MONITORING THE ELITE ATHLETE
By Wm. A. Sands, Ph.D. and Michael H. Stone, Ph.D.

It is hard to imagine a general not constantly monitoring the condition, position, and readiness of his/her troops. It is equally hard to imagine a stock market investor not monitoring his/her portfolio on a constant and regular basis. However, in elite athlete training we often don’t monitor athlete preparedness beyond simple gut feelings and we too often wait only for competitions to show us whether we did a good job. Sadly, if you wait for competition, you’re probably too late. Monitoring provides a means of controlling the athlete development process and thus a window on preparedness. Monitoring provides the link between planning and preparedness, between planning and performance. Monitoring can provide you with insights into athlete preparation that you’ve never had before.

(This is the first of a two part series concerning Monitoring and Evaluation. The second of the series will be in the next Olympic Coach magazine.)


What is "Monitoring?"

Monitoring can be thought of as the periodic or continuous surveillance or testing of some characteristic(s) of interest. Working backward in the previous sentence:

1. the characteristics of interest are athletic preparedness and performance,
2. testing refers to the examination and/or manipulation of some characteristic(s) to determine its status or change,
3. surveillance refers to the observation of some characteristic(s) to determine its status or change,
4. continuous refers to something that is more or less constant or without interruption
5. periodic refers to something that recurs at regular intervals.

Although we surely monitor performance via win/loss records and competition scores, the hard part about monitoring is surveillance and testing of preparedness. Preparedness can be defined as the difference between fatigue and fitness, and refers to the day-to-day status and condition of the athlete – in essence – the results of training and all other stressors that impact the athlete. While the coach and athlete may have only modest control over the competition because judges, officials, referees, weather, and other aspects of the environment may intrude, the coach and athlete have considerable control over training and therefore preparedness. One of the major goals of monitoring is prediction. By monitoring preparedness, we hope to be able to predict whether the athlete will fare well or poorly in the competition.

Monitoring involves the use of information, obtained from a process, to then alter the very process from which the information came. For example, if we have information from an athlete that tells us that the athlete is suffering from fatigue, we might use that information to then reduce the amount of the athlete’s training demands in order to facilitate recovery. We are using information from the athlete and the effects of the athlete’s training and other stressors to then alter the training demands and thus the athlete. This is a cyclic and recursive (i.e., each cycle calls a copy of itself) process that continues throughout the career of the athlete. Of course, we hope that the alterations are positive in nature and that the athlete improves more due to our intervention than he/she would have without the intervention.

Alterations to training result in changes in the athlete that are delayed. The duration of the delay is often unknown. These changes are then also monitored to ensure that the changes match expectations. If the changes do not match expectations then new alterations are selected and implemented. The point is to drive the athlete/system in the direction you want it to go by a cyclic and recursive feedback system that provides useful information in a timely fashion.

Why Monitor?

Training is a process. A process is one or more actions that bring about a result. Preparedness is the goal of training and performance is the outcome of preparedness. Thus, preparedness and later performance are also processes. Training demands monitoring because the actions or tasks of training do not provide the same results at all times, with all athletes, in all circumstances. Moreover, training may provide results that are not only unexpected, they may be unwanted and perhaps dangerous. The effects or results of training are always preceded by the causes or tasks of training such that in the interim between the tasks and the results we may be able to "listen in" on the process and determine whether the tasks are causing the athlete’s changes to head in the right direction. By "listening in" on the training process we may be able to identify training errors and triumphs in order to avoid the first and emphasize the latter.

Training is a long-term process. The effects of training are delayed and cumulative (58). In a sense, training is an investment not a purchase. Athletes and coaches invest time and effort in training to achieve a delayed result. The delayed result is usually measured as increased preparedness such as: greater fitness, efficiency, skill, elegance, and so forth. Due to the delay, we cannot be certain about causes and effects. There may be multiple things impinging on the athlete that later accumulate to result in changes in preparedness and performance.

For example, we may have an athlete who in different competitive seasons uses two very different training programs. Both result in progress, but during one program the athlete is ill more often. Could this be the result of the training program? How would we know? We are concerned with the influence of training on individual athletes and teams, we can’t really devise a kind of experiment that would allow us to impose the two training programs on identical athletes or teams because by definition – there are no identical athletes or teams when we consider "elite" athletes (44).

