Why Your Training Zones Are Wrong (And How to Fix Them Without a Lab)

There’s a good chance your Zone 2 isn’t Zone 2.

If you set your heart rate zones using 220 minus your age, or you let Garmin figure it out, you’re probably training in the wrong zones. Not slightly wrong. Potentially 10 to 15 bpm wrong. That’s the difference between an easy recovery run and a moderate effort that quietly accumulates fatigue, session after session, week after week.

And you’ll never know. Because you’ll feel “fine.” Until you don’t.

The 220 Minus Age Problem

The formula 220 minus age has been floating around since the 1970s. It was never meant to be a prescription. It came from observational data, a rough trend line drawn across a population. Dr. William Haskell, one of the people behind it, has said publicly that it was never intended for individual use.

Here’s why it fails you: the standard deviation on max heart rate predictions is 10 to 12 bpm. That means if the formula says your max HR is 175, your actual max could easily be 163 or 187. For a 50 year old, the formula predicts 170. But tested max heart rates for healthy 50 year olds range from about 155 to 190. That’s a 35 bpm window.

Now cascade that error through five training zones. Your “Zone 2 ceiling” could be someone else’s Zone 3 floor. You’re doing your easy runs too hard. Your threshold sessions are below threshold. Your interval targets are off. Everything downstream of a bad max HR estimate is wrong.

This isn’t a minor calibration issue. It’s a structural problem with how most athletes set up their training.

Garmin’s Auto-Detect Is Better. But Not Good Enough.

Garmin, Polar, and other platforms have moved beyond the simple formula. They use optical HR data, HRV readings, and workout analysis to estimate your lactate threshold and max HR. Credit where it’s due: this is a real improvement over a formula from the 1970s.

But it has limitations.

First, optical wrist HR sensors struggle with accuracy at high intensities. Exactly when you need precision most (threshold and above), the sensor is least reliable. Cadence lock, poor skin contact, and motion artifacts all introduce noise.

Second, auto-detection typically uses a handful of hard efforts to estimate your threshold. Maybe three or four workouts where you pushed into the right intensity range. That’s a small sample. Your threshold shifts with fitness, fatigue, heat, altitude, caffeine, sleep quality, and hydration. A snapshot from last Tuesday’s tempo run isn’t the full picture.

Third, most auto-detection algorithms don’t account for cardiac drift, the gradual rise in heart rate during sustained effort even at a constant pace. If your “threshold” was detected 35 minutes into a long run on a warm day, it’s going to read higher than your true resting threshold.

It’s better than nothing. It’s not good enough for serious training.

The MAF Method: Simple, Conservative, Incomplete

Phil Maffetone’s MAF formula (180 minus age, with adjustments) takes the opposite approach. Instead of estimating max HR, it estimates an aerobic ceiling. For a healthy 40 year old, that’s 140 bpm. Stay under that number for all your easy training.

MAF has real value. It forces athletes to slow down, which most people desperately need. The aerobic base it builds is legitimate. Many runners who adopt MAF see dramatic improvements in pace at the same heart rate over months of consistent training.

But 180 minus age has the same fundamental flaw as 220 minus age: it’s a population average applied to an individual. The adjustments (subtract 5 if you’re coming back from injury, add 5 if you’ve been training consistently for two years) are crude. They don’t account for your actual physiology.

A well trained 45 year old cyclist might have a true aerobic threshold at 155. MAF would cap them at 135. That’s not easy training, that’s walking pace. On the other end, a deconditioned 35 year old might have a true aerobic threshold at 130. MAF gives them 145, which has them grinding in a zone that’s building fatigue instead of base.

Any formula that doesn’t use your actual data is guessing. Some guesses are better than others, but they’re all guesses.

DIY Threshold Testing: What Actually Works

You don’t need a lab. A lab is nice. Lactate testing with a sports physiologist gives you precise, actionable numbers. But it costs $200 to $400 per session, and your thresholds shift as fitness changes, so you’d want to retest every 8 to 12 weeks.

