The Tanaka equation, applied

Every heart-rate zone you have ever seen on a watch is arithmetic wearing a precision costume. The formulas are real, published, and useful — and each one carries an error bar wide enough to matter. This page does the math out loud: where the numbers come from, what your zone-2 band looks like at your age, and when to let the equation steer versus when to let the talk test decide.

Where the Formulas Come From

The arithmetic has a history, and the history matters because it shows how much each formula is actually claiming. "220 minus age" is the folklore — it was never built on a rigorous sample, yet it ships inside almost every fitness tracker on the market. In 2001, Tanaka and colleagues pooled data from thousands of adults and proposed a better general fit: 208 − 0.7 × age (Tanaka et al., Journal of the American College of Cardiology, 2001). A decade later, analyzing healthy women in the St. James Women Take Heart Project, Gulati and colleagues found women's max heart rates ran systematically lower than the general equations predicted, and proposed 206 − 0.88 × age for women (Gulati et al., Circulation, 2010).

The honest framing: these are all estimates of a group average, refined with better data but still an average. The slope of the equation (the 0.7 versus the 0.88) describes how the group's max declines per year — roughly one beat per year by any version. Your personal max is a measured fact that exists only in a lab. Every formula is a guess around that fact, and the width of the guess is the central number of this whole subject.

The Math, Worked by Hand

The equation is small, so do it yourself rather than trusting the watch. Take Tanaka: 208 − 0.7 × your age. At 40, that is 208 − 28 = 180 beats per minute of estimated max. At 50, 208 − 35 = 173; at 60, 208 − 42 = 166. Then the zone-2 band is commonly taken as 60–70% of that max, so at 50 the estimated band is 0.6 × 173 to 0.7 × 173, or about 104–121 beats per minute.

• 🧮 Do the subtraction in your head, not the watch. 208 minus seven-tenths of your age. If you can do it while waiting for coffee, you will spot a mis-set device immediately.
• 📏 Band, not a point. Zone 2 is a range roughly twenty beats wide; aiming at one number turns an estimate into a false precision.
• 👩 For women, run the Gulati version too. 206 − 0.88 × age usually lands a few beats lower; where the two formulas disagree, expect your real number to sit somewhere between them, not exactly on either.
• 🗣️ Then let the talk test arbitrate. The equation gives you a starting target; speech tells you whether that target is right for you on a given day. The talk-test page is the other half of this instrument.

Your Zone-2 Band, by Age

The table shows the worked numbers across the adult range, using the Tanaka equation and the 60–70% band. Read the column as a region to start from, not a commandment — your real ceiling is what the talk test says it is, and the band below is what the math says it probably is.

Notice what the table is quietly telling you: the two formulas disagree by only a few beats at any age, and that disagreement is tiny next to the ±10 bpm spread on top of either one. Losing sleep over which formula your watch uses is misplaced effort — the formula choice barely matters, and the spread is the real story.

The ±10 Problem, Quantified

Here is the number that should govern your trust in every screen on your wrist. The validation literature reports a standard deviation of roughly ten beats per minute around age-predicted max. Concretely: if the formula says your zone-2 ceiling is 124, your real ceiling is plausibly anywhere from about 114 to 134. Two-thirds of people land within that band; about one in three sits further out. That is not a rounding error — it is larger than the entire width of a zone.

The practical consequence is not nihilism — it is the division of labor the whole series runs on. The watch is a trend recorder and a red flag; it is not a governor. When the screen and your throat disagree, the throat wins, because the nonsense in the watch is a known quantity and the throat is a live measurement. The parent page's ranking of the three tests makes the same call.

Karvonen: The Better Backup

If you want the estimate to work harder, switch from a percentage of max to a percentage of heart-rate reserve — the gap between your resting heart rate and your max. The Karvonen method, published in the 1950s and still the standard adjustment, calculates a target as (max − resting) × intensity + resting (Karvonen et al., 1957). Because it accounts for a low resting heart rate — the signature of a trained cardiovascular system — it tends to track individual effort better than a flat percentage of max, particularly for fit people.

Worked for a 45-year-old with a resting heart rate of 60: estimated max from Tanaka is about 177. Reserve is 177 − 60 = 117. The zone-2 band at 60–70% of reserve is 0.6 × 117 + 60 = 130 up to 0.7 × 117 + 60 = 142. Note the reserve numbers land higher than the plain 60–70% of max (106–124 from the table) — that gap is normal and expected, and it is the reserve method correcting for the fact that this person's heart runs slow at rest. Take your own resting pulse a few mornings in a row, average it, and run the arithmetic once per year; you do not need to re-derive it every session.

The Watch as Co-Pilot, Not Pilot

• ⌚ Heart rate lags effort. Your pulse at a given pace takes 30–60 seconds to settle, so a watch reads late — another reason the speech check is quicker than it.
• 📉 Wrist sensors wobble. Optical sensors misread during certain movements and in the cold; a chest strap is more reliable, but still formula-bound upstream.
• 🌡️ Heat and dehydration inflate the reading. The same effort produces a higher pulse in warm conditions — the watch will look like overreaching when it is just the weather.
• 📈 The trend is the treasure. Across weeks, resting heart rate and the effort at a given pace shift in the direction of fitness; that slow trend is what the watch is good at, not the per-minute number.
• 🩺 Medication changes the math. Beta-blockers and other heart-rate-affecting drugs blunt the very signal these formulas use; if you take them, intensity guidance belongs with your clinician, and nothing on this page should override that.

Heart-Rate Math Questions, Answered Briefly

• ❓ Which formula should I put in my watch? Tanaka for most people, Gulati for women, and treat whichever you choose as a starting guess. The variance between formulas is small compared with the ±10 bpm spread both share.
• 😰 My watch's zone 2 feels far too hard. That is the formula's spread colliding with your real threshold. Slow down until the talk test passes; the number will often not match — and that mismatch is the normal condition, not a malfunction.
• 🫀 Is my resting heart rate a useful input? Yes, and it is the most honest number you can collect without a lab. Track its trend across weeks; a drift in the right direction is a genuine adaptation signal.
• 🧪 What would a lab actually measure? A graded exercise test to volitional fatigue gives a real max and real ventilatory thresholds — the VO2-testing page covers how to read the result. It is worth doing once if you can; the site's VO2 max testing page explains what it does and does not change.
• 🙅‍♂️ Why does my watch disagree with itself across apps? Because each app picks its own formula and band boundaries. Same math, different defaults — another reason to do the subtraction yourself and treat every screen as provisional.

Related Topics