Critical Speed Running: Test, Training & Pacing

How fast can you run before your legs give out? Critical speed (CS) answers exactly that question — it is the fastest pace you can sustain for 30 to 60 minutes before lactate and fatigue rise uncontrollably. Unlike a lab-based lactate test, finding it costs nothing but three hard minutes on a track. This guide explains the science behind it, how to measure your CS, and how to use it for training and race pacing.

What is critical speed?

Critical speed comes from the critical power concept in exercise science (originally Monod & Scherrer, 1965). The idea: duration and performance are linked by a stable mathematical relationship. The longer you run, the slower the maximum sustainable pace — but this decline flattens out at a specific value. That asymptotic value is your critical speed.

Physiologically, CS marks the boundary between two metabolic states:

• Below CS: lactate production and clearance stay in balance (steady state). You can hold the pace for a long time; fatigue builds slowly.
• Above CS: lactate, hydrogen ions and inorganic phosphate accumulate inevitably. Time to exhaustion is finite — and the further above CS you run, the shorter it gets.

Functionally, critical speed sits exactly where the anaerobic threshold from a lactate test or LT2 sits. The difference: you determine CS without blood samples, without a lab and without expensive equipment — just a watch and a willingness to hurt.

Typical ballpark figures: a recreational runner with a 50-minute 10K has a CS around 8:00–8:30 per mile (5:00–5:15/km). Ambitious runners at 40 minutes for 10K sit around 6:30–6:50 per mile (4:05–4:15/km), while elite endurance athletes run well under 5:40 per mile (3:30/km).

W′: the second number almost nobody knows

The CS model has a second parameter: W′ ("W prime"), the anaerobic work capacity available above critical speed. Think of W′ as a battery that drains above CS and recharges below it. Run far above CS and the battery empties fast; stay just above it and you can go a long way.

In practice, two runners with identical CS can have completely different race profiles. One has a large W′ and shines from 1,500 m to 5K; the other has a small W′ but a high CS relative to their aerobic base — the born marathoner. Knowing your W′ tells you how aggressively you can dose intervals above CS without blowing up mid-session.

The 3-minute test: measuring CS without a lab

The simplest protocol is the 3-minute all-out test. It produces a surprisingly valid CS estimate — studies typically show deviations under two percent versus multi-trial protocols. Here is how to run it:

• Preparation: Be fresh. No hard training in the 48 hours beforehand; ideally an easy day before. Location: a track, flat bike path or level road without traffic lights or junctions.
• Warm-up: 15 minutes of easy jogging, 4–5 strides of 20 seconds, then 3 minutes walking. The strides matter — without them you will start too cautiously.
• The test: Run absolutely maximally for 3 minutes. Start fast (roughly 5K race pace), hold it, and empty the tank in the final minute. The last 30 seconds should feel like a finishing sprint that cannot get any faster.
• Analysis: Your average speed across the full 3 minutes sits slightly above your CS. Rule of thumb: CS ≈ 95–97 % of your 3-minute average speed. If you averaged 4:00/km, your CS is roughly 4:07–4:12/km.

The most common error: starting too conservatively. Running the first minute 15 seconds slower than possible skews the result downwards. The test is called "all-out" for a reason — it is unpleasant, but it is only three minutes.

Alternative: the two-trial method

For more precision, use two time trials on separate days — for example 1,200 m and 3,600 m, both run maximally. From the two distance-time pairs you can calculate the speed-duration curve: the slope gives you CS, the intercept gives you W′. Online calculators and many training platforms handle the maths. This method costs two hard efforts, but it delivers both model parameters and works well at the end of a training block to make progress visible.

Training at critical speed: three building blocks

Once you know your CS, quality training becomes precise. Three session types cover the relevant spectrum:

• CS intervals (95–100 % of CS): 4–6 × 5 minutes just below CS with 2-minute jog recoveries. These sessions raise CS itself — the main goal for anyone targeting faster times from 10K to the marathon. Total volume per session: 20–30 minutes in the target zone.
• Above-CS intervals (102–108 %): 5–6 × 3 minutes with 2–3 minutes recovery. These train W′ and the ability to work above threshold — decisive for 5K races and finishing kicks. Once a week at most; they are the most fatiguing sessions of all.
• Tempo runs (90–95 %): 20–30 minutes continuous in the classic threshold zone. Less specific than CS intervals but economical: a strong training effect at manageable fatigue, ideal in a base phase.

The weekly split follows the proven principle: around 80 percent of the week stays easy (see 80/20 training), with CS-oriented sessions forming the sharp 20 percent. Two quality days per week are enough — more leads to a stagnating CS rather than progress.

Using critical speed for race pacing

Perhaps the most practical use of CS: it provides reliable pacing anchors for every race distance. Based on your current CS, realistic target paces emerge:

• 5K: roughly 100–105 % of CS. Fresh and on a good day, you can risk the first kilometres just above threshold — the distance is short enough to spend W′ completely.
• 10K: roughly 95–100 % of CS. The classic coincidence: exactly where well-trained runners spend their 40 to 50 minutes.
• Half marathon: roughly 88–93 % of CS. With a 4:00/km CS that means 4:18–4:33/km — for many runners a more honest prediction than any pace chart.
• Marathon: roughly 80–88 % of CS, depending on weekly volume and long runs in your build. Here your energy budget and fueling partly decide how close to 88 % you can get.

These percentages replace guesswork around goal times. If you know your CS, you know before the start which pace is realistic — and you avoid the classic mistake of running the first 10 kilometres too fast. More on strategic distribution is in our article on marathon pacing strategy.

Common mistakes around critical speed

• Treating the test like a workout. A 3-minute test at 90 percent effort returns a CS that reflects 90 percent of the truth — and a training pace too slow to do anything.
• Equating CS with 10K race pace. Only highly trained runners hold their CS for a full 10K. For most people, 10K race pace sits slightly above it — CS is the anchor, not the race pace itself.
• Training off stale numbers. After eight weeks of good training, CS may have improved by 5–10 seconds per kilometre. Training with January values in April leaves adaptation on the table.
• Ignoring heat and altitude. At 28 °C or 1,500 metres of elevation, achievable CS drops by several percent. Tests and target paces belong to comparable conditions — in summer, slower is normal, not weak.
• All speed, no base. CS rises most sustainably through high aerobic volume. Collecting only intervals lifts the threshold briefly — and erodes the base that carries it.

How Peakora steers your threshold paces

In the Peakora engine, pace targets are not static chart values: your threshold and interval paces are calibrated from your actual performance data and updated with every synced session — the same principle as the CS model, without you doing the maths. Quality blocks are periodised automatically: CS-adjacent intervals in the build, above-CS work in the specific phase, race pace in the taper. The recovery monitor makes sure hard sessions only appear on the plan when your form (TSB) can absorb them — not when the calendar happens to suggest them.

Frequently asked questions about critical speed