And, we are concerned with our athlete right now, we have to work with the athlete we have not some "average" athlete that is the result of a statistical procedure (44;50). In order to understand how long-term processes influence our athlete, we need to monitor and then judge whether there is a good argument for assigning a cause (i.e., training, some other stressor, or the accumulation of stressors) to something that precedes the effect (i.e., illness, progress or whatever).

The effects of training are not entirely predictable, and will never be entirely predictable. We commonly suffer from an argument as old as LaPlace (47) that if we had sufficient information, an all-knowing person or perhaps a computer, could predict everything about the future (43). The problem is that we will never have all the information we need to predict the future, and in principle, we can’t have all the information we need. This means that there will never be any kind of "recipe" for training, something like "do this and then your athlete will win." In principle, such a recipe cannot exist and this places further emphasis on monitoring in order to exert some control over the process via intelligent feedback and rational training changes.

Mistakes are expensive. The investment of time, money, and other resources into the development of elite athletes is extraordinary (5;16). Training mistakes usually involve defeat, poor performance in decisive moments, and/or injury (2). These types of problems have an origin in poor preparedness. Monitoring is one of the few tools available to help coaches and athletes discern how training and other stressors might be linked to all aspects of preparedness and performance. Chronic fatigue and overtraining, although beyond the scope of this document, are the prime suspects in poor performance and may be avoided by early detection via monitoring (2;15;18;30;33;38). Elite athletes must push the edge of their adaptation envelope, always coming close to their limits of their training capacity without going beyond.

Monitoring provides a means of better training management. Management means to control the course of an activity. In order to control the course of an activity, you will need to know how an activity changes and what influences an activity to change. In other words, you’ll need to know what to monitor in order to detect that something in the athlete has changed.

What is worth monitoring?

Once you agree that monitoring is important and worth doing, then the question arises, what is worth monitoring (20;21)? To use the metaphor of altitude, monitoring can occur at ground level (sets, reps, weight, skills, kicks, punches, etc.), from 100 feet (lactate, immunity, glycogen, oxygen uptake, etc.), from the 1000-foot level (group dynamics, team tactics, top eight performances, medal counts, etc.), or from the 10,000-foot level (surveys of coaches, assessment of training plans, etc.).

As such, monitoring is an elastic process that can assist individual athletes (the goal of this article) and monitoring can also be used to keep track of how training and performances are going for team leaders, managers, and others who are not involved in the day-to-day decision making of individual athletes and teams. In this article, we will stay at or below the "100-foot level."

Our concern here is monitoring the "dose/response relationship." Borrowed from medicine, the basic idea is that what we do to the athlete in terms of training demands is the "dose." How the athlete "answers" or how the athlete’s body "replies" to the dosage is the "response."

In a medical setting the typical use of this idea is determining how much of a drug a doctor gives a patient to get a certain response. Dosages for common drugs are included on packaging. These dosages are designed to elicit a certain response in the person taking the drug. For example, children often receive lower dosages of certain drugs than adults. Athletes who are in the early stages of training generally receive different training prescriptions than veteran athletes. Athletes preparing for competition should receive different training prescriptions than the same athletes reporting for their first day of training. Monitoring training dose is relatively easy; however response monitoring could easily fill a book.

It helps to have a model. There are two models that are basic assumptions of monitoring. The first model (Figure 1) deals with categorization of the various stresses that impinge on the athlete and provides a general overview of the monitoring process via dosage and response. The second model (Figure 2) deals with an assumption that training is an optimization problem. In other words, athletes can train too little and fail simply because their opponents train harder and smarter, or athletes can train too much and fail because of fatigue, injury, and illness that arise from too much stress.

Dosage. The athlete’s training demands are perhaps the easiest to characterize. Training demands are expressed as volume, intensity, etc. (Figure 1). All coaches and athletes should regularly track their training dosage. This information can be invaluable in determining how much work the athlete has accomplished and can signal when the athlete is approaching too much or too little training.

