As promised, today begins a series of posts on coaching and science, and how the science can be, should be, and sometimes is, and often is not, applied to athlete preparation. Obviously, it comes with an endurance focus, but there's no reason why sprint coaches and team sport coaches can also not glean some information from this.
This is a series that was inspired by my visit to the US Olympic Center in Colorado Springs. I was lucky enough to be invited there by Prof Randy Wilber of the USOC, who had organized a symposium on altitude training. The symposium brought together scientists, coaches, athletes and mangers from 22 different countries, and included 32 Olympic athletes, and numerous sporting codes, Summer and Winter Olympics among them.
An uneasy marriage?
So it was a symposium that created the perfect platform for the marriage of science and coaching. Yet, two days later, and I don't know that too much had been gleaned from the science part of the symposium. At the risk of dismissing its value, the coaches probably gained enormous value from other coaches' presentations - Bob Bowman (Michael Phelps' coach) would have heard some valuable tips from Terrence Mahon (Ryan Hall and Deena Kastor, among others), and vice versa, but I dare say that the science would not have changed the way any of the coaches are approaching altitude training for their athletes. They already had a strategy, and I doubt whether the science showed them anything to improve or change it.
And this is a typical problem - coaches have historically done things well in advance of the science proving (or disproving) that they work. The famous running coach, Arthur Lydiard, was once reported to have said that the coaches already know what works, and the scientist's job is to tell them why it works! Now, that's not always true, because often, coaches do the wrong things, or they fail to do things that would help, or they are unaware of methods to improve performance. So it's not quite as simple as saying that science chases after the truth long after the coach has discovered it. Lydiard said this long before much of today's technology was available, and perhaps his view would now be different
But the eternal questions remain:
- What value does science add?;
- How much science does a good coach need to know? and;
- Is the science sometimes more of a liability than an asset?
One person's view: heart rate monitors + books = science
And co-incidence luck would have it, I woke up this morning to discover LetsRun.com's quote of the day was the following, from Steve Jones:
"What I do is make it simple," Jones, now a teetotaller, says. "There's no science in it – no heart-rate monitors. It's just running – running instinctively. Anyone who saw Steve Jones run in the Seventies, Eighties and early Nineties knew that he ran by the seat of his pants nearly all the time. You don't see that any more and that's what I'm trying to teach these guys. None of it comes out of a book. It all comes out of my own experience."Quite clearly, Jones thinks that science is the sum of heart rate monitors and books. And no doubt, some will agree. I don't. My definition of science, particularly as applied to the coach, is far broader than this, and I dare say that Steve Jones applies a great deal of science to his coaching without even realizing it.
Our purpose and these questions
So what then does this application look like? To begin, the video below is an interview that I did with Training Peaks a week ago, in Boulder, Colorado. For those who don't know, Training Peaks make software to allow athletes and coaches to manage and monitor training. The management of data is, as I mention in the interview, a fundamental aspect of the scientific process applied to training. I'll describe this in a lot more detail in the next post.
If you've not seen Training Peaks software, I really do recommend that you check it out, because it offers so much value that if your big obstacle is managing your training (or that of your athletes), it might be the solution, particularly if you have a large group and you want to keep tabs on many at the same time.
However, in the interview, at around 5:40, Dirk Friel (Training Peaks Chief Marketing Officer, former pro cyclist) asks about coaching, and my answer is really the key to understanding what I believe to be the scientific value to coaching.
Also in the interview are some thoughts on how we began The Science of Sport, and some of the good stories of 2009, but that's really just for interest's sake - the real content is from 5:40 onwards!
Science is the process, not only the theory
To sum it up in a nutshell, there are really two aspects to science for the coach:
First, there is the theory. How the body adapts to stress, overtraining, environmental factors, diet, hydration, biomechanics, cardiovascular system and nervous system, and so forth. Steve Jones would call this the "book", and many coaches would be put off from looking it up, since this is the domain of the white-coated academic in the lab. The better coaches, in my opinion, seek to take from this theory and apply it, and we'll discuss that in more detail soon.
