Super shoes are not only for fast runners: at 7.5-12 km/h the metabolic saving was present and independent of speed
Protocol
n=14 moderately trained runners (6 men, 8 women; age 25.5 ± 2.8; BMI 21.7 ± 2.3; V̇O₂peak 49.8 ± 5.1 ml·kg⁻¹·min⁻¹) ran at 7.5, 9.0, 10.5 and 12.0 km/h — i.e. down to 8:00 min/km — in three shoes: a traditional trainer (On Cloudrunner 2, TRA), an AFT racing shoe (On Cloudboom Echo 3, CBE) and a prototype combining AFT and standard features (MIX). The full protocol was repeated on three different days.
Key Findings
Combining all speeds, oxygen cost was lower in the AFT shoe than in the traditional shoe (−5.4 ml O₂·kg⁻¹·km⁻¹, 95% CI −6.9 to −3.9, p<0.001) and than in the hybrid prototype (−4.1, CI −5.6 to −2.6, p<0.001); no difference was detected between the prototype and the traditional shoe (p=0.269).
The difference between shoes was independent of speed. The benefit did not shrink at the slowest speeds tested.
No clear effect of footwear on cadence, ground contact time or leg stiffness — the saving was not explained by spatiotemporal changes.
Perceived effort was lower in the AFT shoe than in the traditional shoe only (−0.2 points, CI −0.4 to −0.1); perceived comfort did not differ (p=0.377). Shoe preference reflected both lower oxygen cost and higher comfort, with larger effect sizes for comfort.
Authors' conclusion: AFT shoes provide meaningful metabolic savings even at very low running speeds, with no distinguishable speed dependence.
Practical Application
In 14 moderately trained runners (6 men, 8 women, V̇O₂peak ~50), an AFT racing shoe lowered the oxygen cost of running by 5.4 ml O₂·kg⁻¹·km⁻¹ versus a traditional trainer across 7.5-12 km/h, and the difference did not depend on speed. The claim that super shoes "only work if you run fast" is not supported within this range; a 5:30/km runner (about 10.9 km/h) is inside the tested speeds.
Inference, not tested: the purchase decision becomes an economic one rather than a physiological one — the shoe probably works for a recreational runner; whether the money is better spent elsewhere is a separate question.
Mixed-sex sample at recreational fitness — closer to the athlete a coach usually advises than elite-only data, and worth citing when talking to amateurs.
Comfort, not oxygen cost, was the stronger driver of shoe preference — a reason to let the athlete's comfort decide between shoes that both work.
Expect a real-world discount: a laboratory oxygen saving is not a race-time promise (REF-721).
Limitations
n=14, single laboratory, one brand's shoes. Small, and one manufacturer's product line.
Moderately trained runners (V̇O₂peak ~50), not true beginners or high-BMI recreational runners, where foam compression arguments would be most testable.
Slowest speed tested was 7.5 km/h (8:00/km). Many recreational runners race slower than this; the finding cannot be extended below the tested range.
Acute measurements (repeated on three days) — no training or racing outcome.
Published 2026; no independent replication yet, and industry involvement in AFT research is pervasive enough to warrant caution.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from B+: crossover, n=14
Content correction 2026-09-27
Level B+ -> B: crossover with n=14 (rubric: crossover ≥15 for B+).
Spatiotemporal variables corrected (cadence, contact time, leg stiffness — not stride length); added CIs, the prototype-vs-traditional null result, perceived effort and the comfort-driven preference.
"No habituation period" removed (the abstract does not say so; the protocol was repeated on three days); purchase-economics point labelled as inference.