Performance level showed the most consistent relationship with pacing behavior in long-distance running, with faster runners exhibiting more stable pacing patterns and more even or slightly negative pacing generally associated with more favorable performance outcomes.
Key Findings
Results
Pacing profiles differ between half-marathon and marathon distances, with generally more even pacing in the half-marathon and more positive pacing in the marathon.
The review synthesized findings from 37 peer-reviewed studies published up to March 2026.
Pacing variability appears to increase with longer race duration.
Positive pacing refers to a slowing of pace over the course of the race, which was more pronounced in the marathon.
A structured, non-systematic search was conducted in PubMed, SPORTDiscus, Scopus, and Web of Science, with Google Scholar as a supplementary source.
Results
Performance level showed the strongest association with pacing behavior, with faster runners exhibiting more stable pacing patterns and slower runners showing greater variability and a more pronounced pace decline.
Performance level showed 'the most consistent relationship with pacing behavior in long-distance running.'
Slower runners demonstrated greater pacing variability across both half-marathon and marathon distances.
Pacing variability was found to increase with both longer race duration and lower performance level.
The authors note that in mass-participation events, observed pacing profiles may reflect fatigue-related changes and inaccurate pace selection rather than a deliberate strategy.
Results
Sex differences in pacing were more evident in marathon running, where women generally demonstrated more even pacing than men.
Sex-based pacing differences were described as 'more evident in marathon running' compared to the half-marathon.
Women's more even pacing in the marathon may reflect more accurate pace selection relative to their performance capacity.
The review treated sex as a 'secondary contextual factor' for interpreting pacing behavior.
The authors did not prescribe sex-specific pacing recommendations, framing differences as observed associations.
Results
Age-related differences in pacing were small and inconsistent across the reviewed literature.
Age was identified as producing only small and inconsistent differences in pacing profiles.
Age was proposed as a 'secondary contextual factor' alongside sex for interpreting pacing behavior.
The review did not find sufficient evidence to draw definitive conclusions about age-specific pacing patterns.
Practical reference points based on age were described as 'tentative, evidence-informed aids.'
Conclusions
More even or slightly negative pacing has generally been associated with more favorable performance outcomes in long-distance running.
The authors explicitly cautioned that this 'should be interpreted as an observed association rather than a direct prescriptive recommendation.'
Negative pacing refers to running the second half of a race faster than the first.
The association between even/negative pacing and better performance was consistent across the reviewed evidence.
In mass-participation events, pacing profiles may reflect fatigue and inaccurate pace selection rather than deliberate strategy, complicating causal interpretations.
Conclusions
The review proposed practical reference points for pacing behavior based on performance level and race distance, with sex and age as secondary contextual factors.
These reference points were described as 'tentative, evidence-informed aids for interpreting pacing behavior.'
The framework considers both within-race and between-race pacing differences.
The authors acknowledged that practical recommendations remain limited by the observational nature of available evidence.
The narrative review design precluded meta-analytic quantification of effect sizes or statistical pooling of results.
What This Means
This research reviewed 37 scientific studies to understand how runners pace themselves during half-marathons and marathons, examining whether pacing patterns differ based on how fast someone runs, their sex, and their age. The researchers searched multiple scientific databases and synthesized findings narratively rather than statistically, meaning they described and interpreted patterns across studies rather than combining numbers mathematically.
The most consistent finding was that a runner's overall performance level strongly predicts how evenly they pace themselves: faster runners tend to maintain a more steady pace throughout the race, while slower runners are more likely to start too fast and slow down significantly as the race progresses. Marathons generally showed more pronounced slowing over the race compared to half-marathons. Women tended to pace more evenly than men in marathons specifically, though the reasons for this are not fully established. Age differences in pacing were small and inconsistent across studies. Runners who maintained even or slightly faster-finishing pacing strategies tended to achieve better overall performance outcomes.
This research suggests that for everyday recreational runners, the pattern of starting too fast and slowing down significantly—especially in marathons—may often reflect poor initial pace judgment or fatigue rather than a conscious strategy. While more even pacing is associated with better race outcomes, the authors caution that this is a correlation observed across studies rather than a proven cause-and-effect relationship, meaning coaches and runners should treat pacing guidelines as helpful reference points rather than rigid prescriptions. The findings point to performance level as the most important factor when thinking about how to approach pacing in long-distance running events.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Ristanović L, Stojiljković S, Ćuk I, Weiss K, Rosemann T, Knechtle B. (2026). Pacing Profiles in Half-Marathon and Marathon Running Based on Performance Level, Sex, and Age: A Narrative Review.. Medicina (Kaunas, Lithuania). https://doi.org/10.3390/medicina62081503