Zone 2 and intensity: why we train both

Zone 2 alone is not the whole answer

Zone 2 carries our programme. For us it gives a high training load at low risk, it fills the days between the quality sessions, and it lets you train a lot without digging yourself into a hole. But zone 2 is the foundation, not the whole house. Capacity is built by the 1–2 quality sessions a week, from threshold to VO2max.

That is a distribution, not an absence. When we write that more time means more zone 2 rather than more intervals, we mean that the hard sessions stay fixed while the easy hours grow. The intervals never disappear.

The question is what each part actually does. The answer is why the hard sessions cannot be traded away for more easy hours.

What the easy hours build

Easy training builds the engine in the muscle. Endurance training, low and moderate intensity taken together, raises mitochondrial content by about 23 % and capillaries per muscle fibre by about 15 % (Mølmen et al. 2025). Hard sessions do the same when the training work is matched.

These adaptations sit in the muscle, and they are what makes speed cheaper below threshold. In riders who are already well trained, capillary growth appears to level off (Liu et al. 2022).

Training held entirely below the first threshold stands on its own. In a review of 50 randomised studies in healthy adults, such training had a large effect on VO2max and a moderate effect on the threshold itself (Nuuttila et al. 2026). The same review found no lower intensity limit for most outcomes: easy work counts (Nuuttila et al. 2026).

The price is what decides it. Below the first threshold the autonomic nervous system is back in balance within 5–10 minutes, even after two hours, while work above it costs around 30 minutes (Seiler et al. 2007). After a hard interval session it took 90 minutes or more in less trained riders. The boundary is a divide rather than a scale: going even harder did not lengthen recovery further (Seiler et al. 2007).

We have found the same in practice: a zone 2 session needs no recovery day, while a threshold session needs one or two. That is why we steer the total load with the zone 2 volume rather than with the number of intervals.

Fat oxidation, mitochondria and lactate as fuel are covered in their own articles.

What intensity adds

The ceiling for VO2max is set somewhere other than in the muscle: by oxygen delivery, not by extraction (Bassett & Howley 2000). That is where intensity does its work.

What intensity adds sits in the heart and the blood. In a controlled study of 84 young, healthy participants, the training work was matched in every group except the sprint group. The largest VO2max gains came in the interval group and in the hardest continuous group, and the authors found the gain was carried mainly by central adaptations (Inglis et al. 2025). Maximal cardiac output rose in every group except the moderate one, and blood and plasma volume rose only in the two hardest (Inglis et al. 2025). Across 13 studies in untrained and moderately trained people, the VO2max gain tracks maximal cardiac output rather than oxygen extraction in the muscle (Montero et al. 2015).

Intensity also changes how the muscle handles what it produces. Six well-trained cyclists completed six hard sessions over four weeks. Muscle buffering capacity rose and 40 km time-trial performance improved, with no change in the oxidative enzymes (Weston et al. 1997). That is as close to a clean demonstration as the literature gets that hard training gives something easy training does not: the same engine, and a better ability to run it near the ceiling. Intensity also appears to be the main driver of how many lactate transporters the muscle carries, although transporter content and the capacity to move lactate do not always follow each other (Thomas et al. 2012).

Without intensity you get less of the central half: the heart's pumping capacity, blood volume, and the ability to hold high speed. Easy hours build VO2max too. Intervals give a little more, but the margin is small: 1.2 mL/kg/min across 28 studies (Milanović et al. 2015). The difference lies in which systems are reached. The mechanism is not entirely clean either: when researchers removed the blood volume untrained subjects had built over ten weeks of training, the gain in VO2peak, the highest oxygen uptake measured in the test, did not disappear with it (Skattebo et al. 2020).

Why the parts reinforce each other

The two parts solve different problems. The base makes the intensity possible: the purpose of the base period is not peak form, but a body that can take peak form. Easy hours that need no recovery are what leave room for hard sessions that do.

Zone 2 is also the glue. The load is low enough that recovery happens while you ride, so the days between quality sessions become training days that build you back up at the same time. That is how the easy hours pay for the hard ones.

Intensity makes the base worth more. When competitive cyclists replaced 15 % of about 300 km a week with hard intervals, their 40 km time improved by two minutes (Lindsay et al. 1996). The volume was already there. What was missing was the intensity.

The direct comparison points the same way. When 48 well-trained athletes were split across four training approaches for nine weeks, the group that combined a lot of easy training with hard intervals raised VO2peak the most, by 11.7 % (Stöggl & Sperlich 2014). Neither volume alone nor threshold work alone produced measurable gains in the endurance variables, apart from a small improvement in work economy in the threshold group (Stöggl & Sperlich 2014). Neither half on its own was enough.

The split between them is not a matter of indifference either. In a controlled study of 12 endurance runners over five months, with the share of hard training held equal in both groups, more of the volume below the first threshold produced greater improvement than more of the volume in the middle zone (Esteve-Lanao et al. 2007). The authors' own conclusion is that a large share of low-intensity training works as long as the share of high-intensity training stays sufficient. Seiler puts it this way: a predominance of low-intensity work together with fewer, genuinely hard sessions may be complementary, because together they give good adaptive stimulus at an acceptable level of stress (Seiler 2010).

