Heart rate zones are all the rage for many runners. Various experts and authorities tout the benefits of different “zones,” though exactly how many zones there are—and what heart rates define each zone—are points of vociferous disagreement.
Instead of tackling the behemoth that is heart rate training in one post, this article is focused on one narrow but critical aspect of heart rate training: how and why heart rate zones are connected to LT1 and LT2, the first and second lactate thresholds.
My goal is to convince you that the entire point of heart rate zones—as they are currently used and advertised—is to be able to tell whether you are running below LT1, above LT1, or above LT2. The (purported?) benefits and (supposed?) scientific rationale of heart rate zones in popular training advice is almost always couched, either explicitly or implicitly, in the physiology of LT1 and LT2.
We will focus only on the connection between heart rate, heart rate zones, and LT1 & LT2 as key physiological signposts. Our goals are to answer the following questions:
- What is heart rate trying to measure?
- Which measurements of heart rate are most appropriate for quantifying training intensity?
- What is the purpose and underlying science behind heart rate zones as a concept?
Along the way, we will understand %HRmax, %HRreserve (%HRR), and some (but not all) of their flaws.
Some quick terminology and scope
I should note at the outset that I use the terms “LT1” and “LT2” with a bit of hesitation, since they are measurements, not underlying biological phenomena—it’s like the difference between an oral thermometer measurement and your true core temperature.
I am increasingly using the terms “T1” or “first threshold” for the underlying phenomenon that LT1 aims to capture, and “SSmax” or “steady-state max” (which I wrote an entire article about) for the underlying phenomenon that LT2 aims to capture. But, since almost nobody uses those terms, we will go with the more widely-used “LT1” and “LT2” for this article.
I’m going to be as charitable as possible to zone training—if you heard my appearance on the Strength Running podcast (apple podcasts link), you’ll know that I have plenty of criticisms of people who are obsessed with their heart rate zones. But here, I’m taking the premise of heart rate zones seriously.
I will not be covering whether zone training is “good” or “bad” on the whole, nor will I be going in-depth on the training impact of any particular zone—first, because the benefits that are claimed to be associated with certain zones vary from one zone model to the next, and second, because understanding where zones come from is a prerequisite for understanding what possible benefits might accrue from spending time in a particular zone.
As Renato Canova says, training is the answer to a proposal, not the proposal itself. So, in other words, you spend 30 minutes in “Zone 4”—that is the proposal, and that is the topic of this article. The adaptation of your body in response to that proposal is the training, and that is a topic we’ll save for another day.[1]
Finally, I will not be covering problems related to the measurement of heart rate, like cadence lock, wrist-based heart rate accuracy, cardiac drift, and the slow component of heart rate. We want to understand heart rate at its best—accurately measured.
This is the first article in a two-part series. The second article, coming later this week, is concerned with what heart rates actually correspond to LT1 and LT2, and how those heart rates vary from person to person.
But first, we need to understand how heart rate and heart rate zones are connected with LT1 and LT2.
The science of HRmax, HRrest, and HRR for runners
First, let’s make sure we’re on the same page about the underlying physiology behind heart rate zones.
The point of monitoring heart rate is to estimate (aerobic) metabolic intensity
Why is there such a mountain of research and popular content on heart rate anyways? Fundamentally, the reason is that heart rate is an objective estimate of your metabolic intensity—how hard your body’s energy systems (and specifically, your aerobic system) is working.
In contrast, something like rated perceived exertion, or RPE, is a subjective estimate of metabolic intensity.
On a more technical level, physiologists use heart rate as a surrogate for oxygen consumption, a.k.a. VO2.[2]
Since most people don’t have a metabolic cart lying around, heart rate is a more practical way of quantifying and controlling exercise intensity, especially in sports like trail running, elliptical machine use, and cross-country skiing, where objective external metrics of intensity, like speed or mechanical power output are not as readily available, or as meaningful, as they are in running or cycling (respectively).
