But nutrition isn’t that simple. Studies use different levels of athletes, different preparation, and different measures of performance. Regardless I decided to have a crack at pooling the best available science, to at least put a rough figure on the benefits of nutrition to cycling performance.

Cycling studies usually measure performance by either self-paced Time Trials (TT), or Time To Exhaustion (TTE, riding at a constant intensity until you can’t maintain it). Sports scientists favour TT over TTE studies because in the real world the distance is fixed and the pace varies, not the other way around. The TT model still isn’t perfect because in the real world tactics will have a large say in the pace. Think about riding in a breakaway, trying to bring one back, or rolling along in a peloton with minimal effort. A TT doesn’t reflect these variations.

Researchers attempt to simulate the typical pattern of effort in road cycling by preceding the TT with a “steady state” period, where the participant rides at a fixed (moderate) intensity for a period (usually 1-2 hours). This is designed to simulate riding in a bunch before the action happens. The TT then follows and is usually between 30 minutes and 2 hours. Performance is only measured in the TT.

So without further ado, let’s look at the researched benefits of nutrition to cycling performance. I’ve limited this to strategies used specifically for competition. Strategies to enhance the benefits of training are much more complex and beyond the scope of what can be presented in this format.

Carbohydrate Loading

Carb loading is more than just eating pasta the night before a race. In my work with triathletes and cyclists I find very few eat enough to do it properly. The studies here compared proper carb loading to a more moderate carbohydrate intake. There’s three good ones that used TT measures – a 90 minute steady state with a 16km TT, a 2 hour steady state (with interspersed sprints) and 60 minute TT (distance covered in one hour), and a one-off 70km TT. The performance benefits of proper carb loading were 3.3%, 4.1% and 4.0% respectively. That equated to nearly a five minute savings in a 70km TT.

Carbohydrate during a race

What benefit do carbs give you during a race? That depends on three factors – the duration of the race, the total amount and the types of carbs consumed. The main benefit of carb consumption will come from events greater than 2 hours duration.

As for the quantity, the more you consume the faster you’ll go. One study looked at low to moderate doses of carbs (less than 60g an hour) compared with water. After 2 hours steady state cyclists rode a 20km TT. The benefit from carbs increased with the quantity, from 3.2% at 15g an hour up to 4.7% at 60g an hour.  When total carb consumption gets up to 90g an hour the benefits are even greater. In two studies 90g an hour of glucose and fructose (60g/hr glucose, 30g/hr fructose) was 8% and 7.5% better than 90g/hr of just glucose and a whopping 18% better than flavoured water! The marketing of 2:1 glucose/fructose ratio products in the last few years comes from this research. This combination of glucose and fructose is important, but only if total carb consumption is greater than 60g an hour.

To demonstrate this effect in the real world, a study published just last week by Asker Jeukendrup (who works with Rabobank) found that carbohydrate intake accounted for 30% of the variation in finishing time at the 2009 Ironman Triathlon World Champs in Kona.

Relationship between carbohydrate intake and finish times at the 2009 Ironman Triathlon World Championship, Kona Hawaii. R = 0.55, R2 = 0.30, p < 0.001. Source: Med Sci Sports Exerc. 2011 Jul 19. [Epub ahead of print].

Asker’s team also observed how much professional cyclists consume during a race. Carb intakes of Rabobank and HTC-Highroad riders at the 2009 Vuelta (2 mountain stages and 1 flat stage) and Criterium du Dauphine (2 flat stages) averaged 64g an hour, but ranged from 29-107g an hour.

Weight Loss

The effect of weight loss has been mathematically modelled for a 40km TT by body composition gurus Kevin Norton and Tim Olds in Adelaide. They found the benefit s of body weight loss was minimal on flat roads (0.2% for 1kg lost and 0.9% for a 5kg reduction), but the benefit became apparent on a 5% incline (1.90% benefit for 1kg reduction, 9.5% for 5kg). That puts Thor Hushovd’s efforts in this year’s Tour into perspective, since he lugs around almost 15kg more than the main GC contenders.


Hydration is an interesting one, because there’s surprisingly very little quality evidence to draw from. Most of the studies prior to the late 1990s used TTE methods, not TT. Most studies also compared complete fluid intake to no fluid intake at all, not at all reflective of the real world. Only recently have studies started comparing ad libitum fluid intake (drinking according to natural instinct or thirst) to complete replacement of sweat losses. In fact there’s only two studies ever published that compare ad lib fluid intake to complete sweat replacement in a TT format! One found no benefit of any fluid intake compared to no fluid; however the duration was only an hour – not long enough for hydration to impact on performance.