However, the second part of the science is, in my opinion, even more important. It involves the ability of the coach to engage in a scientific process. And what is a scientific process? It is the process by which you ask a question, design an 'intervention' (in this case, a training programme), create a hypothesis (for example, the athlete will run a 5km PB in three months), and then measure and collect data. If the data suggest all is on track, then the coach continues. If it suggests the athlete is not adapting, then the plan is changed, and a new intervention begun.
That's science. Coaching is the application of the same process as a scientist would follow if they wanted to test whether a transcription factor was involved in regulating the activity of a gene. There are differences, sure, but the process of measuring the athlete's response to the training, and then adapting the training to maximize the response requires the same system of thinking, and that is why the best coaches are also "scientific-minded". The interview above hopefully clears that up a little more.
So join us next time when we look more specifically at what scientific theory is valuable for coaches, and how this process should actually work. I think you'll find that good coaches do all of this already, and for them, the improvement in coaching ability comes from tiny adjustments, also made with some science behind them!
Ross
Very interesting, and as always thought provoking.
ReplyDeleteThrough many years of learning about physics and philosophy, then teaching physics, chemistry and mathematics; and of trying to compete more effectively as an adult in running races from 3k to the marathon, I've been obliged to try to wrestle with the differences between empiricism, science and technology, and of course sports coaching.
Space prevents my full response here, but here's my take on empiricism and science (leaving technology and coaching aside for the moment) after the last forty years or so of ponderings on this and related matters:
Empiricism is the philosophical school which in this context, in a nutshell, argues that the collected data is paramount. When I was young and studying physics at school and university, I confess that I was very much a theoretician and not an empiricist. In fact, I intensely disliked lab work, as much for the straightforward reason that the experiments usually failed to work out the way the way they were supposed to! (A common problem for school and undergraduate physics the world over I suspect). Only much, much later did the importance of the raw data slowly dawn on me- data which could be transformed into a chunk of information, which in turn could be, sometimes, massaged further to add to, or confirm one’s stock of “knowledge”, or most excitingly suggest that the prevailing paradigm was wrong – see Kuhn etc for more on this.
So-called "scientific objectivity", I realised eventually, is really inter-subjectivity: the property of any given, well conducted experiment, which tests the truth value of an “If ... then” statement (a hypothesis), which allows the experimental data gleaned from that experiment to be reproduced again and again, by different observers at different times and places, within the experimental uncertainties inherent in the collection of any data set.
It’s of course quite easy to be scientific in this sense in physics. It’s much less easy to be scientific, in the definition I have suggested, in any field of enquiry that involves animals, including Homo. This is in part because once one tries to experiment on animals, there is an ethical dimension that the physicist usually doesn’t need to worry about. Secondly, reproducibility of the gathered data can easily become problematic with Homo and other animals. For example, whilst it might be fairly easy for an experimentalist to ask Bolt to run say, ten, 100m repeats each in say between 11.00 and 12.00 seconds, with a fifteen minute recovery between each one, it becomes somewhat trickier were Bolt to have to run the same number of repeats between say 9.8s and 10.00s under the same conditions.
Consequently, I would argue that sports science is not science in the same sense that physics is science. Rather, I would argue, even at the most fundamental level of sports science research, there are all sorts of methodological issues affecting the inter-subjectivity and repeatability of the raw data, that thankfully, physics researchers don’t have to think about!
There are other issues too to do with the over-simplistic way that data is often reported. By this I mean that every datum has a spread of values – an error that is the datum’s uncertainty. The uncertainty for each datum is estimated from the data and from the limitations inherent in one’s data collection equipment and from the limitations inherent in the human observer. In physics, the uncertainty in each datum is of CRUCIAL SIGNIFICANCE to the significance of the experimental outcome.
Few outside experimental physics recognise the importance of estimating the uncertainty for each and every datum in one's data set. Certainly, none of the many chartered engineers or the mathematicians of my acquaintance have ever been able to “get” the importance of plotting not discrete points in one’s graphs, but rather horizontal and vertical error bars through which one then (before computers, subjectively) draws one's best curvilinear lines (not line) of fit.
I wrote:
ReplyDelete"Certainly, none of the many chartered engineers or the mathematicians of my acquaintance ..."