It is important to realize that it is by accumulation that training dosage matters (58). It is unlikely that a single training day or lesson will be overly demanding enough to harm an athlete in the long-term. Moreover, it is also unlikely that if an athlete misses one training day or lesson that this will upset the progress of training enough to ensure long-term consequences. However, as the training days and lessons accumulate to about a week’s worth of training then accumulated over-demands or lack of demands will likely have an impact on the long-term progress of the athlete. As such we tend to monitor training in week-long periods or "chunks." Training theory literature has named these shorter periods of training "microcycles" (12;34;36;57).

The "what" you monitor in dosage is highly sport dependent. For example, track and field throwers would likely monitor the number of throws and distances along with the typical sets, reps, weight, speed, and so forth from the weight room. A distance runner might monitor distance, speed, time, terrain, intervals, steps, and weight room factors. A gymnast might monitor the number of skills performed and total time required. A wrestler might monitor actual wrestling time along with the number of skills practiced.

Each of these dosage variables should be recorded each training lesson and placed in a training diary. Coaches should be able to prescribe training and simply record what they prescribed as the training demands. However, experience has shown that what coaches prescribe and what athletes actually do can be remarkably different. Because of this potential discrepancy, a diary of what the athlete actually does is crucial for the coach to have so that future training demands can be planned accordingly.

As a personal anecdote, while at the Lake Placid Olympic Training Center, in a lecture one of us asked all the coaches to close their eyes and then asked the athletes a simple question: "When the coach tells you to take a rest day (i.e., no training), do you?" "In other words, when the coach has planned a rest period, do you actually rest?" Only a handful of the dozens of athletes raised their hands. Most of the athletes continue to train even when the coach has prescribed rest.

On the other hand, we have all seen athletes who don’t perform all of the work that was prescribed for them. It is quite difficult to conduct a rational training program when athletes don’t do what the coach prescribed. Moreover, all of the wonderful periodization plans are worthless if the athlete doesn’t actually do the prescribed work or does more than the prescribed work. Unexpected results are likely to occur and the ability to control the training and performance of the athlete becomes impossible. At the very least the coach doesn’t know how to plan for the future without knowing what actually happened in the past. In a very practical sense, if athletes don’t rest during the rest days then they are unlikely to be able to push themselves hard during the high stimulus days and thus don’t get the necessary overload they need to make the progress they’re likely seeking.

Dosage must be monitored. If you do nothing else – monitor dosage. Dosage is the easiest to monitor, will make the most sense to the coach and athlete, and goes a long way in painting a picture that both the coach and athlete can use to modify, and thus enhance, training and performance.

Response. Athletes’ responses to training can fill several large books. Athletes produce measurable responses in physical (physiology) (1;3;6;7;9-11;13;14;19;21;23;27;28;35;48;54;55;59;60), technical (skill/biomechanics) (51), tactical (strategy) (24;26;32;56), psychological (mental/emotional) (4;17;25;29-31;37;38;49;52), and theoretical (knowledge) (53)domains. However, the vast majority of studies of athlete responses to training have been from physiology and psychology.

Responses to training stress can be found in all of the domains above, but psychology and physiology emphasize that, although quite different in orientation – either can usually detect responses to training. Responses to training are neither physiological nor psychological, they’re biological (8). Division of responses into the domains above is somewhat artificial and based on human conventions rather than anything inherent to the processes. This artificial division also points to the fact that there has been little consensus on a single marker or group of markers that are utterly infallible in detecting the status of any given athlete’s response to training.

The attempts to detect markers for overtraining have been such a case. Schiffer describes the problem eloquently, "Although researchers have suggested an impressive array of sophisticated tests to detect overtraining, the best measures continue to be the simplest changes in performance and self-rated perception of fatigue and well-being" (46), p 81. Schiffer’s quotation above emphasizes the usefulness of self-report data. However, it is our opinion that both self-report data and laboratory based tests should be used in monitoring. Both methods have strengths and weaknesses that complement each other.

Response monitoring has involved a host of biochemical tests, psychological surveys and batteries, and other measures. Variables have included heart rate, immune responses, mood states, lactate profiles, oxygen uptake measures, hormone profiles, cardiac variables, and many others. Unfortunately, the lack of consensus on which variable(s) to monitor for each athlete has resulted in interesting problems of interpretation.

There may be a lesson in this lack of consensus. What we may actually be looking for is simply an anomaly. The anomaly may come from any monitored system, physiological, psychological, and so forth. But what you’re really looking for is anything out of place.

