Pacing strategies: all-out only for ≤30 s events; more even pacing improves times in events >2 min; optimal pacing for 1.5-2 min and >4 h events is unknown
Key Findings
Pacing strategies observed by applying power/velocity and force/time modelling to athletic performances: negative, all-out, positive, even, parabolic-shaped and variable
Proposed regulation: self-selected intensity is set in the brain from peripheral sensory feedback and the anticipated workload remaining; the rate and capacity limits of anaerobic and aerobic energy supply are the main determinants of the optimal strategy
Pacing Types (as the abstract states them)
| Strategy | Description | Evidence in the review |
|---|---|---|
| All-out | Maximum from the first second | May benefit extremely short events (≤30 s) |
| Even | Near-constant pace | Prolonged events (>2 min): times may improve when pace is distributed more evenly |
| Positive | Fast start, progressive slowing after peak speed | What well-trained athletes tend to do in middle-distance (1.5-2 min) and ultra-endurance (>4 h) events — optimal strategy for these durations is unknown |
| Negative | Speeding up towards the finish | Described as an observed profile; no duration-specific recommendation in the abstract |
| Parabolic, variable | U-shaped, or varied with terrain/conditions | Described as observed profiles |
Practical Application
Abbiss & Laursen's 2008 review of pacing research finds that an explosive all-out start pays only in events of about 30 s or less, while in events longer than 2 minutes performance times may improve when pace is distributed more evenly; for middle-distance (1.5-2 min) and ultra-endurance (>4 h) events the optimal strategy is not established, and well-trained athletes typically pace positively (slow progressively after peak speed) — so for any race longer than a couple of minutes the default is to avoid the fast start, and beyond 4 h the "even" recommendation is not proven.
Pace is self-regulated by anticipated remaining work, so a known course and a realistic target keep the early effort in check.
Not from this paper: "even pacing especially >2 h", "negative split in the final hour", and "power strictly below critical power" are house advice (see SYNTH-10 for the multi-hour pacing model); durability data (REF-179, REF-181, REF-423) post-date this review and shape what is sustainable late in a long event.
Limitations
Foundational narrative review from 2008, not a meta-analysis: it organises the pacing literature rather than pooling it.
Most underlying evidence is from laboratory time trials of far shorter duration than the multi-hour events the even-pacing idea is often applied to; the review itself states that optimal pacing for >4 h events is unknown.
Terrain, wind, drafting and group dynamics — the dominant pacing constraints in real road racing — are largely absent from the controlled studies underneath.
The entry's earlier sport tag (sprint) has been replaced by "general"; the practical claim concerns events >2 min.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: narrative review without systematic method, from the leading pacing group
Content correction 2026-09-27
Headline "even pacing optimal for >2 h" -> all-out for ≤30 s; more even pacing improves times for events >2 minutes; optimal pacing for 1.5-2 min and >4 h events is unknown, where well-trained athletes tend to pace positively.
Table "positive 30 s-2 min", "even especially >2 h", "negative acceptable last 20 min" -> not in the abstract; replaced with the abstract's statements.
"Negative split in final hour" and "power strictly <CP" moved under "Not from this paper" as house advice.
Level A -> B+ (narrative review).