Here’s what you can do yourself.

The 30 Minute Time Trial. Warm up for 15 minutes. Then go as hard as you can sustain for 30 minutes. Your average heart rate over the last 20 minutes of that effort is a solid approximation of your lactate threshold heart rate (LTHR). This is the method Joe Friel popularised, and it works. It hurts. But it works.

From LTHR, you can set your zones with reasonable accuracy. Zone 2 sits at roughly 70 to 80% of LTHR. Threshold work is 95 to 105%. These aren’t perfect, but they’re anchored to your physiology, not a formula.

The Talk Test. Low tech, surprisingly effective. If you can hold a conversation comfortably, you’re in Zone 2. If you can speak in short sentences but not paragraphs, you’re approaching threshold. If you can only manage single words, you’re above threshold. It’s imprecise, but it catches the athletes who think they’re running easy when they’re actually in no man’s land.

DFA Alpha1. This is the emerging gold standard for non-invasive threshold detection. DFA alpha1 (detrended fluctuation analysis of heart rate variability) measures the fractal correlation properties of your heartbeat intervals during exercise. When alpha1 drops below 0.75, you’ve crossed your aerobic threshold. When it hits 0.5, you’re at your anaerobic threshold.

The catch: you need a chest strap HR monitor (optical sensors don’t have the resolution), an app that calculates DFA alpha1 in real time (HRV Logger, Fatmaxxer, or similar), and some patience to learn the method. But once you have it dialled in, you can verify your zones during any workout. No time trial suffering required.

The Real Cost of Wrong Zones

Getting your zones wrong doesn’t just mean suboptimal training. It means failed training.

Polarised training (the approach backed by the most evidence for endurance performance) requires that roughly 80% of your training volume sits below your aerobic threshold and about 20% sits above your anaerobic threshold. The middle zone, the grey zone between aerobic and anaerobic threshold, should get minimal time.

If your zones are set 10 bpm too high, your “easy” runs are grey zone runs. You’re not recovering. You’re not building deep aerobic base. You’re accumulating moderate stress that feels manageable day to day but compounds into stagnation or overtraining over weeks and months.

Simultaneously, your “threshold” sessions are actually sub-threshold. You’re not getting the stimulus you think you’re getting. You’re putting in the effort without the adaptation.

This is the most common pattern in recreational endurance athletes: too hard on easy days, too easy on hard days. It’s the grey zone trap. And wrong zones are the single biggest reason athletes fall into it.

Where This Is Going: Data That Calibrates Itself

Here’s what a coach does when they’re really good at their job: they look at months of your training data, correlate heart rate with pace and power across different conditions, observe where your performance inflects, and adjust your zones accordingly. They don’t set zones once and forget them. They treat zones as a living, shifting thing.

Now imagine doing that with data, automatically.

If you have heart rate data from hundreds of workouts, paired with pace or power, you have the raw material to identify where your thresholds actually sit. The relationship between heart rate and performance output isn’t linear. There are inflection points. Those inflection points are your thresholds.

With enough longitudinal data (weeks and months, not days), you can observe how those thresholds shift with training load, recovery, season, and fitness. You can detect cardiac drift patterns. You can spot the difference between a genuine fitness gain and a day where caffeine and cool weather made everything feel faster.

This is what P247 is building toward. Not another formula. Not another single-workout auto-detect. A synthesis layer that takes the data you’re already generating from your watch, your chest strap, your power meter, your training log, and uses the aggregate picture to tell you where your zones actually are. Today. Not where they were when you did a threshold test three months ago.

The data already exists on your wrist. What’s missing is the layer that connects it, analyses it longitudinally, and turns it into zones that update as you do.

Your watch knows your heart rate. Your app knows your pace. But nobody is connecting those signals across hundreds of sessions to find your actual thresholds. That’s the gap.