What are you really looking for?

Let’s say you have monitoring data, what do you do with it? How do you interpret data and translate it to a training prescription? Although scientists can argue the finer points of whatever biological mechanism they are surveying, what you really want to know is whether the athlete is adapting to your training demands. Is the athlete improving, staying the same, or getting worse? Does the athlete’s current state of preparedness meet your expectations and fit within your plan? Certainly, there are times when fatigue is the goal of training, sought rather than avoided. Occasionally, the athlete may reach a peak of preparedness that is good, but to do so too early is bad. Moreover, there are times when you are not worried if an athlete is fatigued, frustrated, unhappy with his/her performance, and so forth.

There are only a limited number of things that dosage and response data can tell you: some variable is increasing, decreasing, staying the same, cycling (i.e., repeating some pattern), following another variable, and/or pulsing abruptly (i.e, suddenly changing). Figure 3 shows real data in training intensity, measured as elements (i.e., skills) per minute, over three months leading to the 1988 Olympic Trials. Figure 4 shows increasing and decreasing trends of real data from a former elite female gymnast. Also, Figure 4 shows a rather sudden change in resting heart rate data. Figure 5 shows cyclic behavior, especially during the competitive period, and sudden pulses of change in new injury per exposure records.


What you’re looking for are patterns. Patterns of long-term monitoring data, both dosage and response, can be used to identify whether training demands are going as planned and whether the athlete is adapting to these loads. Perhaps very important for our discussion here, all of the data shown in Figures 3-5 are self-report data. The athletes recorded their elements, scale weight, resting heart rate, and injuries.

A second area of interest, as demonstrated from these data (Figures 3-5) is that there is a fair amount of "noise" in the data. Data is not often nice smooth lines, athlete records of dosage, response, and performance will vary over time. In the second article of this series, we will discuss how to deal with these types of longitudinal or long-term data in terms of processing and interpretation.

One of the most important problems that coaches face with long-term monitoring data is how to make sense of the data. However, there is one aspect of the data that is collected that can be mentioned here – variation. Long-term data tends to vary due to the natural variation present in an athlete’s status from day-to-day. Data will also vary because of the error present in the measurement.

For example, even determining resting heart rate may have error due to miscounting, inability to find a strong pulse, problems with determining "half-beats" during the starting and stopping of counting, and mis-timing. We would like to minimize error variation due to testing so that the variation due to the process under observation can be seen more clearly.

It is important to determine the natural variation present in any variable used for monitoring. Figure 4 shows scale weight and resting heart rate from one athlete over more than three months. The heart rate data (pluses) is much more variable than the scale weight data (triangles). Hopkins (22) has shown that a change of about ½ the natural variation of an athlete is a large enough change to be worthwhile or worthy of our attention. How do you determine variation? Perhaps the simplest is to note the range of data values in a graph such as shown in Figure 3 and then visually estimate about half of this variation. A better means is to determine the standard deviation of the collective long-term data.
How to monitor?

Monitoring should include both training diaries, and if possible, field- and/or laboratory-type tests. The most important aspects of monitoring are to maintain consistency in assessing and recording each variable and to treat the data as longitudinal data. The best way to handle longitudinal data is to graph it. Graphs can be made directly from training diaries (Figure 6), from spreadsheet-type programs (Figures 3-5), and from specialized software and procedures (Figures 7-8). Therefore, you should keep in mind that monitoring data will ultimately need to be graphed so that interpretation is visual and easy.


Training diaries are a simple, common, and easy means of recording monitoring data. Training diaries are usually maintained by athletes with periodic reviews by the coach and sport scientist. A simple means of using a paper diary for training monitoring is shown in Figure 6. A table or matrix of possible values for each monitored variable is included so that the athlete need only put a "dot" or "X" in the column. The appropriate column is dictated by the date of the data recording. The athlete or the coach can later "connect the dots" to provide a line graph that will indicate trends in the data over time.

Figure 7 shows an elite gymnast completing data entry via specialized software using a computer that was kept in the training facility. Specialized software can increase the ability of capturing self-report data and then turning the data around for easy analysis and characterization by the coach and/or sport scientist. However, the computer system has to be more than a simple database. The computer software must also invoke artificial intelligence methods so that the computer sifts through the data for the coach, identifies discrepant data and trends, and then prepares a simple report for the coach so that the coach doesn’t have to "hunt" for relationships.

