February 24, 2016

Nutrition for a marathon – what should you eat?


This was supposed to be a short post about diet for endurance events like a marathon. Typically for me though, I’ve ended up diving pretty deep into the topic. If that’s your ‘jam’ then read on (lucky you), but if you’re wondering whether it’s worth any more of your time, here’s what’s covered:

  • The importance of nutrition in being able to run a marathon.
  • What actually happens when you hit the wall/bonk and how you can try to avoid it.
  • Why anyone running on energy drinks/gels should think very carefully about how they use them.
  • Whether it’s possible – or even beneficial – to run a marathon without taking on any carbohydrates?

Tip: Want to geek out on this subject as much as I have? Start by familiarising yourself with all the different ways that the body produces ATP – our energy currency. Without my uni textbooks to hand I’ve been referring a lot to “Your Inner Engine: An Introductory Course on Human Metabolism” by Jane Vanferkooi (it’s relatively accessible given the topic and, well… Vanferkooi!).

Food or fuel?

For any sport or exercise, diet has a meaningful effect on your ability to train, recover and perform. When it comes to events as long as a marathon, things get a little more serious. Just getting to the finish (let alone in your target time) will be difficult if you fail to fuel properly before and during the race.

As a result, at a time of year when lots of people are giving up certain foods, my concern has been about whether I can eat enough of them! As someone with a – shall we say – ‘generous’ appetite, this isn’t usually an issue for me. Unfortunately, when food is fuel and it’s got to keep you running for over 26 miles, just being able to eat a lot doesn’t solve all your problems.

Dont get ‘hangry’

Before we go any further, I’m aware that conversations about diet & nutrition can get pretty heated. So let’s start by making a general point really clear…

The ‘best’ diet depends heavily on what you’re optimising for. Even then, two physically similar people with a common goal (fat loss, muscle gain, cancer prevention, mental alertness, keeping warm in the winter!? etc.) may react differently to eating exactly the same food.

Despite this, I’d suggest ALL the best diets share a few common components: a variety of quality, nutrient dense whole foods with anything heavily processed kept to a minimum. Pretty basic.

As long as those bases are covered, all other choices are just about figuring out what works for your lifestyle, ethics or goals. There’s a chapter dedicated to this view of nutrition in Unbreakable Runner, but it’s by no means unique. Both Food Rules: An Eaters Manual by Michael Pollan, and a recent post by David L. Katzpretty, do a great job of explaining and summarising these common sense (but less headline worthy) principles.

With that in mind then, this post is focussed specifically on the challenges a marathon poses from a nutritional perspective, and what options we have to try to solve them. In other words:

What is the best diet to optimise performance when training for and running a marathon?

Right, now that’s clear let’s contradict myself and kick off a huge fat vs. carb debate…

Fancy a bonk?

Alongside minimising the chance of injury, a key aim of the way I’m training is to delay the muscle damage that comes with running 26.2 consecutive miles. All the muscular strength in the world is useless however, if you run out of fuel to keep them working; enter the ‘bonk’. Not that sort (mind out the gutter please – this blog is for charity), the sort that tends to affect marathon runners by about mile 20. Also known as ‘hitting the wall’.


Ready and waiting for you to run into…


So what exactly does it mean to hit the wall/bonk? There’s a combination of things going on, but in basic terms it’s glycogen depletion – your muscles running out of fuel.

The science bit (complete with shaky maths)

The amount of glycogen you can store varies from person to person but the maximum is often stated to be around 500g or 2500 calories (kcal) worth. As for me? Very difficult to tell without lab tests, but my best estimate is 1,820 kcal (I won’t bore you with how I got to that figure, but the key is knowing your fat free mass).

Whatever the actual figure is however, the average runner will burn closer to 2,600 kcal over the course of a marathon. So how do you make up for this deficit? Well, there’s a few factors that influence when your glycogen reserves run out, and so a few strategies you can employ to avoid bonking/hitting the wall:

  1. Re-fuel with simple carbohydrates during the event.
  2. Significantly increase aerobic capacity (VO2 Max) and efficiency (oxygen & energy required to sustain a given speed).
  3. Train the body to use a different energy source.

Let’s look at each in more detail…

Option 1 – Carbo-load

This is the approach almost universally recommended for endurance athletes; take on as many carbohydrates as possible before a long run then use energy gels/bars/drinks to top up when glycogen stores start to dwindle. Makes a lot of sense, and there’s a wealth of evidence demonstrating that it helps. There are however, a few problems:

Based on my weight and 3:30 target time the runners world calculator estimates I’ll actually end up burning through closer to 3,000 calories over the course of a marathon. As a result, even if I carbo-loaded my face off, avoiding complete glycogen depletion – and a no hands face-plant into the wall – is going to require somewhere in the region of 500-1,000 kcals during the race, possibly more.

