HIIT: How Hard, How Long, and How Often?

The word people skip in "high-intensity interval training" is the first one. It means repeated near-maximal efforts — breathing too hard to get a sentence out — separated by recovery long enough to let you do it again. Whether you run, cycle, row, or do burpees is beside the point. That intensity is usually written as "85–95% of maximum heart rate", but that is a convenient convention: a fixed percentage of a maximal anchor does not reliably separate distinct physiological states [1], and your maximum was probably estimated from your age anyway. For where these lines come from, the heart-rate zones guide has it.

The "cannot speak in sentences" criterion has real backing — the talk test has been validated against the ventilatory threshold. But note where it applies: the review that recommends it states plainly that it "may not be practical for high-intensity interval training". [2] It is the right tool for confirming your easy days are easy, not for grading yourself mid-way through the third interval.

There is a widely mis-quoted number to be careful with here. A meta-analysis of 28 controlled trials and 723 participants did find intervals ahead — by a small margin. Against no-exercise controls, steady endurance training raised VO₂max by 4.9 ml/kg/min (±1.4) and intervals by 5.5 (±1.2); head to head, intervals led by 1.2 ml/kg/min (±0.9), which the authors word as a "possibly small beneficial effect". [3] So: both work well, intervals edge it, and nothing here is double.

The real difference is the time cost, and there it is large. A twelve-week trial: nine sedentary men did three 20-second all-out sprints inside a 10-minute session, ten others cycled steadily for 45 minutes. Both raised VO₂peak by 19%, with similar improvements in insulin sensitivity and muscle mitochondrial content — at a fifth of the exercise volume and time commitment. [4] That is what intervals actually buy.

While we are here, switch off the calorie story. Excess post-exercise oxygen consumption — the afterburn — is real, but a review of the literature puts it at 6–15% of the net oxygen cost of the session itself, even for the protocols long enough to prolong it; it cannot outweigh the work that caused it. The same review notes the figure may need raising for intermittent protocols, where the oxygen debt repaid during the rest intervals usually goes uncounted. [8] That moves the number, not the conclusion: what intervals save is time, not a free lunch on the energy ledger.

The 4×4 is worth reading together with the people it was tested on: 27 postinfarction heart-failure patients, mean age 75, training three times a week for 12 weeks. [5] Running it in that population means running it supervised; running it as a healthy reader means finding an anchor the trial had and you may not — its intensity was 95% of a peak heart rate that was actually measured, not an age-predicted maximum — apply the protocol to an estimated number and you are no longer running the one that was tested. The Tabata numbers deserve a plain statement as well: the 1996 original had seven subjects on a cycle ergometer at roughly 170% of VO₂max for 20 seconds, resting 10, repeated seven to eight times for four minutes total. [6] Almost no class carrying the name runs anywhere near that intensity.

Recovery is not dead time, but memorising an energy-system ratio table is not enough either: the same 40/20 clock becomes two different sessions at fixed versus self-selected pace. For what 30/30 and 40/20 actually change, which suits a first session, and when neither fits, see the interval-timing guide.

For nearly everyone, once or twice a week — a number reasoned rather than measured, as noted above. The reasoning: interval sessions cost considerably more recovery than easy aerobic work, and the extra one competes for the same budget as your lifting, your sleep, and the quality of the next interval session. A third session usually is not one more good session; it is two worse ones. When you want to add, add easy minutes — intervals sharpen the edge, the aerobic base is the steel.

This is also why twenty minutes of intervals a week does not replace the rest of your cardio. On VO₂max alone intervals go further than you would think, but the rest of what cardio buys — blood sugar, mood, resilience to fatigue, and the dose-response curve tied to weekly minutes — accrues with accumulated time rather than peak intensity. That curve belongs to the cardio guide, which is where the argument for it lives. Intervals do not replace volume; they make a small slice of it more efficient.