The critical speed test: 12 minutes then 3 minutes running as far as you can, with 30 easy minutes between. Your two distances give your CS, the speed you can hold without blowing up, and your D′, the distance reserve above it. 70 minutes, on a track.
| Target | Zone | Time | vVO₂max 14 km/h | vVO₂max 16 km/h | vVO₂max 18 km/h |
|---|---|---|---|---|---|
| 103–112% | 6Anaerobic | 3 min | 4:10/km–3:50/km 721–784 m per rep | 3:38/km–3:21/km 824–896 m per rep | 3:14/km–2:59/km 927–1008 m per rep |
| 90–98% | 5VO₂max | 12 min | 4:46/km–4:22/km | 4:10/km–3:50/km | 3:42/km–3:24/km |
Put this session on your watch
Do the test from your watch, guided block by block. Then enter your new reference speed and every session follows.
This test doesn't make you fitter: it measures both parameters of the critical speed model, the running counterpart of the critical power cyclists use. Your CS is the line between paces you can keep going for a long time and paces that will run you into the ground: above it, fatigue builds until you stop. Your D′ is the distance you can run above that speed, a reserve in metres drained by surges, hills and the finish.
The principle is a straight line: the distance you cover flat out rises in a straight line with the length of the effort. The slope of that line is your CS, its starting point your D′, and two distances are enough to draw it. Your CS sits below your vVO₂max and above your marathon pace, which makes it a useful marker for your threshold sessions.
Experienced runners who can pace a 12-minute maximal effort well and want a fuller marker than vVO₂max alone. It suits 5K and 10K runners in particular, whose races are run close to CS, and anyone who wants to know how many surges they can afford. For a first test, or if all you want is your vVO₂max, the Cooper test or the half Cooper is simpler.
At the end of a training cycle, or every 4 to 8 weeks: a common guideline. With the long recovery the session runs to 70 minutes, so save it for a day with time to spare, after one or two easy days, never the day after a hard session. Avoid heat and strong wind, which weigh more on the 12 minutes than on the 3. Repeat it on the same track, in the same shoes, so your results compare.
10′ at 60–65% vVO₂max + 5′ at 66–72% vVO₂max
10 minutes of easy running at 60–65% of vVO₂max, then 5 minutes a little brisker, at 66–72%. Add two or three strides if you like, and spot the 100 m markings on the track: you'll use them to note your distances.
12′ at 90–98% vVO₂max
12 minutes, as far as you can, at an even pace. Guide: 90–98% of vVO₂max, or 2,880 to 3,136 m at 16 km/h. Count your laps, keep a little in hand for the first 8 minutes, then build towards the end. Note the distance to the metre: it doubles as a Cooper test.
30′ at 50–60% vVO₂max
30 minutes of very easy jogging at 50–60% of vVO₂max, or walking. This recovery is borrowed from the cycling studies: it gives you time to recover from the 12-minute effort. Drink, and get back to your start line a few minutes before the end.
3′ at 103–112% vVO₂max
3 minutes, as far as you can. Not a sprint: go off quick but even, or the last minute falls apart. Guide: 103–112% of vVO₂max, or 824 to 896 m at 16 km/h. Note the distance to the metre.
10′ at 55–65% vVO₂max
10 minutes of very easy running at 55–65% of vVO₂max, starting with a walk. Then the maths: CS (m/s) = (d12 − d3) ÷ 540, multiplied by 3.6 for km/h, and D′ = d3 − CS × 180, in metres. 3,300 m and 1,000 m give a CS of 4.26 m/s, or 15.3 km/h, and a D′ of about 233 m.
The 12 minutes feel like a 3K race: hard but manageable for eight minutes, then very hard. The 3 minutes arrive at exactly the moment you were starting to feel human again, and they're maximal from start to finish. Guide: 9 to 10 out of 10 on each effort.
Over 3 minutes, a few metres of error carry a lot of weight. With 3,300 m and 1,000 m your D′ is about 233 m; if your watch adds 20 m to the second effort, it becomes about 260 m. On a track, count laps and read the markings: far more reliable.
With only two points, a poor effort doesn't show up: it just moves the line. A cautious 3-minute effort inflates your CS and shrinks your D′; a cautious 12-minute effort does the opposite. Both need to be genuinely maximal.
Thirty minutes of jogging is a long time. But a shorter recovery leaves fatigue in your 3 minutes, and the line comes out wrong. If you're short of time, run the two efforts on separate days.
Your CS is lower than your vVO₂max: enter it in its place and every session set as a % of vVO₂max becomes too slow. Keep vVO₂max for your paces, and CS as a threshold marker.
The distance-duration model is the running counterpart of the critical power model put forward by Monod and Scherrer (Ergonomics 1965). Galbraith and colleagues (2011) tested it in the field with 10 trained distance runners over 1,200, 2,400 and 3,600 m, roughly 3, 7 and 12 minutes: CS was reproducible from one trial to the next, with a coefficient of variation of 1.3 to 2.0%. In the studies consulted, D′ is clearly less reliable than CS. The two-effort version and the 30-minute recovery, borrowed from the cycling studies, have not been validated as such in running. The % vVO₂max ranges, the warm-up and the cool-down are Wattson choices. Bottom line: good evidence for CS, weak for D′.
Subtract your 3-minute distance from your 12-minute distance, then divide by 540, the difference in seconds: that's your CS in metres per second. Multiply by 3.6 for km/h. 3,300 m and 1,000 m give 4.26 m/s, or 15.3 km/h. Your D′ is then d3 − CS × 180, here about 233 m.
It's your distance reserve above CS. As soon as you run faster than CS you start drawing on it; once it's empty, you have to slow down. A big D′ helps with surges, hills and the finish. Two runners with the same CS can have very different D′ values, and race in very different ways.
vVO₂max is the speed at which you reach your maximal oxygen uptake, which you can hold for a few minutes. CS is lower: the highest speed you can hold in a steady state. Sessions set as a % of vVO₂max stay anchored to vVO₂max; CS is a marker for your threshold sessions and race paces.
Yes, if you did it recently, fresh, on the same track: your 12-minute distance is the first point on the line. All that's left is to run the 3 minutes on another day, in the same conditions. That's the separate-days version, which is more reliable for D′.
On a track if you can: the distances are exact, and a few metres of error on the 3 minutes noticeably changes your D′. On the road, pick a flat, measured route with no stops, and run both efforts in the same place. Don't rely on GPS alone.
An intuitive workout editor to build cycling and running sessions in no time, then sync them to your watch or bike computer: Wahoo, Garmin, Suunto or Coros.
Try this session