What does that look like in practice? Well, a typical energy gel is 45g and provides about 100 kcal so I could need as many as 10 during the race. Aside from whether I could keep down that much sugar whilst running (!), there’s also an absorption problem. The most optimistic estimates suggest you can absorb 1.7g of carbohydrate per min – or about 100g per hour. Any more, and your intestines just don’t have the capacity to transport it into your blood stream. This gives me a theoretical limit of 700 kcal of usable carbs I could take on during the race, equivalent to an energy gel every 30 mins. Possibly enough, but no guarantees.

Assuming I could take the large amount of extra carbohydrate on board (did I mention I’m pretty good at eating?), the other thing to consider is how it would actually make me feel during the race. Ingesting simple carbohydrates gets glucose to your muscles quickly (good times), but the accompanying insulin response may cause peaks and troughs in energy levels (bad times). Perhaps more importantly however, insulin also inhibits the breakdown of fat into fatty acids (Coyle EF, 1995) – the downside of which will become apparent in the next option.

Option 2 – Turn down the burn rate

This strategy is essentially just ‘get fitter’. Sounds obvious, but lets break down what that actually means in terms of energy requirements.

The body is fuelled by two main sources; carbohydrates and fats (a simplification but for the purposes of this post accurate enough). How much of each is used depends on a range of factors, including what is available & the total energy requirements at any given time. It also varies between individuals, but will usually look something like this…

So at rest fat typically supplies around two thirds of total energy requirements – the remainder coming from carbohydrates. As we become more active however, this ratio (called your Respiratory Quotient or ‘RQ‘) starts to change. At some stage a cross-over point is reached where the body is using equal amounts of carbohydrates and fats (Wolfe, 1998Brookes & Mercier, 1994). Carrying on increasing intensity past this level results in carbohydrate becoming the predominant fuel source, and eventually the body is entirely reliant on them to meet energy demands.

In basic terms then, the higher the exercise intensity the more dependent muscles become on carbohydrate for energy. The reason for this relationship? Again, it’s pretty nuanced but one of the factors is the way the body metabolises different substrates. Whilst fat contains more calories than carbohydrate (9kcal per gram vs. 4kcal per gram), it takes longer to liberate that energy. As a result, carbohydrates are the preferred option when energy requirements are high.

Despite triggering a whole range of physiological adaptations, one of the outcomes of training is the ability to use more fat at a given intensity (Holloszy & Coyle, 1984Kiens et al., 1993). In other words, exercise can result in the crossover point (switch from fat to carb use) moving to the right, sparing glycogen. This is a vital part of what ‘fitness’ actually means when it comes to endurance events.

Here’s where it gets interesting…

Given the limited amount of glycogen you can store, the hotter you’re ‘burning’ the faster those reserves get used up. This highlights one of the possible downsides of using simple carbohydrates before or very early in the race: it may increase the chance of you hitting ‘the wall’.

Why? If a rush of fast absorbing carbohydrates causes a strong insulin response, your ability to draw on fat as an energy source is reduced, leaving you more carb reliant. In essence, you risk losing some metabolic flexibility. So by all means, if you’ve experimented with pre-race energy gels & it works for you then go ahead. Just be aware that it may actually raise the total amount of carbohydrate you need to take on during the race, to avoid running out of fuel.

From a fitness perspective then, there’s two areas that I can focus on to try and delay the point at which I run out of fuel:

  • Increase aerobic fitness such that running at my target pace requires an effort below the point where carbohydrates become the main energy source.
  • Train my body to use more fat for any given level of effort.

The problem with relying on these to guide how I fuel long runs is that without a friendly sports science lab I can only estimate baseline data such as V02 max and RQ. Even with this data, I still don’t know if it would be possible to improve any of the markers enough to significantly change the amount of fuel I’d need to take on during the race.

So, improving aerobic capacity & efficiency helps preserve glycogen by training your body to use less of it for a given intensity of exercise. Despite this, it seems carbohydrates will always be a limiting factor whilst you’re relying on them to keep you going…

…which brings us on to the third approach.

Option 3 – Kick in the ketones

If you can’t guarantee avoiding ‘the wall’ through carb supplementation or training, could diet help shift the balance? Possibly, but it’s tad controversial.

Whilst glycogen stores are limited, fat reserves are almost limitless (even the leanest people store tens of thousands of kcal). The theory goes therefore, that if we could continue to use fat as the primary energy source, running out of fuel no longer becomes a major factor in how long we can keep going.

