How to Find Your Max Heart Rate: Formulas, Field Tests and Their Error Bars

Your maximum heart rate is the highest rate your heart can sustain during all-out effort. It is not a fitness marker, it cannot be trained upward, and it tells you nothing about how good an athlete you are. Its only real job is to act as the anchor from which training intensities are calculated, which is why getting it roughly right matters and getting it exactly right matters less than people assume.

The honest answer to the question in the title is that you cannot know your maximum heart rate from your age. Every age-based formula produces a central estimate for a population, and individuals scatter around that estimate by about ten beats per minute in either direction. What follows is where the three common formulas came from, why the choice between them matters less than the error bar around any of them, how a field test works, and why a genuine maximal test is a medically screened procedure rather than a weekend project.

Three formulas, and where each one came from

The formula almost everyone has heard is 220 minus your age. It is usually attributed to Fox, Naughton and Haskell, writing in 1971, and its origins are less rigorous than its ubiquity suggests. Robergs and Landwehr traced its history in the Journal of Exercise Physiology Online (5:1-10, 2002) and found it was not a regression fitted to a dataset at all. It was a line drawn through pooled observations from roughly a dozen sources, offered as a convenience, and it acquired the status of fact through repetition rather than validation. Robergs and Landwehr concluded it had no acceptable scientific merit as a research tool, while conceding it survives as a rough guide because it is trivially easy to remember.

Tanaka, Monahan and Seals produced the replacement most laboratories now use, in the Journal of the American College of Cardiology (37:153-156, 2001). They ran a meta-regression across 351 published studies covering more than eighteen thousand subjects, then validated the resulting equation prospectively in a separate laboratory cohort of 514 healthy people. The result was 208 minus 0.7 times age. They reported that the relationship was independent of sex and, notably, independent of habitual physical activity status, which is to say fit people do not have higher maximum heart rates than unfit people of the same age.

Gellish and colleagues came at it from a different direction in Medicine and Science in Sports and Exercise (2007). Rather than pooling cross-sectional snapshots, they followed adults longitudinally through a university fitness programme, accumulating repeated measurements on the same people across many years. Their linear equation, 207 minus 0.7 times age, has the same slope as Tanaka's and sits one beat lower. The convergence of two independent methods on the same slope is the most reassuring thing about either.

The formulas disagree in a specific, predictable direction

Because Fox has a slope of one beat per year and the other two have a slope of 0.7, they diverge at both ends and cross in the middle.

Read that carefully, because it is routinely stated backwards. Fox reads higher than Tanaka for anyone under 40, the two agree exactly at 40, and Fox reads lower than Tanaka for anyone over 40. A 20-year-old given 200 by the old rule is being handed a target six beats above the better estimate. A 70-year-old given 150 is being told to stop nine beats short of where the evidence puts them, which for an older adult trying to train hard enough to matter is the more damaging error of the two.

That older-adult underestimate is the main reason the field moved. It is also why our heart rate zone calculator uses Tanaka rather than the version printed on gym posters, and displays the Fox figure alongside it so the size of the disagreement is visible rather than hidden.

Individual variation is larger than the gap between formulas

Here is the part that reframes the whole exercise. Tanaka and colleagues reported a standard deviation of roughly 10 beats per minute around their regression line. That is not measurement noise; it is genuine biological spread between people of the same age.

Take the 40-year-old whose predicted maximum is 180 under every formula in the table. About two-thirds of 40-year-olds will have a true maximum somewhere between 170 and 190. Roughly one in twenty will fall outside 160 to 200. Those are not exotic outliers. In a gym class of thirty people, one or two of them will have a true maximum twenty beats away from the number on their watch.

Now compare the scale of that spread against the largest disagreement in the table above, which is nine beats between Fox and Tanaka at age 70. The gap between the best and worst formula is smaller than one standard deviation of the person-to-person variation that every formula shares. Choosing the better equation is worth doing, and it is nowhere near sufficient.

The propagation downward is what actually bites. If your true maximum is 190 and your watch assumes 180, every zone boundary is set about seven beats too low, and you will spend months believing an easy session was a moderate one. The five-zone scheme and how the bands are drawn is covered in what each heart rate zone trains; the point here is simply that all of it inherits whatever error sits in the anchor.