HRmax: Maximal heart rate
HRmax is your maximal heart rate: the highest heart rate you can achieve during an all-out effort. Your heart rate during the last 30 seconds or so of an all-out race is often the best estimate of your HRmax; if you haven’t raced recently, the highest heart rate you attain in a tough interval session at 5k pace or faster—ideally with repeats lasting at least three minutes—is another good indicator.[3]
Contrary to popular belief, HRmax can change over time—but not in the way you’d expect. As you get in better shape, HRmax goes down, to the tune of three to seven percent. HRmax decreases as you get older, though the rate of decline is quite slow (roughly 0.6 bpm per year, but this number varies from person to person too!).
Most importantly, it is not possible to accurately predict HRmax given your age. Even state-of-the-art regression formulas have a 90% margin of error of ± 18 beats per minute (bpm), which is way too much error to be useful for training—and don’t get me started on “220 - age,” which performs even worse.[4]
From Nes et al. 2013
HRrest: Resting heart rate
HRrest is the lowest your heart rate gets in a normal context (i.e. at rest). Measuring HRrest is easy: just lie down for five minutes or so, then check your heart rate. HRrest is about 7 bpm lower when lying down vs sitting, but many physiology studies measure HRrest in a seated position anyways.
Like HRmax, there is massive individual variation in HRrest, even among healthy well-trained athletes. One study of NCAA athletes found a 90% range of 45–81 bpm (and a full range of 35–110 bpm!). While HRrest was 5.5 bpm lower in cross-country runners, even these athletes showed enormous individual variation.
Absolute heart rate values are meaningless for training
In the very early days of training, interval training pioneers like Woldemar Gerschler touted the benefits of specific heart rate targets (e.g. hitting 180 bpm on intervals, then beginning the next repeat when heart rate returned to 120 bpm). Today, you still see people quoting specific heart rates as workout or training targets, even when talking generally about all runners, not a specific individual.
Given what we’ve already learned about the individual variation in HRmax and HRrest, it should be obvious that it makes no sense to talk about an absolute heart rate (e.g. “zone 2 = 110–120 bpm”) in the context of training. This is even before considering the variation in LT1 and LT2 in relative terms.
So, let us dispense completely with absolute heart rates. If we have any hope of using heart rate for training, it must be normalized.
%HRmax: A common training metric with a number of flaws
By far the most common form of heart rate normalization is %HRmax: expressing heart rate relative to HRmax. For example, a heart rate of 120 bpm in a runner with a heart rate of 180 bpm is 67% HRmax.
%HRmax has many flaws—some of which we’ll get into in Part II of this series—but one flaw is obvious already: normalizing to HRmax only takes care of individual variation in HRmax, not HRrest. Moreover, it doesn’t really make sense to talk about, say, 10% or 20% HRmax, since these heart rates are going to be physiologically impossible for most people.
HRR: Heart rate reserve and %HRR: A less-common metric with fewer flaws
A better alternative is to normalize heart rate to HRmax and HRrest. This is the idea behind heart rate reserve, or HRR: a scale that runs from HRrest to HRmax.
Percent of heart rate reserve (%HRR) is essentially a “throttle” from 0% to 100%, with 0% set to HRrest and 100% set to HRmax.
Heart rate reserve correlates well with %VO2max, which is often used as the gold standard in physiology for relative metabolic intensity—much better than %HRmax, in fact.[5]
From Swain et al. 1997
Though the %HRreserve–%VO2max correspondence is not perfect, it’s good enough that %HRreserve is a commonly-used proxy for %VO2max targets in exercise science studies (e.g. studies comparing “HIIT training” to moderate-intensity continuous training).
Despite the advantages of %HRR, shockingly few physiology studies on runners use %HRR; most opt for %HRmax. Likewise, most heart rate “zone” recommendations use %HRmax. My best guess as to why is that calculating %HRR is slightly more complicated than %HR.
