Nutrition for Zone 2 Training: Eat Enough to Train Well

The problem is rarely what you eat

The most common nutrition problem among cyclists isn't the wrong food. It's too little food. Most rides happen without any intake, and you run out of energy on many sessions a week.

Even with low training volume, you can quickly experience poor form if you eat poorly or too little before, during, and after training — and deplete yourself of energy the times you actually do train. The consequence creeps up on you. We start with the daily diet. That's where most of the work is done.

The daily diet matters more than the timing

What you eat across the day matters more for your training than whatever you manage in the last twenty minutes before heading out. Total intake decides whether your stores are full when you get on the bike.

Protein is the nutrient with the clearest numbers. A daily intake of 1.4–2.0 grams per kilo of body weight covers the need for most people who train (Jäger et al. 2017). For a 75-kilo cyclist that is 105–150 grams a day. Distribution counts too. Roughly 0.25 grams per kilo per meal, or 20–40 grams, spread evenly every three to four hours (Jäger et al. 2017).

Carbohydrate need follows training load. When the training week gets heavier, the plate should get fuller. How much you need around any single session, though, depends on what kind of session it is.

What powers a zone 2 session

The body uses fat and carbohydrate at the same time, and the balance between them shifts with intensity. In the lower part of zone 2, fat dominates. Closer to LT1, the first lactate threshold, where lactate begins to rise at the top of the zone, carbohydrate gradually takes over. Carbohydrate delivers higher power, while fat offers greater capacity (Hargreaves & Spriet 2020).

On a long ride it is the glycogen that disappears, not the fat oxidation. That is why you have to eat along the way. Why that is so is explained in Fat Burning and Metabolic Flexibility.

Before, during and after the session

How hard you ride decides which fuel you burn. How long you ride decides how much you have to replace. The advice below therefore follows duration.

Before the session

Under 90 minutes

You don't need to eat anything special. If you've eaten normally through the day, your stores are good enough.

From 90 minutes

Start with full glycogen stores (Hargreaves & Spriet 2020). A normal meal in good time before you leave is enough.

Over three hours

The advice is the same, only more so. Start full, and carry food from the beginning.

During the session

How much you need follows the length of the ride.

Under 90 minutes, water is enough.
From 90 to 180 minutes you need 30–45 grams of carbohydrate per hour.
Over 180 minutes you need 45–60 grams per hour. Start after three quarters of an hour, not when you feel empty.

The research on longer sessions lands at 30–60 grams per hour, and up towards 90 grams for very long races (Burke et al. 2011; Jeukendrup 2014). On easy sessions you sit in the lower part of that span.

After the session

What matters most happens over the next day, not in the first twenty minutes.

Eat a normal meal within a couple of hours. Fully rebuilding glycogen stores takes 24–48 hours and depends on how much carbohydrate you take in during that period (Egan & Zierath 2013). If you have a hard session tomorrow, then it is urgent. If tomorrow is easy, eating as usual is enough.

Fasted training: when it's fine

Go ahead and ride an easy zone 2 session on an empty stomach. East African runners have used the model for a long time: an easy session in the morning without breakfast, a quality session in the afternoon after a proper meal (Baar 2014). Hard sessions, on the other hand, should always have carbohydrate underneath them. Without it the quality of the intervals drops (Yeo et al. 2008), and you get a worse session than you could have had.

What does riding fasted actually buy you? Less than its reputation suggests. Low glycogen stores activate AMPK and p38, two signalling proteins that set adaptation going inside the muscle cell (Baar 2014). A meta-analysis also finds higher mRNA levels of PDK4 and UCP3, two genes that govern which fuel the cell selects (Diaz-Lara et al. 2025).

But the mRNA levels of PGC-1α and TFAM, the genes that drive mitochondrial building itself, do not rise (Diaz-Lara et al. 2025). The effect on PGC-1α appeared only in a subgroup where glycogen stores differed by more than 200 mmol per kilo dry weight between the two conditions. That number measures how much glycogen the muscle has stored. It is a difference between two measurements, not a level you are meant to hit.

The performance numbers are clearer. In well-trained cyclists who placed two sessions every second day, and therefore rode the second on low stores, oxidative enzymes and fat oxidation rose more than in those training daily. Both groups improved performance by about 10 percent, and the difference between them was none (Yeo et al. 2008). The gain may in any case come from the sessions sitting close together rather than from the glycogen level itself (Andrade-Souza et al. 2020).

When you eat too little over time

If you eat too little across weeks, it shows up as a pattern we recognise from our own training. You feel weak from the start and get unusually hungry early in the ride. Your legs resist even moderate effort, and you need longer than usual to come round.

