Every autumn, when runners move from the road to the treadmill, the same piece of advice resurfaces: “set the incline to 1% so it feels like the road”. The advice is old, and it has an author and a date: Jones and Doust, Journal of Sports Sciences, 1996. It also has a problem — in thirty years almost nobody has re-checked it in the form in which it is actually used. In 2026 three groups did so at once.

Where the rule came from

In 1996 nine trained men ran 6 minutes at each of six speeds — from 10.5 to 18 km/h — on a treadmill at 0, 1, 2 and 3% grade and on a flat road. The logic of the incline is simple: indoors there is no air to push against, while outdoors overcoming it costs energy. The incline is meant to add roughly the same amount.

The result was subtler than the way it was later retold. At the two slowest speeds (10.5 and 12 km/h) the road was no different from either 0% or 1%. At 13.5 km/h only 1% matched. At 15–16.5 km/h both 1% and 2% did. At 18 km/h the road fell between 1% and 2%. The authors' conclusion — “1% is appropriate for speeds of 10.5–18 km/h” — is correct, but it is the middle of the range, not an exact match at every speed.

What was tested in 2026

Study one: running economy. Shahidi and colleagues (PLoS One, August 2026) took 12 male runners aged 21 (VO₂max determined by a ramp test) and put them through the same incremental protocol — 8–15 km/h in 3-minute stages — first on a treadmill at a 1% grade, then on an outdoor 400-metre track. Breath-by-breath gas analysis, averaged over the final 60–90 seconds of each stage.

Running economy (VO₂ at a given speed): 51.3 ± 2.2 versus 49.6 ± 3.1 ml/kg/min outdoors and on the treadmill respectively. The gap widened at intensities between 70 and 100% of the second ventilatory threshold.
Oxygen cost per kilometre: 204.2 ± 11.3 versus 197.1 ± 12.3 ml/kg/km — roughly 3.6% more expensive outdoors, and the same at every intensity (no condition × intensity interaction).
Energy cost per metre: 3.87–4.60 versus 3.65–4.10 J/kg/m, with the gap growing with intensity (p = 0.045).
Carbohydrate oxidation approaching VT2 was also higher outdoors.

In other words, even with the 1% grade, the track remained 3–4% more expensive than the treadmill.

Study two: lactate threshold in elite runners. Suzuki and colleagues (Scandinavian Journal of Medicine & Science in Sports, 2026) compared nine Japanese Tier 4 runners — personal bests of 13:35 for 5,000 m and 27:44 for 10,000 m — on a 300-metre indoor track and on a treadmill with no incline. Six stages from 17.2 to 22.5 km/h.

Lactate on the track was higher at every stage, and the difference grew: from 1.0 mmol/L at the first stage to 3.7 mmol/L at the sixth.
Speed at 4 mmol/L: 329.6 versus 345.1 m/min (19.8 versus 20.7 km/h). At 6 mmol/L: 344.2 versus 359.4. Converted to pace — 7–8 seconds per kilometre.
Ground contact time at the higher speeds was longer on the track.
Speed at 6 mmol/L from the track predicted the season's 10,000 m result with r = −0.973; from the treadmill — only r = −0.695, and without statistical significance.

The authors say it outright: thresholds obtained on the treadmill and on the track are not interchangeable, and the correction depends on speed.

Study three: the review. Bottura and Fletcher (Applied Physiology, Nutrition, and Metabolism, 2026) gathered thirty years of data and concluded that the equivalence of treadmill and overground running depends on speed, environmental conditions, methodology and individual characteristics, so a fixed 1% cannot be universal.

This agrees with the 2019 meta-analysis by Miller and colleagues (34 studies): at submaximal speeds VO₂ on the treadmill and overground is practically the same — at 0% as at 1% — whereas lactate on the treadmill is lower (by 1.26 mmol/L at 0% and by 0.52 at 1%), and endurance-test performance on the treadmill is worse (SMD −0.50).

Why this happens

Air resistance grows with the square of speed, and the power needed to overcome it with the cube. At 10–12 km/h that is a few per cent of the cost of running; at 20 km/h it is already around 8%. One per cent of incline adds a fixed “step” and cannot keep up with a cubic relationship — hence, already in Jones and Doust, “at slow speeds 0% is enough, at fast speeds 1% is too little”.