How much intensity is needed

A little goes far, and more is not better. Seiler finds no convincing evidence that more intensity produces long-term gains in athletes who are already well trained (Seiler 2010).

In 15 elite cross-country skiers and biathletes, with training volume held equal, only the group with fewer and longer interval sessions improved its time-trial performance measurably (Tønnessen et al. 2020). The authors' own explanation is recovery: more time between sessions let the athletes absorb the training better. That is the same logic behind our 1–2 quality sessions a week. We have found that it pays to put a little more time at a little lower intensity rather than the other way around.

Too much intensity costs something measurable. Four weeks of intensified training in elite athletes produced about 20 % lower mitochondrial respiration, even though the markers of mitochondrial content rose (Cardinale et al. 2021). Performance was nevertheless better a few days later, so the finding should not be overstated. In a study of healthy volunteers where the load was increased week by week, intrinsic mitochondrial function fell alongside poorer glucose control (Flockhart et al. 2021).

Some common assumptions do not hold. There is no validated marker for overtraining: none of the hormones, tests or biochemical measures meets the criteria for general use (Meeusen et al. 2013). And in an eight-year study of 37 cross-country skiers, it was not training volume that was associated with infections, but travel, competition and large swings in load (Svendsen et al. 2016).

A high VO2max counts for health too

This is not only a performance question. In a cohort of 122 007 patients with 13 637 deaths, fitness was inversely associated with mortality, with no upper limit of benefit found, and elite fitness gave lower mortality than high fitness (Mandsager et al. 2018). The difference between low and elite fitness was comparable to or greater than the difference associated with smoking, diabetes or coronary artery disease in the same material (Mandsager et al. 2018).

A meta-analysis of 33 cohorts points the same way: 13 % lower mortality per MET of higher fitness (Kodama et al. 2009). Change counts as well. In 579 men followed for 15 years, 1 mL/min/kg higher VO2max at a second measurement was associated with 9 % lower mortality (Laukkanen et al. 2016).

Here we have to be precise about what has not been shown. Oxygen uptake at the ventilatory threshold, a breathing-based measure of the same divide, taken during a submaximal test, also predicts mortality — but the association disappears once researchers adjust for maximal oxygen uptake (Kunutsor et al. 2017). The two measures are so closely related in that study (r = 0.90) that it cannot separate them, and the study does not state which of the two ventilatory thresholds it used. What is missing is documentation, not a finding against the threshold. We train threshold because it carries the speed you can hold, not because it has been shown to extend life on its own.

All of this is observational data. It shows association, not cause. The health effects of easy training are covered separately.

What the research does not agree on

Our argument has to survive its counter-evidence, so here it is.

Per hour, zone 2 is the least efficient route to mitochondria: sprint intervals were about 3.9 times more efficient (Mølmen et al. 2025). The answer lies in the denominator. Per hour, that is right. But when the training work is matched, the adaptation is just as large (Mølmen et al. 2025), and easy hours can be stacked day after day, week after week, without having to be rested off.

In untrained people, 2.5 hours of sprint training over two weeks matched what 10.5 hours of endurance training gave (Gibala et al. 2006). In a 10-week study of 35 untrained and recreationally active adults, hard training built durability as well as easy training did, on a quarter of the hours (Matomäki et al. 2023). Durability is the ability to hold your level when you are tired. Both findings are in untrained people. For them almost all training is new, and what holds there does not transfer directly to a cyclist who has trained for years.

The distribution is not settled either. The largest individual-participant meta-analysis in the field found no difference between polarised and pyramidal distributions, for VO2max or for time-trial performance (Rosenblat et al. 2025). What that means for your week belongs in the article on intensity distribution.

On intensity distribution →

Nor is a lot of volume free. Among joggers, the association between dose and mortality was U-shaped, with the greatest benefit in those who ran least (Schnohr et al. 2015), but the confidence interval for the hardest group is wide, so the finding is weak.

A lifetime of high training volume is associated with more coronary calcification, with a more benign plaque composition (Aengevaeren et al. 2017), and endurance sport is associated with a higher risk of atrial fibrillation (Newman et al. 2021). Note that this is a different exposure from the previous section: there it was how fit you are, here it is how many hours and years you have put in. The two do not automatically follow each other.

The field itself does not consider the case closed. A 2025 perspective article sets several competing explanations for why low-intensity training works against each other, and its author does not rule out that the volume could in principle be replaced (Matomäki 2025).

How the combination looks here

What does not change through the year is the structure: the quality sessions stay fixed, and zone 2 is placed around them. In the base period, zone 2 builds the foundation. In season, it maintains that base and secures recovery between hard sessions and races.

If you want to see how that looks in practice, the distribution across the week is in the article on intensity distribution, and the zones with their numbers are in the zone article. Ready-made weeks are in the programmes, and you work out your own zones in the calculator.