Formulas for heart rate reserve and %HR reserve in runners
The two relevant formulas are as follows:
%HRreserve = 100 × (Target HR - HRrest) / (HRmax - HRrest)
Target HR = HRrest + (%HRreserve / 100) × (HRmax - HRrest)
(I am working on a web app to make these calculations easy; I’ll update this article once it’s finished)
…which is indeed slightly more involved than %HRmax = Target HR / HRmax.
An aside: we are assuming you are accurately measuring your heart rate!
Now, everything we’ve covered so far assumes you have an accurate measure of your heart rate. The accuracy of optical heart rate estimate from GPS watches is another topic I’d rather leave for another day; suffice it to say that for some runners, wrist-based heart rate estimates can be adequate, but if you’re serious about heart rate training, you need a chest strap or (possibly) an arm band, as these devices are much more accurate and far less prone to cadence lock and other issues.
The science of LT1 and LT2 for runners
Having gotten heart rate out of the way, let’s briefly cover LT1 and LT2, the first and second lactate threshold. Both LT1 and LT2 are typically estimated using blood lactate measurements during an incremental treadmill test, so in the definitions below “slowest speed” should be understood as “slowest speed during an incremental treadmill test,” with the hope—but not the guarantee—that the same concept will translate to continuous runs at different speeds.
Exceeding LT1 causes a loss of efficiency and an increase in fast-twitch fiber recruitment
In terms of physiological metrics, the first lactate threshold (LT1) is defined as the slowest speed that produces blood lactate levels that are significantly elevated from baseline levels.[6] While it’s traditionally associated with blood lactate levels of 2.0 mM, LT1 can occur at a wide range of absolute blood lactate levels.
However, elevated blood lactate levels are merely the result of a deeper physiological shift. As you reach LT1, your slow twitch fibers start losing efficiency: they produce less force for a given amount of energy. In response, your body calls on more muscle fibers—many of them being fast-twitch fibers—to pitch in.
These fast-twitch fibers are the source of most of the additional ambient lactate in your bloodstream, but there’s another physiological sign you’ve crossed LT1: a gradual increase in oxygen consumption over the first 10–15 minutes at a given intensity. This increase—called the VO2 slow component—is the other canonical sign that you’ve departed the easy to moderate domain and entered the high-end aerobic domain.[7]
In terms of the physiology of running, when you exceed LT1, you face a markedly greater aerobic challenge, and your body generates a qualitatively different pattern of physiological response. The mechanisms of fatigue are also different above versus below LT1.
I have a much longer deep dive in the works on LT1 (join my email list to find out when it comes out!), but for now, this very brief overview should be enough to convince you that LT1 is a fundamentally important threshold along the spectrum of running intensities.
Alternative terminology for LT1
In some training systems, LT1 is sometimes also referred to as the aerobic threshold (AT or AeT): I used this terminology in my first book, though today I avoid it because it isn't very precise: speeds both above and below LT1 are fully aerobic!
In the physiology literature, LT1 is sometimes also referred to as just "lactate threshold" or LT, the gas exchange threshold or GET, the ventilatory threshold (VT or sometimes VT1), and, most confusingly, the anaerobic threshold (AnT). The term I like the most is simply "T1": first threshold.
(Note that many training systems use AnT and LT to refer to LT2, the second lactate threshold, hence the confusion!)
LT2 is an estimate of the highest intensity that produces a metabolic steady-state
LT2 is defined as the slowest speed that produces a “sudden and sustained” increase in blood lactate.[8] While it’s traditionally associated with blood lactate levels of 4.0 mM, LT2 can occur at a wide range of absolute blood lactate levels.
LT2 is really just an estimate of your steady-state max (SSmax)—the highest metabolic intensity that you can achieve while maintaining stable values of oxygen consumption, intramuscular acidity, and other key physiological metrics.
(I also have a separate in-depth article on steady-state max for runners, so check that out for much more on how LT2 relates to SSmax)
The short version is this: LT2 estimates SSmax within about 5–10%. If your LT2 perfectly estimates SSmax (not a guarantee!), then below LT2, you will be able to maintain stable values of oxygen consumption (VO2), blood lactate[9], and intramuscular acidity. Below LT2, fatigue will be driven by the same factors that drive fatigue at intensities just above LT1.