But air is not the only difference. A treadmill belt is softer and returns energy differently from asphalt or tartan; on a treadmill there is no wind to cool you; there is no visual flow and no bends; foot placement changes — Suzuki saw longer ground contact on the track. That is exactly why lactate diverges more than VO₂ does: the body spends roughly the same amount of oxygen but recruits the muscles differently.

How to apply this

Easy running up to 12 km/h — the incline does not matter. 0 or 1%, whichever is more comfortable; the difference is within the noise.
Tempo and threshold sessions from 16 km/h — 1% is not enough. Either add 0.5–1 km/h to the target speed, or set 1.5–2% of incline, or simply accept that on the treadmill you are running a little easier than the display says.
Do not transfer heart-rate and lactate zones obtained on a treadmill to the track one-to-one. In elite runners the threshold speed on the track is 7–8 s/km slower than on the treadmill. For amateurs at 12–15 km/h the gap is smaller, but the direction is the same: calculate your training zones in the environment where you will be training.
If you test twice a year, test the same way. Comparing a “summer” threshold from the track with a “winter” one from the treadmill is meaningless: the methodology will be sitting inside the difference.
The reverse correction works too: to reproduce the track's threshold intensity on a treadmill, treadmill speed needs to go up by the same 7–8 s/km.

Honest limitations

The samples are tiny: 12 and 9 people, men only, young and trained. Shahidi had 21-year-old male amateurs, Suzuki had national-level elites. This cannot be extrapolated to women, to older runners or to speeds below 8 km/h.

The conditions were not fully equalised. Outdoors also means temperature, surface, footwear, and a bend every hundred metres on the track. Part of the “extra” 3.6% may be the bends rather than the air.

Suzuki compared a treadmill with no incline. His 7–8 s/km is the price of the absence of a correction at elite speeds, not a test of the 1% rule. By the logic of Jones and Doust, at 21 km/h even 1% would be too little, but the study did not test that directly.

Blinding is impossible. Every participant knew where they were running.

The bottom line

The 1% rule (Jones and Doust, 1996) was derived from nine runners and, by the authors' own data, matches the road exactly only at around 13.5 km/h; at 10–12 km/h 0% is enough, at 18 km/h even 1% is too little.
In 2026, in 12 runners, the track turned out 3.6% more expensive in oxygen per kilometre even at a 1% grade.
In nine elite runners lactate on the track was higher at every stage (a gap of 1.0 to 3.7 mmol/L), and threshold speed on the track was 7–8 s/km slower.
A meta-analysis of 34 studies: VO₂ is the same on the treadmill and overground, lactate is lower on the treadmill, endurance-test performance is worse.
In practice: up to 12 km/h the incline does not matter; at tempo speeds 1% is too little; calculate your zones where you train.

Sources: Shahidi S.H., Can R., Paça F.M., Zengin M.D. “Overground running incurs a higher energetic cost than treadmill running at a 1% grade: A comparison of running economy, oxygen cost of transport, and energy cost in endurance athletes”. PLoS One, 2026;21(8):e0355988. DOI: 10.1371/journal.pone.0355988 · Suzuki Y., Takei N., Tanji F., Yamanaka R., Saeki T. “Track and Treadmill Lactate Assessments Are Not Interchangeable for Training Prescription in Elite Distance Runners”. Scandinavian Journal of Medicine & Science in Sports, 2026;36(4):e70286. DOI: 10.1111/sms.70286 · Bottura R., Fletcher J. “Beyond the 1% treadmill rule: 30 years of evidence on treadmill and overground running economy”. Applied Physiology, Nutrition, and Metabolism, 2026;51:1–7. DOI: 10.1139/apnm-2026-0109 · Miller J.R., Van Hooren B., Bishop C., Buckley J.D., Willy R.W., Fuller J.T. “A Systematic Review and Meta-Analysis of Crossover Studies Comparing Physiological, Perceptual and Performance Measures Between Treadmill and Overground Running”. Sports Medicine, 2019;49(5):763–782. DOI: 10.1007/s40279-019-01087-9 · Jones A.M., Doust J.H. “A 1% treadmill grade most accurately reflects the energetic cost of outdoor running”. Journal of Sports Sciences, 1996;14(4):321–327. DOI: 10.1080/02640419608727717