These are signs that you should eat more, not train less. When carbohydrate availability is low, the body increases the breakdown of amino acids to cover its energy need (Egan & Zierath 2013). Left to run, you draw on the muscle you are training to build.

If the pattern stands for months, it has a name. Low energy availability means too little energy relative to what the training expends. That is the core of Relative Energy Deficiency in Sport, a syndrome with documented consequences for health and performance in both women and men (Mountjoy et al. 2023). If you recognise this over months, it is a matter for a doctor or a dietitian, not an adjustment to the training plan.

Supplements: antioxidants and caffeine

Most cyclists need no supplements beyond ordinary food. Free radicals have a bad reputation, but during exercise they act as signalling molecules that start the adaptations you are training for (Gómez-Cabrera et al. 2008). Damp the signal with high doses of antioxidants and you damp the response as well.

Vitamin C and E can slow adaptation

In a study of 54 young recreational athletes over eleven weeks, PGC-1α and COX4, an enzyme that measures mitochondrial capacity, rose in the placebo group. In those given vitamin C and E, they did not (Paulsen et al. 2014). At the same time VO₂max rose equally in both groups. Short-term performance was not affected, but the cellular adaptations were.

The advice is simple: eat fruit and vegetables, and skip high-dose C and E supplements during periods when you are training to improve.

Caffeine occupies a different place. It raises fat oxidation at doses above 3 mg per kilo, but the effect is largest in untrained people and smallest in trained athletes (Collado-Mateo et al. 2020). The underlying studies were all done fasted and at submaximal intensity. Caffeine also offsets some of the power loss on sessions with little carbohydrate (Rothschild & Bishop 2020).

Summary

Six principles

Eat enough. Riding yourself empty week after week is one of the most common causes of poor form.
Prioritise the daily diet. 1.4–2.0 grams of protein per kilo per day, spread across the day (Jäger et al. 2017).
Let duration decide. Under 90 minutes you need little. From 90 minutes you need 30–45 grams of carbohydrate per hour, and 45–60 grams once the ride passes three hours.
Fasted training belongs to the easy sessions. Hard intervals should always have carbohydrate underneath them.
Skip high-dose C and E during training periods. They damp adaptation, and no study shows a gain that outweighs it.
Know the signs. Hungry early, heavy from the start, slow to recover. The answer is more food.

References

Andrade-Souza VA, et al. (2020). FASEB J 34(1):1602–1619. https://doi.org/10.1096/fj.201901207RR
Baar K (2014). Sports Med 44(S1):S5–S12. https://doi.org/10.1007/s40279-014-0146-1
Burke LM, Hawley JA, Wong SH, Jeukendrup AE (2011). J Sports Sci 29(sup1):S17–S27. https://doi.org/10.1080/02640414.2011.585473
Collado-Mateo D, et al. (2020). Nutrients 12(12):3603. https://doi.org/10.3390/nu12123603
Diaz-Lara J, Prieto-Bellver G, Guadalupe-Grau A, Bishop DJ (2025). Sports Med 55(1):79–100. https://doi.org/10.1007/s40279-024-02119-9
Egan B, Zierath JR (2013). Cell Metab 17(2):162–184. https://doi.org/10.1016/j.cmet.2012.12.012
Gómez-Cabrera MC, et al. (2008). Am J Clin Nutr 87(1):142–149. https://doi.org/10.1093/ajcn/87.1.142
Hargreaves M, Spriet LL (2020). Nat Metab 2(9):817–828. https://doi.org/10.1038/s42255-020-0251-4
Jäger R, et al. (2017). J Int Soc Sports Nutr 14:20. https://doi.org/10.1186/s12970-017-0177-8
Jeukendrup A (2014). Sports Med 44(S1):S25–S33. https://doi.org/10.1007/s40279-014-0148-z
Mountjoy M, et al. (2023). Br J Sports Med 57(17):1073–1097. https://doi.org/10.1136/bjsports-2023-106994
Paulsen G, et al. (2014). J Physiol 592(8):1887–1901. https://doi.org/10.1113/jphysiol.2013.267419
Rothschild JA, Bishop DJ (2020). Sports Med 50(1):25–53. https://doi.org/10.1007/s40279-019-01185-8
Watt MJ, et al. (2002). J Physiol 541(3):969–978. https://doi.org/10.1113/jphysiol.2002.018820
Yeo WK, et al. (2008). J Appl Physiol 105(5):1462–1470. https://doi.org/10.1152/japplphysiol.90882.2008