When does a measurement become a useful decision?

Fatigue can change one measure while another stays stable

In 12 trained trail runners, device-derived contact measures changed during a 216-minute treadmill protocol while running economy and several power measures remained largely stable. During the fixed-speed phases, heart rate and perceived effort increased. Some uphill power differences were no longer significant after adjustment for speed.

Limitations: Small sample:6 women and 6 men pooled, with no sex-specific conclusion. Laboratory running with standardized fueling differs from a technical ultra. Proprietary sensor measures lacked concurrent criterion validation in the fatigued uphill trials. Speed adjustment is model dependent, not an isolated causal effect of fatigue. No individual cutoff, ideal form or training dose was validated.

Evidence level: Moderate evidence

Supports checking what a fatigue measure actually represents and retaining terrain, effort and fueling context; it does not validate a watch-based readiness decision.

• Power-Based Running Durability in Trained Runners: Preliminary Evidence of Speed-Adjusted Mechanical Drift as a Marker of Fatigue. (opens in a new tab) · 2026-10-19 · Research from this week

A model that fits these intervals still needs an independent test

In 27 trained runners, establishedW′-balance models overpredicted the intervals completed in two specific track sessions. A refitted linear model fitted those sessions more closely. Its reported errors came from the same 54 sessions used to fit it, not an independent prediction test.

Limitations: Mostly men, one fixed intensity/rest design and voluntary stopping. The model's empirical coefficient is not an established physiological trait. Small in-sample errors do not establish accuracy for a new runner, different session or ultra race. No automated interval prescription or individualized coaching tool follows.

Evidence level: Moderate evidence

Reinforces the existing distinction between model fit and validation; retain the established plan while checking independent validation before using a model to change it.

• Original W'-balance models do not transfer to high-intensity interval training in trained adult runners: Proposal of a single-coefficient linear alternative. (opens in a new tab) · 2026-10-20 · Research from this week

Palm cooling depends on the test and the comparison

A review of 24 reports found mixed results for palm cooling across exercise tasks. Its authors rated endurance-performance evidence low and resistance-volume evidence very low. Positive findings coexist with null results; an older 12-person running trial did not detect a difference in total exercise time including warm-up; time to exhaustion was shorter with cooling.

Limitations: Nonregistered narrative synthesis, mostly small studies and only one validated sham comparison. Devices, timing and environments differed; two reports reused the same 11 participants. The older running trial tested one device, not every cooling method. Ultra performance and heat-illness prevention remain unestablished. No product or cooling schedule is recommended.

Evidence level: Emerging evidence

Supports preserving the current context-dependent heat guidance and checking the actual performance outcome before adopting a cooling device. The older negative study is a foundation, not a new October trial: https://pubmed.ncbi.nlm.nih.gov/23444094/.

• Palmar Cooling as a Thermoregulatory Intervention: Mechanistic Basis and Implications for Exercise and Clinical Applications. (opens in a new tab) · 2026-10-18 · Research from this week
• Palm cooling does not improve running performance. (opens in a new tab) · 2013-02-26 · Foundational evidence

Actionable items

1. Before using a new fatigue score or interval model to change next week’s training, check whether it was tested on different runners or sessions from those used to fit it. Keep your established plan when that validation is missing.
2. Before buying or adding a palm-cooling device for a hot race, check whether the evidence measured running performance in comparable conditions. Use your existing heat plan; cooling a hand or changing a temperature measure does not establish heat-illness protection.