Sources

1. Mølmen KS, Almquist NW, Skattebo Ø (2025). Sports Medicine 55(1):115–144. doi:10.1007/s40279-024-02120-2
2. Liu Y, Christensen PM, Hellsten Y, Gliemann L (2022). Med Sci Sports Exerc 54(10):1714–1728. doi:10.1249/MSS.0000000000002955
3. Nuuttila OP, Matomäki P, Raitanen J, Sievänen H, Vasankari T (2026). Scand J Med Sci Sports 36(1):e70208. doi:10.1111/sms.70208
4. Seiler S, Haugen O, Kuffel E (2007). Med Sci Sports Exerc 39(8):1366–1373. doi:10.1249/mss.0b013e318060f17d
5. Bassett DR, Howley ET (2000). Med Sci Sports Exerc 32(1):70–84. doi:10.1097/00005768-200001000-00012
6. Inglis EC, Rasica L, Iannetta D, et al. (2025). Med Sci Sports Exerc 57(8):1669–1680. doi:10.1249/MSS.0000000000003697
7. Montero D, Diaz-Cañestro C, Lundby C (2015). Med Sci Sports Exerc 47(10):2024–2033. doi:10.1249/MSS.0000000000000640
8. Weston AR, Myburgh KH, Lindsay FH, et al. (1997). Eur J Appl Physiol Occup Physiol 75(1):7–13. doi:10.1007/s004210050119
9. Thomas C, Bishop DJ, Lambert K, Mercier J, Brooks GA (2012). Am J Physiol Regul Integr Comp Physiol 302(1):R1–14. doi:10.1152/ajpregu.00250.2011
10. Milanović Z, Sporiš G, Weston M (2015). Sports Medicine 45(10):1469–1481. doi:10.1007/s40279-015-0365-0
11. Skattebo Ø, Bjerring AW, Auensen M, et al. (2020). Eur J Appl Physiol 120(5):985–999. doi:10.1007/s00421-020-04336-2
12. Lindsay FH, Hawley JA, Myburgh KH, et al. (1996). Med Sci Sports Exerc 28(11):1427–1434. doi:10.1097/00005768-199611000-00013
13. Stöggl T, Sperlich B (2014). Frontiers in Physiology 5:33. doi:10.3389/fphys.2014.00033
14. Esteve-Lanao J, Foster C, Seiler S, Lucia A (2007). J Strength Cond Res 21(3):943–949. doi:10.1519/R-19725.1
15. Seiler S (2010). Int J Sports Physiol Perform 5(3):276–291. doi:10.1123/ijspp.5.3.276
16. Tønnessen E, Hisdal J, Rønnestad BR (2020). Int J Environ Res Public Health 17(9):3190. doi:10.3390/ijerph17093190
17. Cardinale DA, Gejl KD, Petersen KG, et al. (2021). J Appl Physiol 131(1):388–400. doi:10.1152/japplphysiol.00829.2020
18. Flockhart M, Nilsson LC, Tais S, et al. (2021). Cell Metabolism 33(5):957–970.e6. doi:10.1016/j.cmet.2021.02.017
19. Meeusen R, Duclos M, Foster C, et al. (2013). Med Sci Sports Exerc 45(1):186–205. doi:10.1249/MSS.0b013e318279a10a
20. Svendsen IS, Taylor IM, Tønnessen E, Bahr R, Gleeson M (2016). Br J Sports Med 50(13):809–815. doi:10.1136/bjsports-2015-095398
21. Mandsager K, Harb S, Cremer P, et al. (2018). JAMA Netw Open 1(6):e183605. doi:10.1001/jamanetworkopen.2018.3605
22. Kodama S, Saito K, Tanaka S, et al. (2009). JAMA 301(19):2024–2035. doi:10.1001/jama.2009.681
23. Laukkanen JA, Zaccardi F, Khan H, et al. (2016). Mayo Clin Proc 91(9):1183–1188. doi:10.1016/j.mayocp.2016.05.014
24. Kunutsor SK, Kurl S, Khan H, Zaccardi F, Laukkanen JA (2017). Int J Cardiol 236(2):444–450. doi:10.1016/j.ijcard.2017.01.156
25. Gibala MJ, Little JP, van Essen M, et al. (2006). J Physiol 575(Pt 3):901–911. doi:10.1113/jphysiol.2006.112094
26. Matomäki P, Heinonen OJ, Nummela A, et al. (2023). Front Physiol 14:1128111. doi:10.3389/fphys.2023.1128111
27. Rosenblat MA, Watt JA, Arnold JI, et al. (2025). Sports Medicine 55(3):655–673. doi:10.1007/s40279-024-02149-3
28. Schnohr P, O'Keefe JH, Marott JL, Lange P, Jensen GB (2015). J Am Coll Cardiol 65(5):411–419. doi:10.1016/j.jacc.2014.11.023
29. Aengevaeren VL, Mosterd A, Braber TL, et al. (2017). Circulation 136(2):138–148. doi:10.1161/CIRCULATIONAHA.117.027834
30. Newman W, Parry-Williams G, Wiles J, et al. (2021). Br J Sports Med 55(21):1233–1238. doi:10.1136/bjsports-2021-103994
31. Matomäki P (2025). Eur J Appl Physiol 125(9):2401–2407. doi:10.1007/s00421-025-05843-w