So how do you go about using diet to increase the contribution fat makes to your energy requirements? The most effective way is to get into a state called nutritional ketosis. This however, requires eliminating nearly all carbohydrates and so runs completely counter to the years of evidence and advice regarding diets for improved athletic performance. Hence the controversy.

First of all then, let’s start where people tend to agree.

What is ketosis? 

For an in depth, professional explanation it’s worth reading this post by the keto-curious (and none too shabby athlete) Dr Peter Attia. In brief however, here’s my attempt…

Under normal conditions (fed, at rest) energy requirements are met by a combination of fatty acids stored as triglycerides, and glucose stored as glycogen. Not all cells can use fatty acids to produce energy however, so if glycogen levels drop significantly the liver responds by creating a group of molecules called ketone bodies (a process called ketogenesis). In all but a few cases ketones can be used as an alternative to glucose, so under extreme or sustained glycogen depletion, the rate of ketogenesis goes up. Nutritional ketosis begins when ketones are produced faster than they are used – typically defined by having circulating blood ketone levels >0.5 mmol/L.

With this in mind, it’s also fairly widely accepted that ketogenic – or more broadly low carbohydrate, high fat (LCHF) – diets can increase the contribution fat makes to overall energy requirements. One recent study even suggests that we may have underestimated the extent to which we can use fat as a fuel source during exercise (Volek et al., 2015).

Where the evidence is less clear, is whether this actually improves performance in endurance athletes (Burke et al. 2002Hawley & Lecky, 2015). As a result, it’s difficult to make a judgement on whether it’s something the average marathon runner should think about pursuing. From a very limited perspective then, here’s my take:

The research to support ketosis being advantageous for endurance athletes is incomplete. Despite this, there’s a growing number of rigorous self experimenters, elite race winners, and endurance record holders successfully using ketogenic or low carbohydrate diets. Clearly a few n=1 experiments don’t prove that it will be beneficial for everybody. It does however, give cause to question whether there might be an alternative to the high carb diets so widely recommended.

When it comes to benefits of high carb diets, there’s no shortage of research outlining how and why they help athletic performance. Where the evidence seems weaker, is in studies that suggest LCHF diets are not beneficial (or are even detrimental). Here, issues including relevancy of study design for endurance events and funding links to sugary sports drinks manufacturers make it more difficult to assume the conclusions are applicable.

So the potential benefits of a LCHF diet for endurance events are certainly intriguing. As touched on earlier though, it does require a fairly drastic change in diet.

The diet sacrifice

To guarantee getting into nutritional ketosis, you really need to limit carbohydrate intake to somewhere in the region of 30g-50g day. For context, it turns out a pint of beer – a reference point for much of my life – has about 13g. This means even a relatively sensible night out (with no other carbohydrates for the rest of the day!) would drop you straight back into carb burning mode. “Hang on, why don’t you just switch to drinks with less carbs? And should you be drinking at all whilst training!?” Well, yes I could and no I probably shouldn’t, but lets be realistic here.

What’s more, a process called gluconeogenesis allows glucose to be created from non-carbohydrate sources. This is an integral part of how your body maintains blood glucose levels and so happens continuously at a low level. It’s thought however, that even a high protein meal may allow your body to ‘make’ enough glucose to drop you out of ketosis. The experimental evidence isn’t conclusive on this, but as a result most ketogenic diets recommend eating predominantly fat (75%-80%), with moderate protein (15%-20%) and very low carbohydrate (5%) intake. There’s a few tools online such as the calculator over at KetoDiet Blog if you’re interested in examples.

The real problem with adopting a strict ketogenic diet in my case however, isn’t a penchant for alcohol or cake (honest), but how suitable it is for the way I’m training. The Crossfit Endurance approach involves a lot of days training at high intensity, and there’s a consensus that your ability to work near maximum effort is reduced compared to when carbohydrates are the primary fuel source. Whilst a strict ketogenic diet may offer an endurance advantage on race day then, it seems it’s also likely to compromise the effectiveness of my training. As such, it doesn’t seem like a viable option.

A fourth option?

So I’m not going to scrap carbohydrates completely. But is it possible to get some of the implied benefits of a LCHF diet (reduced glycogen use, more stable energy, less fuel required during the race) without actually being in ketosis?

To maximise fatty acid utilisation whilst exercising, you almost certainly need to be in nutritional ketosis. There is research however, that suggests being on a LCHF for diet for as little as 5 days results in an ability to up-regulate fat metabolism – even after reverting back to eating higher levels of carbohydrate (Raper et al. 2014Yeo et al. 2011, Stellingwerff et al. 2006). 

