Sport Science advice must be unbiased.

 

Imagine you are tossing a coin. Suppose you got tails five times in a row. Now you are preparing for the next throw. What do you expect more: heads or tails?

It is forgivable if you answered “heads” because we may intuitively think that if just one outcome from different possible events repeats it cannot last forever and a probability that such outcome will end increases with every subsequent throw. However, the right answer is that probability remains 50/50 independently of what happened in previous throws. The coin has no memory to remember how it fell before. Of course, the coin must be fair https://en.wikipedia.org/wiki/Fair_coin. Yes, it is forgivable mistake to think otherwise if you are working in a non-scientific field, but is it acceptable for a scientist?

Now have a look at following citation: “I am aware that the All Blacks have not been beaten there since 1996. (In statistical terms, this only means that the probability that the All Blacks will lose there becomes increasingly more likely with each successive game that they win. Clearly it is impossible that the All Blacks will never ever lose another game at Eden Park. So the only reality is that the probability that they will lose their next game at Eden Park increases with each new game. Not the opposite, as some would want you to believe.)”

The author of the citation is Tim Noakes, renown sports scientist from South Africa. This paragraph is from a letter which he wrote to the South Africa rugby team during their preparation for the challenging game against invincible All Blacks from New Zealand. I found it in his book “Challenging Beliefs: Memoirs of a Career.” Tim wanted to inspire his countrymen, and since he is a very respected scientist, he brought this “scientific” argument to support South Africans winning hope. However, scientifically it is a nonsense. Probability in this case, similar to the coin toss, is independent of past games.

I deliberately pick up this Tim Noakes citation because I greatly respect him. I respect Tim as an athlete who ran numerous super marathons, and I appreciate him as a sports practitioner who worked with the great coaches and athletes. I can say that his scientific philosophy and views are often very close to mine and he is more scientist than I am. So should I carp at this silly mistake? Should we demand, for example, from a professor of History or Literature knowledge of basics in Probability Theory?

Well, perhaps not. Maybe sports scientists should know statistics better than historians because they very often use it in their papers but, at the end of the day, if they manage without such knowledge — it’s ok. Actually, Tim Noakes honestly admitted in his book that he is weak in math and stats and that he is always trying to have a team of people around him who can help with this.

Fair enough, I have no problem with that. However, I don’t like the tone of this citation. I don’t like the words: in statistical terms, clearly, the only reality. I don’t like this unreserved overconfidence, especially from the man whose scientific credo is “ Challenging beliefs”. Did he demonstrate how overconfidence coupled with ignorance easily create another erroneous belief?

Tim Noakes is an apologist for some scientific concepts which sometimes go against generally accepted theories. Often I support him. I almost completely agree with his views on water replacement strategy ( see article). I think his Central Governor theory makes a lot of correct conclusions. Maybe I am, like many nutritionists, sceptical about his support of “Low- carbs” diet but I have to honestly admit that I am not a prominent specialist in nutrition https://en.wikipedia.org/wiki/Low-carbohydrate_diet.

However, what if his level of overconfidence and bias in these theories is the same as it was in “flipping coin problem”? Reading his book, I have a feeling that if something goes against his assumptions ( like, for example, that game which South Africans lost), he tends to explain this by conspiracy theories rather than trying to review his conceptions. Perhaps he, like many other revolutionaries, started as a fighter against old beliefs but, eventually, becomes a passionate defender of the new dogmas.

Nevertheless, in this post, the question is not about Tim Noakes theories. Moreover, it is not only about him. It is about Sport Science in general. In my opinion, the most dangerous trap for a scientist is a bias in favour of the theory that he/she created and ignoring objective reality. Unfortunately, it very often happens in Sport Science. And if even Tim is not immune from that what we can say about thousands less scrupulous scientists?

Many sports practitioners, athletes, and coaches believe scientists and follow their advice. These “field workers” probably have no time for detailed analysis and critical approach. So, they trust scientific theories, especially if such theories packed in a beautiful wrapper and promise a great result. Unfortunately, a nice wrapper does not mean good stuff. Sport scientists should improve a genuinely scientific quality and ethical standards of their recommendations. Biased coin produces unfair flipping.
.

Peter

Written by: Peter