As a note on nutrition and so called nutritionists.

Much of the solid research about food in general and health has been in terms of the possible effects on chronic diseases (cardiovascular disease-CVD and the cancers mostly). Food is extensively studied because it feels like something we can control. But it's also quite hard to measure accurately. One Swedish professor told me that the obesity epidemic per year is such a small imbalance of energy intake that it could be explained by a person have three cups of a tea a day for a year with a teaspoon of sugar in each. Even that is a very hard thing to get accurate reports about.

Almost all of this research has been observational studies (which are not as good as experiments). Some of these observational studies have shown things like food with high levels of antioxidants are better for you because they reduce the risk of CVD. But when you experimentally give antioxidants to humans in a large and long term experiment antioxidants may actually increase the risk of death (Bjelakovic and others, Journal of the American Medical Association (JAMA) 2007). It's immensely complex and some foods can be shown to be good for some conditions and bad for others.

Now lets bare in mind that all of this biomedical research has many hundreds of thousands of people in it. It costs millions to do properly and you have to have highly trained scientists and clinicians to design and run the experiments or studies properly and then analyse and interpret them correctly.

This is where the rubber hits the road. What do you tell people to do and who do you get to do the telling? Because lets not forget that if food can make you healthy the wrong advice could also make you sick.

Dieticians are highly trained health care professionals who deal with diet issues in people with disease. They are a basically trustworthy profession but they tend to deal with sick people and not athletes- their might be some around who specialise in sports people though.

Nutritionists run the gamut from world leading researchers and clinicians who really know what they are on about to people who have read a few pages on the internet and hung out a shingle. Sometimes they might even have a dodgy diploma from a private training establishment to back them up.

So beware! There are some really really bad ones. Some of these really really bad ones write books. Writing books is not a sign that you're well respected and know what you're on about. It can be a sign that you're making stuff up and can't get published through the usual channels.

You'll note that I haven't yet mentioned anything about general principles in diet and performance in highly tuned athletes? Why? Because I don't know anything. I imagine that the problems are this. 1. There's no money to do it properly. 2. Sports science is often not very scientifically attuned- they have issues getting the right study design in my short and limited experience. 3. If you're doing research in elite athletes the problem by definition is that there isn't a lot of them (small sample size). 4. As they are elite there's proboably not a massive improvement that you can cause by tinkering with their diet (ceiling effect or improvement to an asymptote AKA the learning curve). So it's really hard to show that performance is improved by particlar dietary interventions. It may also be that diets that are recommended in certain circumstances actually make your performance worse.

If you want a book I might recommend the CSIRO diet book. It's written with the best available scientific information in mind and with minimum distortion but is still correctly criticized on some levels. But again it's designed for weight loss and disease pervention- not for optimum performance. And as Carl has already mentioned performance is a very hard thing to measure- especially in underwater hockey.

PS. I've flatted with Liam. His diet is best described as eccentric.