBURGH EH8 9TJ
Email Dave.Saunders=AT=ed.ac.uk
Phone 0131-650-2589
FAX 0131-557-6398

While Will has already posted a summary to his enquiry about "what's worth monitoring in athletes", my colleagues and I still wanted to respond, after returning from a very successful Commonwealth Games campaign. We are routinely and professionally involved in testing elite athletes and therefore like to think we can actually make a difference diagnostically and prescriptively (I accept this may actually be the point underlying the debate, but will try to add some rationale for our positions).

Our underlying belief in testing physiological (and mechanical and perceptual) variables in an athlete rather than simply using time trials or competitions is that a finishing time will tell you how well an athlete performs, but not why. A particular performance can be achieved in many different ways by different or the same athlete at different times, but a stopwatch does not tell you the breakdown of its component parts.

Will is right in saying that if you can't selectively train/change an identified component then you are perhaps wasting your time measuring it, and I will address that later. However, his first up example of measuring VO2max perhaps typifies the continuing need to explore SPORT rather than EXERCISE science: I was guided years back in England to see VO2max as a potential for performance in some sports (rather than as an indicator), irrelevant to performance in some sports, and central to performance in others. This comes down to an argument of whether to measure capacities or powers, and essentially highlights the need to analyse the particular sport and assess what contributes to performance, then focus measurement there; for example, measuring VO2max in rugby players provides a number that everyone can relate to, but is arguably irrelevant to performance.

In that vein, we employ a wide range of tests with different sports, most of which we have examined for reliability. The chosen statistic for this, under the guidance of the Laboratory Accreditation Scheme of the Australian Sports Commission, is the Technical Error of Measurement, basically looking at test-retest reliability and interclass correlation. Taking a typical incremental test, using an average of seven 4 or 5 min stages, we can then measure heart rate, lactate, and power at the individual anaerobic threshold with TEM of around 2.4%, and at the lactate threshold to 1.8%; TEM for VO2max in such a test (not specifically designed for VO2max testing) is around 2.7%. We have similar data for field tests such as electronically timed sprints, the multistage fitness test, vertical jumps, etcetera: in fact anything we forward under athlete testing for lab accreditation.

In parallel to this, variability during training of maximal time trial performances is quite high. I have seen an Australian male representative 100 m freestyle swimmer (best time 49.9 sec) struggle to break a minute, and an Olympic track cyclist peak under 1000 W in a simulated sprint (best around 1800 W). It is likely that variability of such maximal - or even approaching maximal - tests is higher than for submax efforts, making diagnostic tests at submax intensities preferable substitutes during heavy training.

The commonality, whether using tests or trials, is the need for standardisation in environment and athlete preparation, and we are confident - as scientists - that when this is adhered to, variability in chosen diagnostic tests is within the variability seen in elite performance. Foremost in the factors to control is (perhaps obviously) the athlete's state of recovery on the day.

We then explore as many contributors to performance to performance as will help paint a picture of the athlete's current status. This includes, for example with swimmers, heart rate, lactate, stroke rate, stroke count, and perceived exertion at submax and maximal swimming speeds. For track cyclists in a laboratory - where control is commensurately increased - we may examine the same variables plus VO2 if endurance testing, or VO2 dynamics, accumulated oxygen deficit, and acid-base balance if looking more at speed. The list becomes exhaustive as more activities are considered, extending to the gym, the field, or the video camera.

Where possible we draw from established literature and apply in to sport. So, for example, we make calculations of critical velocity from swimmers' race performances (Wakayoshi, 1992; Day, 1996); to take up David Rowbottom's point, we look at Snyder's (1993, 1995) work on the Lactate:RPE ratio as a differential of low lactate in well-trained or overtrained athletes; and we might look at Tegtbur's (1993) method of determining maximum lactate steady state, independent of an athlete's glycogen levels. Ultimately, we are striving for a complete picture of the athlete with as much scientific control as possible to augment the coach's immediate view of final performance.