When you exceed LT2, you are in a fundamentally unsustainable metabolic state: oxygen consumption will spiral higher, eventually reaching VO2max; blood lactate will rise rapidly, and other key metrics (including heart rate!) will rise steadily towards their maximal limiting values.
Once you reach these limiting values (HRmax, VO2max, minimum tolerable intramuscular pH, etc.) you will become too exhausted to continue within a matter of minutes—or less.[10]
Physiologically speaking, above LT2, you are steadily drawing down your anaerobic energy reserves, and your body’s pattern of physiological response is qualitatively different than below LT2.
Again, see my SSmax article for details, but hopefully this quick summary is enough to convince you that LT2 is also a fundamental threshold that’s important for understanding running intensity.
Alternative terminology for LT2
In some training systems, LT2 is simply just called "LT" or "T pace." It is sometimes also called anaerobic threshold (AnT), as a contrast with "aerobic threshold."
In the scientific literature, LT2 is often called the lactate turnpoint (LTP), the onset of blood lactate accumulation (OBLA, or sometimes OPLA for "plasma"), and is sometimes estimated using the second ventilatory threshold (VT2) or the respiratory compensation point (RCP).
(I do not think VT2 and RCP are valid estimates of LT2, but that's a story for another day).
Note the possibilities for confusion here: if a coach says "anaerobic threshold" or "lactate threshold", they probably mean LT2. If a physiologist says the same thing, they probably mean LT1!
LT1 and LT2 demarcate two key physiological transitions between three domains of intensity
To underscore the differences: below LT1, blood lactate is indistinguishable from baseline levels, and oxygen consumption is stable over time. These paces pose only a modest challenge to your aerobic system.
Above LT1 (but below LT2), blood lactate is elevated but stable over time, and oxygen consumption rises gradually over the first 10–15 minutes of a run or repeat before stabilizing. Both of these changes are caused by a loss of efficiency in slow-twitch fibers, and a resulting increase in recruitment of additional muscle fibers, many of which are fast-twitch fibers. These paces pose a significant challenge to your aerobic system, but do not draw down your anaerobic energy reserves.
Above LT2, blood lactate and oxygen consumption spiral ever-higher, never reaching a steady-state, and you draw on your anaerobic energy reserves to meet the rising demands of the pace. Soon, your VO2 reaches VO2max, you run out of anaerobic energy, you become exhausted, and you have to stop.
Clearly, LT1 and LT2 demarcate two key transitions between three distinct domains of intensity, each of which generates a characteristic pattern of physiological response from your body.
The entire purpose of heart rate zones is to know where you are in relation to LT1 and LT2
Now, we’ve finally arrived at a point where we can connect LT1 and LT2 with heart rate and heart rate zones. The entire point of heart rate zones—and the explicit physiological claims made by all zone training models—is to garner specific fitness benefits by spending time at specific metabolic intensities that produce the desired pattern of physiological response.
Put more plainly, if you are running in “Zone 2” (in a five-zone model), the whole justification of Zone 2 and its benefits is that you are in a physiological region where the aerobic challenge to your body is low, your blood lactate levels are not elevated compared to baseline, you are not losing efficiency in slow-twitch fibers, and you are not recruiting a significant number of fast-twitch fibers.[11]
Likewise, if you are running in Zone 4 (in a five or six-zone model), the whole justification of Zone 4 is that your body is very close to, but not above, its steady-state max: your aerobic system is working at close to its steady-state limit, and you do not have lactate rapidly accumulating in your muscles or in your bloodstream.
So, you can use as many zones as you want, as long as you can ground them somehow in relation to LT1 and LT2.[12]
A few examples of zone models explicitly tied to LT1 and LT2
I said above that I’m not going to wade into arguments about whether there “should” be three, five, six, or seven zones, but here are a few representative examples of zone-based systems tied to LT1 and LT2.
