October 27, 2026

What Is Anaerobic Threshold? Six Tests, Different Numbers

Anaerobic threshold explained: the definition, six ways to estimate it, and a worked step test where 4 mmol/L and a curve method land 28 W apart.

Kristian Hoffmann · SaaS founder and operator

The anaerobic threshold is the highest exercise intensity you can sustain while lactate appearing in the blood is still cleared as fast as it is produced; above it, lactate keeps rising. That is the core answer to "what is anaerobic threshold". The catch is practical: no single test measures this point directly. Lactate threshold, OBLA, maximal lactate steady state and ventilatory threshold are all estimates of it, and on the same athlete they can land 30 watts or more apart.

The definition, and why it is a concept rather than a number

Svedahl and MacIntosh define the anaerobic threshold (AnT) as "the highest sustained intensity of exercise for which measurement of oxygen uptake can account for the entire energy requirement", and add that at the AnT the rate of lactate appearance in the blood equals its rate of disappearance (Svedahl & MacIntosh, 2003). A second, shorter version comes from medical reference works: the level of oxygen consumption above which aerobic metabolism is supplemented by anaerobic processes (ScienceDirect Topics).

Read those two definitions carefully. Neither one names a lactate value, a heart rate or a wattage. They describe a balance point.

The same review lists the tests developed to locate that balance point: maximal lactate steady state, the lactate minimum test, lactate threshold, OBLA, the individual anaerobic threshold and the ventilatory threshold. Each one, the authors write, "has consistent and predictable error depending on protocol and the criteria used". Their recommendation is blunt: call the result by the name of the method that produced it, not "the AnT".

That one sentence explains most of the confusion online. When a coach, a watch and a lab report all quote an "anaerobic threshold" for the same athlete and the three numbers differ, nobody has necessarily measured badly. They have used different definitions.

Is "anaerobic" even the right word?

Only loosely. The same 2003 review states there is no evidence that lactic acid production above the AnT results from inadequate oxygen delivery. Muscle does not suddenly run out of oxygen at a threshold. The name stuck for historical reasons, and in endurance sport today "anaerobic threshold", "LT2" and "second threshold" are often used for the same upper boundary.

Six ways to estimate the anaerobic threshold

Each method answers a slightly different question. The table sorts them by what you need to run them, which is usually what decides which one you actually use.

Three of the six can be run from a single home step test with a portable lactate meter. That is the space where LactateThreshold works: you enter the stage data, and the analysis places the curve-based thresholds side by side so the method is visible rather than hidden. The step-by-step detail of those methods is in Calculate lactate threshold: methods and formulas explained.

Where the 4 mmol/L figure comes from

OBLA is the oldest fixed-value shortcut. In Sjödin and Jacobs' 1981 study of 18 Stockholm Marathon runners, the running velocity at a blood lactate of 4 mmol/L (VOBLA) accounted for 92% of the variation in marathon velocity. The figure is a group-level convention, not a personal constant. Two athletes can sit at very different points of their own curve when both reach 4 mmol/L.

Ventilatory versus lactate thresholds

Ventilatory thresholds come from breathing data, lactate thresholds from blood. Both estimate the same underlying transition, but they are measured on different signals. If your old report says "VT2" and your new one says "LT2", treat them as two separate baselines rather than a change in fitness.

A worked example: one test, three "anaerobic thresholds"

The following numbers are a fictional example built to show how the method choice moves the answer. Same cyclist, same test, same blood samples.

Using straight-line interpolation between stages:

• First clear rise (baseline + 0.5 mmol/L = 1.6): between 210 W (1.5) and 240 W (2.2) → 210 + (0.1 / 0.7) × 30 = 214 W.
• Baseline + 1.5 mmol/L (= 2.6): between 240 W (2.2) and 270 W (3.4) → 240 + (0.4 / 1.2) × 30 = 250 W.
• Fixed 4 mmol/L (OBLA): between 270 W (3.4) and 300 W (5.6) → 270 + (0.6 / 2.2) × 30 = 278 W.