That's an exaggeration, of course. Should more precisely have written:
"Certainly, not many of the many chartered engineers or the mathematicians of my acquaintance ..."
Dr. Tucker,
ReplyDeleteGreat post about a crucial issue. I'm looking forward to the follow-up threads.
In my experience, the key to evaluating information - and more broadly, to improving practice through science - is to be disciplined enough to first consider the right questions; and then take a "best practices" approach to finding the answers. Best practices, in turn, takes us to a place that's principled as well as evidence-based - two extraordinary things to be.
This implies a few things that help put science into perspective for practitioners:
* Spell out a set of principles that you have enough conviction in to write down, post in plain view for all to see, and explain in terms a child could understand.
* Consider the body of evidence.
* Embrace the concept of levels of evidence.
That's the short version. Here's a longer one, from an article along the same lines posted on our site last month:
http://excelsiorsports.blogspot.com/2009/09/science-professional-practice.html
Best regards,
Steven Plisk
Excelsior Sports
Shelton CT (USA)
I believe science and empiricism have to go hand in hand to get the most progress. At times, science can evaluate the best input parameters to obtain a certain result. But other times, because of the willingness of a coach to try unexplored possibilities, empiricism can end up suggesting science in what direction to look. It's up to both coaches and sport scientist to get the best out of both.
ReplyDeleteScience hardly ever "proves" that one thing is better than another when it comes to coaching. The area is so complex that one study hardly ever is so good that everyone says "boy, I need to start doing that."
ReplyDeleteHowever, it seems to me that one area that science can have a real and immediate impact on coaching is if research shows that something the coach is doing is detrimental to performance. Time is the enemy of every athlete and if a coach can be shown that with his practices he is likely wasting athlete's time or substantially increasing injury risk for no performance benefit, he is likely to say "Boy, I need to stop doing that."
Very interesting and thought provoking..... with scientific training... the sportsmen would do a lot better....it seems to me that one area that science can have a real and immediate impact on coaching. keep posting. Will be visiting back soon.
ReplyDeleteRoss,
ReplyDeleteThanks for posting this. I've had the opportunity to work in a number of settings where scientists and coaches are supposedly working together for the benefit of the athlete. In reality it's usually an unhappy marriage!
For mine, the issue is a partially legitimate pride and prejudice on the part of both parties:
The scientists are prejudiced against the coaches because their conclusions are based on gut feelings that come from general observations over a long period of time.
The coaches are (legitimately)prejudiced because the scientific perspective comes from making conclusions based on studies that work with folks (sometimes 'real' athletes, often not) over very short periods of time in the context of real world athlete development.
Neither of these are likely to change with the financial and bureacratic constraints that come with scientific studies but this is a problem not of the scientific method but of the nature of how science is supported. When it comes to application of the scientific method, the best coaches in the world are as accomplished as any 'white coat' out there :-)
Best regards,
Alan Couzens, MS (Sports Science)
Exercise Physiologist/Coach
Endurance Corner
alan@endurancecorner.com
http://www.endurancecorner.com/ac_blog
For a practicing athlete, solutions come from an eclectic mix: the empirically derived ideas of Arthur Lydiard, which he developed by experimenting with training his own body; and the (very occasional) scientific study that employs actual training methods and evaluates the results in terms of performance. E.g., the study that showed intervals improve 10K race times far better than tempo runs.
ReplyDeletePossibly the most interesting experiment in the world of elite running concerns aerobic metabolism, i.e., the Africans whose success is based on years of running up to 100 miles per week aerobically, in _elementary school_. This is Lydiard-style training. The best runners in the world train empirically; science merely tries, with very mixed success, to explain why it works.
For the individual athlete, the lab is the body, and the gauge of the "rightness" of training is the instrument by which the body speaks to us: the calm, intuitive feelings of the heart.
Could'nt it be that the good coaches get new experiences (empiricism) from trying the theories that science shows could work. Would'nt this be a relationship between science and empiricism coaching. I agree that a good coach thinks (should think) like a scienctist though my lack of experince with coaching make my ideas just ideas, not theory nor experienc.
ReplyDeleteMartin