What a field test can and cannot give you

A measured value beats an estimate, with two large conditions attached.

The first is medical. A maximal exercise test is a deliberate push to volitional exhaustion, and it is the single most cardiovascularly demanding thing most people will ever do on purpose. In clinical and research settings it is conducted with supervision, a graded protocol and often ECG monitoring, precisely because that is when latent cardiac problems declare themselves. This is not something to attempt casually. If you have known or suspected cardiovascular, metabolic or renal disease, if you have any symptoms suggestive of them, if you are over about 40 and have been sedentary, or if you take medication affecting heart rate or blood pressure, the decision to perform a maximal test belongs to a clinician who knows your history, not to an article. Chest pain, unusual breathlessness, palpitations or light-headedness during exercise means stopping and seeking medical advice, not repeating the test with more determination.

The second condition is that field tests systematically under-read. You cannot exceed your true maximum, so any number a test produces is a floor rather than a measurement. Reaching it requires a level of self-inflicted discomfort that most people cannot access on the first attempt, and motivation, pacing skill and how recently you ate all affect the result. A first test almost always lands short.

If a test is appropriate for you

Assuming clearance, the protocols that work share a structure: a long warm-up, several progressively harder efforts to fatigue the system, and a final effort taken to genuine failure.

A common running version is fifteen to twenty minutes of easy jogging, then three efforts of about three minutes up a moderate hill, jogging back down between them, with the last one run as hard as it is possible to run and finished with a sprint. Peak heart rate almost always appears in the last thirty seconds or the first few seconds after stopping. A rowing equivalent is a 2000 m piece with a sustained final 500. On a bike, expect a value three to five beats lower than your running figure, because a smaller active muscle mass and a supported position both suppress peak heart rate. Swimming is lower again. The number is mode-specific, which means a cyclist and a runner are not comparing the same quantity.

Use a chest strap. Optical wrist sensors read blood flow through skin and become unreliable at exactly the intensities the test depends on, sometimes latching onto cadence instead of pulse. Record the peak from the trace and discard obvious artefacts. A two-second spike thirty beats above the surrounding line is noise, not a reading. Repeat the test two or three times across several weeks, keeping the highest genuine value.

The quieter way to find it

There is a version of this that requires no test at all, and for most people it is the better option: pay attention to the highest number you have ever legitimately seen.

Wear a chest strap during hard sessions, races, and the last kilometre of anything competitive. Log the peak. Over a season of genuinely hard efforts, the highest reliable value you record is a lower bound on your true maximum, and if it exceeds your age-predicted figure, your age-predicted figure is simply wrong and should be replaced. This costs nothing, carries no additional risk beyond the training you were already doing, and accumulates evidence rather than resting on one anxious afternoon.

Several things move the number in ways worth knowing. Beta blockers and some calcium channel blockers lower both resting and maximum heart rate substantially, which makes every age-predicted figure meaningless for the people taking them. Heat, altitude, illness, dehydration and accumulated fatigue all shift peak values, usually downward. And training does not raise maximum heart rate. Endurance training lowers resting heart rate and raises stroke volume, which is why fitness improves; the ceiling itself declines gently with age regardless of what you do. If you want the number that actually responds to training, that is VO2 max and what changes it, not this one.

Anchoring intensity without trusting the anchor

The most useful conclusion is that maximum heart rate deserves less weight than the calculators built on it imply.

Perceived effort and the talk test cost nothing and require no calibration: an easy aerobic effort permits complete sentences, a threshold effort permits a few words, and above that permits none. For structured endurance work, threshold-anchored methods are more robust than maximum-anchored ones, because your threshold is measurable directly: the average heart rate across the final twenty minutes of a thirty-minute all-out time trial is a serviceable estimate. Threshold is also the boundary that actually governs how long you can hold a pace. The case for and against spending most of your time below it is examined in Zone 2 training.

If you are new to endurance work, none of this needs solving before you start; how to begin running without breaking down is a question about progression rate, not about heart rate ceilings. And if you want to see how a session's intensity converts into an energy cost, the calories burned calculator uses population-average MET values, which carry their own generous error bars for the same reason this number does: it is describing an average person, and nobody is one.