Figure 8 shows the monitoring form used for data entry by USA Gymnastics. This form was designed specifically for gymnastics and includes both dosage and response information. Although the data are all self-report, the analysis was extensive involving scanning of the forms, reduction of data, analysis of data, storage of data, and reporting of data (22;39-42;45). Analysis of the obtained data involved an "expert system" developed from a computer language called Prolog which is often used in artificial intelligence settings. The program was coded with over 200 rules obtained from the literature on overtraining and training theory regarding trends in training that should be flagged so that the coach could be alerted quickly and easily.

Figure 8. Computer "dot" sheet for recording training dose and response information for USA Gymnastics. This data form was used for several years to record national team training data. The resulting forms were mailed to a central location for scanning, storage, and analysis and then a report was returned. Today, this type of information would be obtained and turned around much faster via the internet.



Laboratory tests provide more sophisticated, more valid and more reliable data than the self-reported data in training diaries. However, laboratory tests are usually performed less frequently resulting in a lower resolution of data. Laboratory tests are generally selected based on the sport, the information sought, available equipment and expertise, and access. Because laboratory tests tend to be more sensitive than training diaries, a "test microcycle" should precede each laboratory testing session. By having a "standard" microcycle prior to testing, the coach can be more assured that the athlete is in a similar state of fitness as he/she approaches the test. Moreover, the athlete should undergo one or more tests prior to the major laboratory assessments to ensure that the athlete is not fatigue, well hydrated, and otherwise ready to produce a maximal effort.

Conclusion

Monitoring training requires that training dosage and response variables be recorded over the long-term. These variables should be recorded with the intent of determining how the athlete is performing in the time domain as opposed to comparing one athlete with another or one team with another. You’re looking for patterns in monitoring data that demonstrate that training is or is not proceeding as planned. With regard to the actual method of recording data, some methods are more cumbersome than others, and the choice of a method is likely a practical as opposed to scientific issue. Any form of data recording and storage is acceptable as long as the data are later used to enhance performance.

Dr. Bill Sands is the head of the Biomechanics Department for the United States Olympic Committee at the Colorado Springs Olympic Training Center. Bill was formerly at the University of Utah.

Dr. Mike Stone is the Director of the Exercise Physiology Laboratory at East Tennessee State University in Johnson City, TN. Mike was formerly the head of the USOC Physiology Department.


References

1. Adlercreutz H. Effect of diet exercise on hormones: Implications for monitoring training in women. Clinical Journal of Sport Medicine 1991;1:149-53.

2. American Academy of Orthopaedic Surgeons. Sports participation, injury and cost data. American Academy of Orthopaedic Surgeons, American Orthpaedic Society for Sports Medicine, www.aaos.org/wordhtml/press/boomer/sports.htm. 2000.

3. Banister EW, Wenger HA. Monitoring training. In: MacDougall JD, Wenger HA, Green HJ, editors. Physiological testing of the elite athlete. Ithaca, NY: Mouvement Publications; 1982. p. 163-70.

4. Berglund B, Säfström H. Psychological monitoring and modulation of training load of world-class canoeists. Medicine and Science in Sports and Exercise 1994;26(8):1036-40.

5. Bernard, A. B. and Busse, M. R. Who wins the Olympic Games: Economic resources and medal totals. 2002. Dartmouth University, Dartmouth, MA, National Bureau of Economic Research.

6. Bishop D. The validity of physiological variables to assess training intensity in kayak athletes. International Journal of Sports Medicine 2004;25:68-72.

7. Dare B. A review of heart rate monitoring. Track Technique 1983;85:2716-8.

8. Dishman, R. K. Biological psychology, exercise, and stress. Quest 46, 28-59. 94.

9. Flynn, M. G., Pizza, F. X., Boone, J. B., Andres, F. F., Michaud, T. A., and Rodriguez-Zayas, J. R. Indices of training stress during competitive running and swimming seasons. International Journal of Sports Medicine 15(1), 21-26. 94.

10. Foster C. Monitoring training in athletes with reference to overtraining syndrome. Medicine & Science in Sports & Exercise 1998;30(7):1164-8.

