With marathon season underway, there are surely several nutrition questions running through people's minds. "What should I eat the day before the race?" "What should I eat the morning of the race?" "Should I alter my morning routine at all?" "Should I do a carbo-load?" "What is a carbo-load?" Et cetera. Nutrition is a big topic that's impossible to cover in one blog post (or even one blog!), but in this post, I'll attempt to explain carbohydrate loading.
For aerobic exercise lasting longer than about two minutes, the primary fuel for muscles is glycogen. The muscles and liver are able to store a certain amount of glycogen, and the amount of stored glycogen is generally enough to get an athlete through most races and workouts, provided the athlete is eating a healthy diet. However, it is not enough to get an athlete through a marathon (or really, any race lasting longer than 90 minutes), so in order to keep from running out of energy, the athlete either needs to spare glycogen and use fat as fuel, store more glycogen than normal, or ingest easily digested carbohydrates during the race. The problem with using fat as fuel is that as intensities get harder, the ratio of fat:carb usage moves toward a greater carbohydrate usage. Additionally, the high-carbohydrate diet favored by athletes (for good reason, considering carbs are the primary source of energy for workouts, as mentioned earlier) has a fat-sparing effect during workouts. Getting accustomed to a high-fat diet will help the body learn to spare glycogen during endurance workouts, but, as Noakes explains in Lore of Running, athletes generally perform better off of a high-carbohydrate diet, making training with a high-fat diet less than optimal. That's the short version anyway, and for anyone interested in the physiology of this, I suggest you pick up a copy of the aforementioned book, because Noakes does an excellent job of explaining this.
So the problem now becomes how to get used to the high-fat diet while still doing the majority of your training while eating a high-carbohydrate diet. In the 1960s, athletes discovered the original form of carbohydrate loading, which caused glycogen supercompensation, and the muscles and liver, which had previously been starved of glycogen, stored more than usual, allowing the athlete to perform longer at a given intensity and resulting in faster race times. The original carbo-load involved an exhaustive workout at the beginning of the week to deplete the body's glycogen stores, followed by about three days of carbohydrate depletion (high-fat, low-carb diet, with ~10% of calories from carbs), and finally, with about four days of carbohydrate loading (high-carb, with ~90% of calories from carbs). This is known as the Ahlborg method. However, as anyone who has ever tried a low-carb diet knows...this sucks! For those three days of carbohydrate depletion, you'll be miserable and fatigued, there's a chance your digestive system will rebel, and worst of all, in those couple days when you really want that final confidence boost, you'll run like crap! Clearly, the original carbo-load is not practical for the majority of runners today.
Luckily for runners, scientists later realized that eating a high-carbohydrate diet without the depletion phase raised muscle glycogen levels and resulted in nearly the same performance benefits as the Ahlborg carbo-load. This approach is favored by many more athletes today, because it does not have all the nasty side effects of the Ahlborg method. This method involves eating a normal diet until about three days before the race, and then switching to a high-carb diet (~80% of calories from carbs) for the final three days.
A few years ago, that's where this post would have ended. However, in 2002, researchers at the University of Western Australia found that the carbo-load could be condensed into a single day, and with the same results as the Ahlborg method. In the Western Australia approach, athletes perform a very high intensity three minute workout consisting of 2:30 at mile race pace followed by a 30 second sprint. This workout is followed by a high carbohydrate diet (~12g per kg of lean body mass). I have not personally tried this method, since I've done a three mile premeet the day before the race for years and have become a bit superstitious about it, but the short workout should not actually be enough to take away from your performance, and may be worth a try, as it supposedly increases glycogen storage up to 90%(!).
There is, of course, the question of what type of carbohydrates to consume. Really, it's up to you and what your body is used to eating. I limit my intake of high fiber carbohydrates, like whole grains, fruits, and vegetables, and turn to refined and easily digested foods, like white breads and pastas, in the days immediately before the race. I figure less fiber means less chance of a nasty surprise on race day, and this method has worked pretty well for me. Other people may find that different food combinations work better for them. I also like to eat a decent high-carb breakfast on race morning, usually consisting of a bagel with peanut butter and an apple a few hours before the race, and a Powerbar right before I start my warmup. I'll also carry Gu for half-marathons or longer. Again, figure out what works for you, since for some people, the protein and fat from the peanut butter may be too much (or not enough), or the fiber from the apple could get them in trouble, etc. Also, people prone to transient hypoglycemia may not want to eat large amounts of food within 90 minutes of their race, and should make sure they are awake early enough to allow sufficient time for digestion after breakfast. No matter what you choose to eat, make sure to give it a trial run before race day (preferably for a speed workout...I can get away with a lot more for easy runs than I can for quality workouts).
Carbohydrate loading is extremely valuable to the endurance athlete in that it can allow an athlete to sustain a given intensity for 20% longer than without the carbo-load, which can equate to as much as a 3% time improvement. However, it is important to note that all of this research was done on athletes who did not consume food during the race, and that eating during the race can give similar results as a carbo-load. Even with a carbo-load, it can be beneficial to consume carbohydrates during longer races. Additionally, it's beneficial to taper your training (volume, not necessarily intensity) going into the race, such that your muscles are recovered and have a chance to build their glycogen stores.
Finally, remember that water stores with glycogen in your muscles. Don't be surprised if you see a couple pounds of weight gain before the race. This is a good thing. You will need that energy and water for the race. If you're prone to freaking out over weight fluctuations, just stay away from the scale in the days leading up to the race.