Thankfully the other study used a longer TT protocol (80km with no steady state). In 33oC heat participants were 3.2% quicker with ad lib fluid intake (which replaced ~50% of sweat losses) compared to no fluid at all, however the small number of participants in the study meant that this result is not clear cut. There was no benefit in replacing 100% of sweat losses compared to ad lib intake (-0.09%). Whether ad lib fluid intake is sufficient for rides longer than 2 hours remains unknown.


In contrast to the lack of quality research on something as fundamental as hydration, there’s a gold mine of data about caffeine. Without going through each study, most involved 1-2 hours of steady state followed by a 30-60 minute TT. The benefits across a range of caffeine doses (from 2-6mg per kg body weight, the equivalent of 2-5 strong coffees or energy drinks) are remarkably similar at around 3-5%.

Nitrate Supplementation – The New Kid on the Block

You may have read of the supposed benefits of beetroot juice (which contains nitrates). Nitrates are reported to reduce the amount of oxygen required to produce energy, but how this works is not yet understood. The hype has outgrown the research though, and until last month there wasn’t a single study that actually measured TT performance after nitrate supplementation. This single study compared beetroot juice to processed nitrate-free beetroot juice. The study measured 4km and 16km TT performance and found 2.8% and 2.7% benefits from the nitrates. Whilst these short TT’s might be relevant to track cyclists, there’s still no data over longer distances.

Putting it all together

One comment about these studies is that in many cases they don’t take all nutrition factors into account. For example in some hydration studies participants don’t consume any carbohydrate during their ride, and certainly didn’t carb load beforehand. So it’s likely that the 3 and 4 percent benefits seen for each individual factor in many of these studies may be different when you start to combine strategies together. Add in the inevitable tactics of cycling and the picture becomes even less clear.

In terms of the most beneficial nutrition strategies, the largest gains are seen with very high carbohydrate intakes during exercise, especially combining glucose and fructose in doses of 90g an hour or above. Deliberately drinking to replace sweat losses beyond normal thirst may not be any advantage, but there’s not enough data to drawn definitive conclusions. Losing excess body fat will help in the hills, but makes little difference on flat terrain. Nitrates might help in shorter events, but haven’t been studied over longer durations. Caffeine does seem to be consistently beneficial though.

Nutrition versus Equipment

Finally coming back to CT’s original post on TT equipment, the benefits from improving your nutrition appear to be as beneficial (and in many cases more so) as aero bars, skinsuits, helmets and disc wheels, at least in a TT following a couple of hours of riding. Nutrition’s also a fairly cheap strategy – even if you need a couple of appointments with an expert it’ll cost a lot less than a new set of wheels. So next time you’re preparing for a long ride, think about what you can do with your eating and drinking. It might just make the difference between feeling flat or riding your mates off your wheel.

Research Summary:

Carbohydrate Loading

  1. Rauch et al. Br J Sports Med 2005;39:34–38. 1hr TT preceded by 2hrs at 65% VO2peak.
  2. MacLaren et al. Eur J Appl Physiol. 2011 Feb 1 [Epub ahead of print]. 16km TT preceded by 90min at 65% VO2max.
  3. Widrick et al. J Appl Physiol. 1993 Jun;74(6):2998-3005. 70km TT (no steady state).

Carbohydrate During Riding

  1. Currell & Jeukendrup. Med. Sci. Sports Exerc. 2008 40(2):275–281. TT equiv. to 60min at 75% Wmax, preceded by 2hrs at 55% Wmax.
  2. Triplett et al. Int J Sports Nutr Ex Metab 2010, 20:122-131. 100km TT with 4 X 1km & 4 X 4km sprints.
  3. Smith et al. J Appl Physiol 2010; 108: 1520–1529. 20km TT preceded by 2hrs at 77% VO2 peak.

Weight Loss

  1. Norton & Olds (1996). Anthropometrica. UNSW Press. 40km TT (mathematically modeled).


  1. Dugas et al. Eur J Appl Physiol (2009) 105:69–80. 80km TT (no steady state).

Caffeine Supplementation

  1. Desbrow B et al. Med Sci Sports Exerc. 2009; 41(9):1744-51. TT equiv. to 30min preceded by 2hrs at 70% VO2 peak.
  2. Cox G et al. J Appl Physiol. 2002; 93(3):990-9. TT equiv. to 30min preceded by 2hrs at 70% VO2 peak.
  3. Ivy JL et al. Int J Sport Nutr Exerc Metab. 2009; 19(1):61-78. TT equivalent to 60min at 70% Wmax.
  4. Irwin et al. J Sports Sci. 2011 29(5):509-15. TT equivalent to 60min at 75% Wmax.
  5. Hulston & Jeukendrup. Med Sci Sports Exerc. 2008 Dec;40(12):2096-104. TT equivalent to ~45min preceded by 105min at 62% VO2 max.
  6. McNaughton LR et al. Int J Sports Physiol Perform. 2008 Jun;3(2):157-63. 60min TT (distance covered in 60min).