Whether these things can then be trained selectively will always remain contentious. However, to give but one example, we have calculated periodically through a rowing season the relative functional buffering capacity described by Rocker (1994), and seen it respond differently to endurance or high-intensity training. In short, we saw sufficient evidence that the lactate and individual-anaerobic thresholds are independent variables, not just cross sectionally but longitudinally in the same athlete. If we carry this belief to other tests, and apply our scientific and coach's artistic skills, we have to believe that we can selectively influence performance, else the training program would be irrelevant and anyone that just worked hard would achieve success.

I look forward to any feedback, supportive or opposing, and comment on published tests others may be applying in similar situations as ours. Thanks.

Graeme Maw, PhD
High Performance Manager
QAS Swimming
Queensland Academy of Sport
PO Box 8103
Woolloongabba
QLD 4102
Australia
Tel 61 7 3404 3222
Fax 61 7 3404 3205
Email graeme.maw=AT=dtsr.qld.gov.au

Refs.

Day & Lin (1996). Critical velocity as a predictor of female front crawl swimming performance. Med Sci Sports Ex, 28(5), s158.

Rocker et al (1994). Relative functional buffering capacity in 400-meter runners, long-distance runners and untrained individuals. Eur J Appl Physiol, 68, 430-434.

Snyder et al (1993). A physiological/psychological indicator of over-reaching during intensive training. Int J Sports Med, 14, 29-32.

Snyder et al (1995). Overtraining following intensified training with normal muscle glycogen. Med Sci Sports Ex, 27, 1063-1070.

Tegtbur et al (1993). Estimation of an equilibrium between lactate production and catabolism during exercise. Med Sci Sports Ex, 25, 620-627.

Wakayoshi et al (1992). Determination and validity of critical velocity as an index of swimming performance in the competitive swimmer. Eur J Appl Physiol, 64, 153-157.

Personally, I really appreciate this particular discussion, and the input from Will H., Carl Foster etc.

A few more comments:

I think it is clear that we are VERY hard pressed to quantify the small and transient functional changes associated with top level performance variation, at least in the endurance sports where we this discussion has been focusing. What about strength and power sports? I remember a Bulgarian coach now at UT-Austin named Angel Spasov telling a female shotputter I mentored at the time "When you can bench press XX kg, you will be ready to throw XX meters. This will happen the first week in June." He missed the date by a week but got evrything else right. Is there any value in the volleyball coach measuring vertical jump to monitor if his athletes are overreaching? Can we predict the onset of a dead baseball throwing arm before it strikes, with non-throwing tests?

Back to the endurance folks, my experiences with testing have been primarily limited to developmental athletes, where the changes can be significant from test to test. Here, much of the value of testing I have observed comes from:

1) the teaching opportunity that is presented, if we use it to explain to athletes what we are measuring and how we think it relates to their training and performance. A coach-athlete-physiologist discussion following testing seems really effective, if we make time for it.

2) the value of regular testing; One-off tests have VERY limited value. However, when we are able to tests athletes at regular intervals, in the company of their coaches who know precisely how they have trained/performed in that interval, then I feel like we are contributing to the fine-tuning of their training program in a positive way. Often the coaches like the numbers in order to put a stamp of validation on the feedback they are giving their athletes, especially young athletes whose training efforts can be quite variable.

3) Pure psychology: A great example is Bjorn Dæhlie, who has won 8 gold medals in the Olympics. He is the king of his sport of XC skiing. Bjorn Dæhlie is Bjorn Dæhlie's coach. He is one of the best ever at being tuned in to his own body. However, he still wants that regular treadmill test of his VO2 max, performed by the same guys in the same lab in Oslo. He still performs his 6 minute On- ski performance test on the same home course he has used for 15 years. He still needs those pieces of feedback to confirm for himself that he is at the right level of fitness at the right time as he moves towards a new season and a new performance peak.

Stephen Seiler PhD
Assistant professor
Institute for Sport
Agder College
Kongsgard alle 20
4604 Kristiansand, Norway
email: Stephen.Seiler=AT=hia.no
phone: (47) 381 41 347
fax (47) 381 41 301
Endurance performance physiology website (MAPP):
http://www.krs.hia.no/~stephens/index.html
Sportscience research/education website: http://sportsci.org