Due to the possible performance benefits associated with this adaptation, I’m interested in improving my ability to use fatty acids during exercise. In essence, the aim is greater metabolic flexibility.

With this in mind, I plan to experiment with a low carb, high fat approach over a limited time span (5-7 days). If successful (i.e. it’s not horrific and I still feel able to train) I may then try using a 24-48hr fast and/or synthetic ketones to enter ketosis without the slow adaptation period (Cox & Clarke, 2014). This all means I’m yet to decide on a nutrition strategy for the actual marathon. We’ll see how things pan out before I commit to that!

Hight fat vs. high carb – do we have a winner?

In a word, no. Stay with me though…

In some cases eating lots of carbohydrates is undoubtedly a good approach. What I’ve become sceptical about, is whether it should be put forward as THE solution for every runner. The fact that most people will have heard the term ‘carbo-loading’ hints at how ubiquitous the advice has become. Just because it’s right however, does that mean any other option is by default wrong? Much like any training approach, it seems short sighted to suggest there’s a definitive answer. For example, the diet that improves a sub 3 hour club runners performance may not work for a rhino costume wearing charity runner. As per my earlier point; it depends on what you’re optimising for.

So why isn’t there more people suggesting this? Well, the cynical answer is that it’s not easily actionable advice, so not very headline worthy. Which are you more likely to click on….

“Take quantitative and qualitative baseline data, experiment by substituting and adding back different foods one at a time and and monitor performance. Over time your optimal diet will become clear”


“Eat this one food for two weeks to guarantee running a PR”

Both are pretty ridiculous as headlines, but the point I’m making is that strong opinions and quick fixes grab our attention. Woolly, hedged recommendations not so much. In order to avoid defining my own post as useless then here’s my attempt at some sort of conclusion…

Choose your goals

As more research on diet and athletic performance accumulates, the ideal mix of carbohydrate & fat for endurance athletes may become clearer. Based on what we seem to know right now, here’s my view:

High Carbohydrate, Low Fat – Greater outright performance potential, inflexible.

Low Carbohydrate, High Fat – Limits top end performance, but may increase absolute endurance potential. Unlocks metabolic flexibility.

For a competitive athlete then, fuelling with carbohydrates makes a lot of sense. One of the reasons for this is that it helps maximise your ability to make quick changes of pace – something that may be the difference between winning and losing a race. For this group of runner, and particularly for events lasting less than 2 hrs a high carb approach is probably the best bet.

Past 2 hrs of continuous exercise however, it becomes increasingly difficult to take on enough carbohydrates to avoid running out of fuel. If you do and the body is forced to switch to fatty acid metabolism, that process is going to hurt (oh, hello ‘the wall’). In contrast, a LCHF diet could increase your ability to maintain a steady pace for longer. This potentially makes it a more effective approach for endurance/ultra-endurance events, especially for runners who are only racing the clock. The caveat of course, is that the relative effort needs to be low, and/or you need to have trained your body to be effective at burning fat.

If you can make a LCHF diet work, the real benefit seems to be the flexibility it offers. Start fuelling with simple carbohydrates and you’re committed to that approach. A fat to carbohydrate energy switch on the other hand seems less likely to affect performance, giving you the option to change strategies during an event.

Of course, implementation is the difficult part when it comes to diet. If making any significant changes, it’s going to require plenty of time for your body to adapt and subsequently to figure out what works best for you.

So there you go, a conclusion of sorts. As ever though, NOT advice.

If anybody does have advice they can offer me, please help out by leaving a comment below or sharing with people who actually have some expertise/experience with nutrition for a marathon. All views welcome, including those that point out I’m being an idiot – just as long as you tell me why!

Spread the word:

2 Comments on “Nutrition for a marathon – what should you eat?

Daniel 'Bear on a Bike' Francis
February 25, 2016 at 4:51 pm

As someone who has been part of a endurance exercise type thing, enjoying hitting that dark place you never thought you would get too. Think eating/killing your best friends and weighing up the ease of killing yourself. I would scrap the glucose gels and just haribo it up (other jelly based sweets are available), only because they taste like glucose gels and that is not great. During training make sure you get all the nutrients you can think of, lots of greens, yellows and any other colour veg you can think off. It will reduce your DOMs and help you get back in the swing of things the next day.
As everyone has said mate, you are f*cked. It is not because I don’t believe in you, it is just that I know how much you are going to hurt.


[…] Nutrition for a marathon – what should you eat? […]


Leave a Reply

Your email address will not be published. Required fields are marked *