All of these are reasonable, in the sense of tying zone boundaries to LT1 and LT2 somehow, but notice how they differ in terms of what each zone is supposed to represent.
Again, I'm not endorsing or criticizing any of these specifically; these are just illustrative examples.
Above: A five-zone system proposed in Jamnick et al. 2020
Above: A different five-zone system (from The Athlete Blog)
Above: A three-zone system (from Matthew Boyd Physio)
The big question: what heart rates correspond to LT1 and LT2?
So, for heart rate zones to be physiologically justifiable, they need to be tied somehow to LT1 and LT2. Different zone systems might disagree on the number of zones, but every scientifically based system had better anchor at least some zones to LT1 and LT2.
That conclusion leads to two critical questions: first, what heart rates correspond to LT1 and LT2? And second, can specific heart rate zones—as either %HRmax or %HRreserve—correctly position you below or above these key thresholds?
For the answer, stay tuned for Part II: Individual variation in heart rates at LT1 and LT2 in runners, and the implications for zone training, coming later this week.
Learn more about the science of running
If you enjoyed this article, subscribe to my email list! It’s the best way to find out when I’ve got a new article on training, exercise science, or when I have a new tool available like my threshold and CV pace calculator.
I also have a book, Modern Training and Physiology for Middle and Long-Distance Runners, that focuses on the science of performance and training for events from 800m to the 10k. Check it out!
Footnotes
[1] It is important to remember that “Run at X pace” is not the only, or even the best, way of improving the physiological capabilities that might manifest at “X pace.” An extreme example comes from research on so-called “sprint interval training”—the following cycling protocol is an effective way of boosting VO2max over a four-week period in sedentary adults doing exercise bike training: 5 x 30 seconds all-out with 3 minutes of unresisted cycling in between.
Clearly a short all-out intensity is far above VO2max in terms of nominal intensity, yet it provokes an effective stimulus to the body (in sedentary adults, of course). So, the sprint is the proposal; the VO2max increase in response is the training.
[2] Or, in some cases, heart rate is being used as a proxy for energetic expenditure (e.g. calories burned per hour), which is basically the same thing as VO2 for high-intensity running, but can differ by ~5% at lower intensities because of the different energetic yields from carbohydrates vs. fats.
[3] Some popular training resources recommend all-out uphill runs as a way to get an estimate of HRmax. Hills are nice because they essentially force you to run all-out, but people tend to gas themselves by running hard early, then slowing down after a minute or two. This tends to artificially decrease your heart rate. I prefer using workouts to get a HRmax estimate, but if you insist on a field test, here is a better protocol:
20 minutes progressive run, then 4 strides with jogging rest, then:
3 x (1 min at 3k effort, 1 min jog)
4 minutes done as (2 min at controlled mile/1500m effort + 2 min all-out)
The short repeats at the beginning help “prime” your heart rate, letting you go into the four-minute segment with an already-elevated heart rate, and doing the first two minutes backed off from maximal effort prevents you from getting out too hard, which is the most common mistake people make when field-testing their maximal heart rate.
[4] If you really insist on an age prediction formula, the range given by the following formulas (source, n=3,320 healthy adults) give the 5th, 50th, and 95th percentile heart rate values (in bpm) as a function of age (in years):
95th percentile: 229 - 0.64*age (in years)
50th percentile: 211 - 0.64*age
5th percentile: 193 - 0.64*age
Again, I actively discourage using formulas because, as you can see, the margin of error is massive! One in ten people will be outside the upper and lower predicted by the formula (indeed, my own HRmax is nearly below the 5th percentile prediction!).
[5] As the title of the linked paper indicates, heart rate reserve is actually more closely tracking %VO2reserve, which is the analogous “reserve” concept applied to VO2max and VO2rest. For runners, though, VO2rest is so low (3.5–5 mL/kg/min) compared to typical VO2max values (45–80 mL/kg/min) that the difference is pretty trivial. Accounting for VO2reserve is more important in sedentary people or people with chronic illnesses, whose VO2max can be as low as 15–20 mL/kg/min.