The two "upper threshold" candidates, 250 W and 278 W, are 28 W apart. That is about 10% of the OBLA value, from identical data. Neither number is wrong. They answer different definitions, which is exactly the warning in the Svedahl review.

The athlete whose test never reaches 4 mmol/L

Now a second fictional rider, stopping at exhaustion on the last stage:

The fixed 4 mmol/L method has no answer here without extrapolating beyond the measured data. The baseline + 1.5 mmol/L method (2.4) gives 232 + (0.5 / 0.9) × 24 = 245 W, inside the measured range. Low peak lactate values are a common reason a fixed-value threshold misleads: the 4 mmol mark may sit above anything the athlete reached in the test. If your report shows an OBLA value higher than your final stage, it was extrapolated, and you should read it with that in mind.

For a deeper look at how far curve-fitting methods can spread on one data set, see Step Test Analysis: 4 Threshold Methods, 33 Watt Difference.

What typically goes wrong when people compare threshold numbers

Most "my threshold dropped" stories are comparison errors, not fitness changes. The recurring ones:

1. Method switched between tests. A lab used OBLA last spring, your home analysis uses a curve-based method now. The 28 W gap in the example above can appear with no physiological change at all.
2. Protocol switched. Stage length and step size change the shape of the curve. Svedahl and MacIntosh describe the error of each method as dependent on protocol, so a 3-minute-stage test and a 5-minute-stage test are not interchangeable.
3. Watch estimate compared with a blood test. A wearable "anaerobic threshold" is a model output from heart rate or pace data, not a lactate measurement. Use it as its own trend line.
4. Reading the threshold as a heart rate first. Heart rate drifts with heat, fatigue and caffeine. Record the power or pace at threshold first and the heart rate second.

The decision rule is simple: compare a threshold only with a value produced by the same method, the same protocol and, ideally, the same meter.

Testing in November and December

For many endurance athletes the weeks after the autumn race season are a quieter block, and indoor trainers or treadmills become the default. That is a practical window for a baseline test, because indoor conditions are easier to repeat than outdoor ones: same trainer, same room, no wind.

What to record so the retest after the winter block is comparable:

• Ergometer or treadmill model, and calibration state of a smart trainer
• Room temperature
• Stage length, step size and start load
• Lactate meter model and strip batch
• Time since last meal and last hard session

If the winter test is indoors and the spring test is outdoors, you are comparing two protocols. Plan both tests on the same setup.

How this relates to your training zones

Threshold values become useful once they feed zones. LactateThreshold derives LT1, LT2 and zones from your entered step-test data; the reasoning behind the zone boundaries is laid out in Lactate threshold calculator: estimate LT1, LT2 and training zones. How hard to train relative to those numbers is a coaching decision. If you have a heart condition, are returning from illness or are unsure whether maximal testing is appropriate for you, clear that with a physician before running a step test to exhaustion.

FAQ

Is anaerobic threshold the same as VO2 max?

No. VO2 max is the highest rate of oxygen uptake you can reach; the anaerobic threshold is the highest intensity you can sustain before lactate keeps accumulating. Ghosh's review reports that the anaerobic threshold correlates more strongly with distance-running performance than VO2 max does, because holding a high fraction of VO2 max for a long time delays metabolic acidosis (Ghosh, 2004).

What is a good anaerobic threshold?

There is no single good value, because the number depends on the method, the sport and the athlete's body mass. A more useful question is whether your own threshold, measured the same way, moves over a season. Compare power or pace at the same method, protocol and meter rather than against someone else's report.

How do I improve my anaerobic threshold?

Ghosh's review states that training at or a little above the anaerobic threshold intensity improves both aerobic capacity and the anaerobic threshold level. How much of that fits your week depends on your history, goals and recovery, so set the actual plan with a coach. A repeat test with an identical protocol is how you check whether anything changed.

Is anaerobic threshold the same as lactate threshold?

They overlap but are not identical. "Lactate threshold" is one of several measurement approaches used to estimate the anaerobic threshold, alongside OBLA, maximal lactate steady state and ventilatory threshold. Svedahl and MacIntosh recommend naming the method you used instead of calling every result "the anaerobic threshold".