11. Francaux M, Jacqmin P, Sturbois X. Effect of training on lactate clearance. In: Ramstetter E, Zieres-Nauth C, Mack M, editors. Workshop Report Accusport. Mannheim, Germany: Boehringer Mannheim GmbH; 1995. p. 52-6.

12. Harre D. Principles of sports training. Berlin, German Democratic Republic: Sportverlag; 1982.

13. Hartmann U, Mader A. Importance of the lactate parameter for performance diagnosis and for the regulation of training in top competition athletics and in recreational sports. In: Ramstetter E, Zieres-Nauth C, Mack M, editors. Workshop Report Accusport. Mannheim, Germany: Boehringer Mannheim GmbH; 1995. p. 14-20.

14. Held T, Grossenbacher A. Improvement of the informative power of simple endurance tests by lactate determinations. In: Ramstetter E, Zieres-Nauth C, Mack M, editors. Workshop Report Accusport. Mannheim, Germany: Boehringer Mannheim GmbH; 1995. p. 57-60.

15. Hoffman JR, Kaminsky M. Use of performance testing for monitoring overtraining in youth basketball players. Strength and Conditioning Journal 2000;22(6):54-62.

16. Hogan K, Norton K. The 'Price' of Olympic gold. Journal of Science and Medicine in Sport 2000;3(2):203-18.

17. Hollander DB , Meyers MC, LeUnes A. Psychological factors associated with overtraining: Implications for youth sport coaches. Journal of Sport Behavior 1995;18(1):3-20.

18. Hooper, S. L. and Traeger Mackinnon, L. Monitoring overtraining in athletes. Sports Medicine 20(5), 321-327. 95.

19. Hooper SL, Traeger Mackinnon L, Howard A, Gordon RD, Bachmann AW. Markers for monitoring overtraining and recovery. Medicine and Science in Sports and Exercise 1995;27(1):106-12.

20. Hopkins, W. G. Quantification of training in competitive sports. Sports Medicine 12(3), 161-183. 91.

21. Hopkins WG. Measurement of training in competitive sports. Sportscience 1998;2(4):http://www.sportscie/org/jour/9804/wgh.html.

22. Hopkins WG. How to interpret changes in an athletic performance test. Sportscience 2004;8:1-7.

23. Hopkins WG, Hawley JA. Monitoring training and racing of an elite cyclist. New Zealand Journal of Sports Medicine 1989;17(1):2-4.

24. Hughes, M. and Franks, I. M. Dynamic patterns of movement of squash players of different standards in winning and losing rallies. Ergonomics 37(1), 23-29. 94.

25. Hughson S. Performance monitoring - the psychology of slalom. 1999 International Slalom Coaching Conference; Belconnen, ACT, 2616: Australian Sports Council; 1999.

26. Ishii T, Masuda T, Kurogi H, Yabuno S, Yamamoto H. The distance covered of soccer and rugby referees during the match using a mobile "GPS". In: Gianikellis KE, editor. Scientific Proceedings of the XXth International Symposium on Biomechanics in Sports. Caceres, Spain: Universidad de Extremadura, International Society of Biomechanics in Sports; 2002. p. 322-5.

27. Janssen PGJM . Heart rate monitoring for estimation of training intensity. Medicine in Sports Training and Coaching. Basel, Switzerland: Karger; 1992. p. 115-59.

28. Jones J. The value of heart rate monitor training. Strength and Conditioning 1994;16(1):36.

29. Kellmann M. Current status and direction of recovery research. In: Kellmann M, editor. Enhancing recovery: preventing underperformance in athletes. Champaign, IL: Human Kinetics; 2002. p. 301-11.

30. Kellmann M. Psychological assessment of underrecovery. In: Kellmann M, editor. Enhancing recovery: preventing underperformance in athletes. Champaign, IL: Human Kinetics; 2002. p. 37-55.

31. LeUnes A. Monitoring athlete mood using the Profile of Mood States: a case study. Applied Research in Coaching and Athletics Annual 1997;12:85-90.

32. Lothian, F. and Farrally, M. Estimating the energy cost of women's hockey using heart rate and video analysis. Journal of Human Movement Studies 23, 215-231. 92.