[6] As you might expect, what exactly constitutes “significantly elevated” is a matter of debate in physiology. Disagreements about how to define a “significant elevation” in blood lactate have real consequences for variability in LT1 as a percentage of HRmax and HRR, as we’ll see in Part II.
[7] Physiologists call these the moderate domain (below LT1) and the “heavy” domain (between LT1 and SSmax or LT2); I don’t find these terms to be particularly helpful for runners, though. A good cruise interval workout should feel anything but “heavy.”
[8] As above, what constitutes a “sudden and sustained” increase in blood lactate is up for debate. One paper counted over a dozen different definitions of LT1 and LT2!
[9] Well, blood lactate is almost stable. There is a narrow range of intensities below SSmax where blood lactate levels drift upward, but VO2 is stable over time. Nobody really knows why this happens, but it’s the strongest argument for why critical speed is a better estimate of SSmax than maximal lactate steady-state, or MLSS.
[10] About 50% of athletes—even international-level elites—reach total exhaustion more or less instantly when they reach their VO2max: they never show the textbook “plateau” in VO2.
[11] One specific purported benefit to Zone 2 that I cannot resist commenting on: Zone 2 is widely pointed to as the optimal training intensity for burning fat. Lactate itself is a powerful inhibitor of fat oxidation, so running above LT1—when lactate levels are significantly elevated above baseline, and when carb-hungry fast-twitch fibers come online—is clearly suboptimal if the desired proposal to your body is burning a lot of fat.
[12] Alternatively, you could argue “zones are completely arbitrary and don’t need to be connected with any specific acute physiological response from the body, but it is useful to have a name for the various different intensities that we use in training.”
This is more or less the idea behind full-spectrum percentage-based training: 90% of 800m pace is not producing the same acute response from the body as 90% of marathon pace, but we can talk about both of them as being race-supportive endurance for the goal race.
In the percentage-based view, physiology is a subordinate or instrumental tool along the path towards building the race we want. I have never heard anyone make this argument for heart rate zone training—zone frameworks always prioritize physiology and physiological stimuli as the primary goal, with race-specific considerations as an afterthought.
Related articles
New web app: Predicting LT1 pace and Zone 2 pace from 5k time
Designing a plyometrics program for improving bone strength in young runners
Three theories of tissue damage accumulation during running
Lecture: The science behind modern marathon training
What assumptions are baked into your race prediction model?
Tendons do not store energy for free
In windy conditions, running at a constant effort is usually better than running at a constant speed
A comprehensive guide to the science of cadence for runners
Understanding tissue loading, tissue damage, and running injuries
A great read and I am looking forward to part 2. Thanks!
If having pace zones normalized to critical speed pace is part of coach's way of prescribing training zones, would a second heart rate based zone system normalized to HRR be redandant or complementing the pace based zone system? I'm speaking about road running.
What is the difference between critical speed pace and LT2 as far as oxidative metabolic control is concerned. Do they always agree? I thought MLSS was not the gold standard ..?
Hi John,
Yet another masterpiece of an article—thank you for consistently sharing such valuable insights.
On a different note, I wanted to ask you a couple of questions regarding percentage-based full-spectrum training.
I’m currently in the process of designing a personalized 16-week marathon plan. Based on what I’ve gathered from your articles, the general, supportive, and specific phases could reasonably span 6, 4, and 6 weeks, respectively. You’ve astutely pointed out that the foundation of an effective program begins with identifying the key race-specific workouts (100% MP), followed by specific/speed endurance sessions, and finally, supportive speed/endurance work. I also understand that the "funnel" approach should be applied, gradually increasing workout specificity as race day approaches.
Given your extensive experience with Canova’s system—having tested and implemented it with numerous athletes—I was hoping you could provide a ballpark estimate of how many key, specific, and supportive workouts you would typically prescribe within each mesocycle. Additionally, if possible, I’d love to get a sense of how you would structure these sessions within a sample microcycle for each phase.
I greatly appreciate your time and insights!
Best regards,
Jorge B