33. Mackinnon L, Hooper S. Training logs: an effective method of monitoring overtraining and tapering. Sports Coach 1994;17(3):10-2.

34. Nadori L, Granek I. Theoretical and methodological basis of training planning with special considerations within a microcycle. Lincoln, NE: National Strength and Conditioning Association; 1989.

35. Olbrecht J. The effective monitoring of training with science. Fort Lauderdale, FL: American Swim Coaches 31st Annual World Clinic; 1999. p. 115-21.

36. Plisk SS, Stone MH. Periodization strategies. Strength and Conditioning Journal 2003;25(6):19-37.

37. Raglin, J. S. and Morgan, W. P. Development of a scale for use in monitoring training-induced distress in athletes. International Journal of Sports Medicine 15(2), 84-88. 94.

38. Rushall, B. S. A tool for measuring stress tolerance in elite athletes. Journal of Applied Sport Psychology 2(1), 51-66. 90.

39. Sands, W. A. National women's tracking program pt. 2 - response. Technique 10(1), 23-27. 90.

40. Sands, W. A. Monitoring the elite female gymnast. National Strength and Conditioning Association Journal 13(4), 66-71. 91.

41. Sands, W. A. AI and athletics. PC AI 6(1), 52-54. 92.

42. Sands, W. A., Henschen, K. P., and Shultz, B. B. National women's tracking program. Technique 9(4), 14-19. 89.

43. Sands, W. A. and McNeal, J. R. Predicting athlete preparation and performance: A theoretical perspective. Journal of Sport Behavior 23(2), 1-22. 2000.

44. Sands WA, McNeal JR, Stone MH. Plaudits and pitfalls in studying elite athletes. Perceptual and Motor Skills 2005;100:22-4.

45. Sands, W. A., Shultz, B. B., and Newman, A. P. Women's gymnastics injuries. American Journal of Sports Medicine 21(2), 271-276. 93.

46. Schiffer J. Overtraining. New Studies in Athletics 2001;16(4):79-122.

47. Shermer, M. Exorcising LaPlace's demon: chaos and antichaos, history and metahistory. History and Theory 34(1), 59-83. 95.

48. Smith DJ, Norris SR. Training load and monitoring an athlete's tolerance for endurance training. In: Kellmann M, editor. Enhancing recovery: preventing underperformance in athletes. Champaign, IL: Human Kinetics; 2002. p. 81-101.

49. Steinacker JM, Lehmann M. Clinical findings and mechanisms of stress and recovery in athletes. In: Kellmann M, editor. Enhancing recovery: preventing underperformance in athletes. Champaign, IL: Human Kinetics; 2002. p. 103-18.

50. Stone MH, Sands WA, Stone ME. The downfall of sports science in the United States. Strength and Conditioning Journal 2004;26(2):72-5.

51. Strand BN, Wilson R. Assessing sport skills. Champaign, IL: Human Kinetics; 1993.

52. Taylor SR, Rogers GG, Driver HS. Effects of training volume on sleep, psychological, and selected physiological profiles of elite female swimmers. Medicine and Science in Sports and Exercise 1997;29(5):688-93.

53. Thomas, K. T. and Thomas, J. R. Developing expertise in sport: The relation of knowledge and performance. International Journal of Sport Psychology 25, 295-312. 94.

54. Tonkins, Jr. W. P. Analysis of the relationship between exercise capacity and heart rate variability in trained and untrained individuals. 99. University of Memphis, Memphis, TN, University of Memphis.

55. Troup J. Monitoring performance with an applied approach. Las Vegas, NV: American Swim Coaches Association; 1983. p. 95-110.

56. van der Mars, H. and Darst, P. W. Improve your coaching environment through behavior analysis. The Journal of Applied Research in Coaching and Athletics 3(4), 262-273. 88.

57. Verkhoshansky, Y. Main features of a modern scientific sports training theory. New Studies in Athletics 13(3), 9-20. 98.

58. Verkhoshansky YV. Programming and organization of training. Moscow, U.S.S.R.: Fizkultura i Spovt; 1985.

59. Viru A, Viru M. Biochemical monitoring of sport training. Champaign, IL: Human Kinetics; 2001.

60. Viru M, Viru A. Monitoring of training. Track Coach 2002;Fall(161):5154-5.