Marathon Pacing Guide
How to pace a marathon — four strategies backed by research, how hills and weather change your splits, what wind actually does to your effort, and why the wall is a fueling problem disguised as a pacing problem.
Quick summary
- →Even effort is the safest strategy and produces the best outcomes for most runners.
- →True negative splits are rare (1–8% of finishers) — don't plan for one unless you've done it before.
- →Hills cost more going up than they save going down. A +2% grade adds ~24s/mi; -2% saves only ~14s.
- →Heat stress (WBGT) is the biggest external pace killer — it accounts for dew point, wind, and sun, not just temperature.
- →Wind matters most on exposed, out-and-back courses. Loopy urban courses largely cancel out.
- →The wall at mile 20 is a fueling problem, not a pacing problem. See our fueling guide.
4 Pacing Strategies
Every marathon plan needs a pacing strategy. Not "I'll see how I feel" — an actual decision about how you want to distribute effort across 26.2 miles. The research is clear on what works:
Constant effort adjusted for terrain. Uphill slows the clock, downhill speeds it up, but your perceived exertion stays flat. This is the safe default and what most coaches recommend.
Evidence: A study of 1.7 million marathon finishers (Smyth 2018) and a 39-study systematic review (2024) confirm even pacing produces the most consistent results. Elite non-fallers achieve just 2.9% speed variation across the race (Hanley 2016).
Best for: First marathons, unfamiliar courses, conservative goal times.
First half held 0.5–1% slower than average, then progressively faster through the back half. The acceleration is concentrated in the final 45% — a J-curve, not a straight line.
Evidence: Elite world record holders pace this way (Díaz 2018). But only 1–8% of recreational marathoners achieve a true negative split (Smyth 2018). The early conservatism preserves glycogen and reduces cardiac drift. Especially effective in warm conditions.
Best for: Experienced racers, hot weather, courses with late downhills.
First half about 2–3% faster than average, trending to 2–3% slower by the finish. Rather than pretending you'll negative split and panicking when you slow, this strategy plans for a controlled fade.
Evidence: 87% of marathon finishers positive split (Smyth 2018). Men fade about 14%, women about 9% in the second half (Deaner 2015). A planned +3% fade is very different from an unplanned +15% blow-up.
Best for: Net-downhill courses, runners who historically fade, windy races with early tailwind.
Slightly conservative (~0.5% slower) through mile 20, then progressive acceleration over the final 10K. Bets that your disciplined early pacing preserved glycogen that other runners burned.
Evidence: 43–56% of recreational marathoners hit the wall, with 73% collapsing after mile 19 (PLOS ONE 2021). This strategy turns that statistical cliff into your advantage — you're accelerating while the field is fading.
Best for: Experienced racers who trust their fueling plan and can access another gear.
How Hills Affect Your Pace
The key insight: uphill costs more than downhill saves. This is the Minetti polynomial (Minetti et al. 2002) — the metabolic cost of running on a grade is asymmetric. A +2% uphill adds roughly 24 seconds per mile, but a -2% downhill only saves about 14 seconds. Over a hilly course, this asymmetry means your average pace is always slower than flat equivalent, even if the course is net-zero elevation.
Grade-Adjusted Pace (GAP) — what your watch should show
GAP converts your actual pace on hills to what that effort would equal on flat ground. If you're running 8:30/mi up a 3% hill, your GAP might be 7:50/mi — meaning you're actually working harder than your watch says. Run by GAP on hilly courses to maintain even effort. Most GPS watches can display GAP in real time.
Weather & Heat Penalties
Heat is the single biggest external factor that slows marathon performance. Raw temperature alone is misleading — a dry 80°F is very different from a humid 80°F. The scientific model uses WBGT (Wet Bulb Globe Temperature), which combines air temperature, dew point, wind speed, and solar radiation into a single heat stress index. WBGT is what race medical teams use to set flag conditions, and it's what your body actually responds to.
Ideal marathon conditions — no adjustment needed.
Mild heat stress — most runners won't notice.
Moderate — expect 15–30s/mi slower. Effort-based pacing essential.
High — redefine your goal. Run by effort, not by watch.
Extreme — black flag territory. Walk aid stations. Prioritize finishing.
The practical takeaway
When WBGT exceeds 65°F, switch from pace-based to effort-based running. Your watch pace will be slower, and that's correct — it's not fitness loss, it's thermodynamics. The same effort that produces a 3:30 marathon at WBGT 50°F might produce a 3:45 at WBGT 70°F. Accepting this before the race prevents the desperation spiral of chasing a pace your body can't sustain in heat. A 60°F race day with high dew point and full sun can have a higher WBGT than a 70°F day with low dew point and cloud cover — that's why temperature alone is unreliable.
Wind on Course
Wind is often overlooked but can add 2–5% to your effort on exposed courses. The key physics: headwind costs more than tailwind saves — the same asymmetry as hills. Air resistance increases with the square of relative wind speed, so a 15 mph headwind hurts far more than a 15 mph tailwind helps.
Worst case
Out-and-back, open road
Strong wind on an exposed out-and-back means half the race is a headwind slog. The headwind half costs more than the tailwind half saves. Net effect: always slower.
Best case
Loopy urban course
Courses that zigzag through city streets (Chicago, London) largely neutralize wind — you get headwind, crosswind, and tailwind on alternating blocks. Trees and buildings provide shelter.
Practical advice: on headwind miles, tuck behind other runners (drafting reduces air resistance by 30–40%). On tailwind miles, don't surge — the free speed feels great but you're still burning glycogen. Stay disciplined.
The Wall
The wall is not a pacing problem — it's a fueling problem. Glycogen depletion occurs at approximately 120 minutes of sustained marathon-pace effort, regardless of your speed. A 3:00 marathoner hits it around mile 17–18. A 4:30 marathoner hits it around mile 11–12. The mile is different; the time is the same.
43–56%
of recreational marathoners bonk
~120 min
glycogen depletion threshold
73%
hit the wall after mile 19
The fix is not pacing slower — it's fueling properly. Start taking carbs at 40–45 minutes, maintain 60–90 g/hr with dual-transport gels, and train your gut in the weeks before race day. See our marathon fueling guide for the full protocol.
Half Marathon Pacing
The half marathon is raced at a higher percentage of VO2max than the full — which means the lactate cost of going out too fast is steeper. The strategy principles are the same (even effort is safest) but the execution is different:
No glycogen wall
At 75–120 minutes, most runners have enough stored glycogen. The limiter is lactate threshold and pain tolerance, not fuel.
Tighter splits matter more
Every second counts in a shorter race. A 5s/mi positive split costs you 65 seconds over 13.1 miles. In a marathon that's background noise; in a half it's a PR or not.
Negative split is more achievable
With no wall risk and shorter duration, the half is the best distance to practice negative splitting before trying it in a marathon.
Get splits built for your race
Your course. Your weather. Your pace.
The guide above is universal — it works for any marathon. A racecast.io premium dossier takes it further: per-mile splits computed from your actual course GeoJSON, adjusted for race-day hourly weather, wind direction vs course bearing at every mile, and your chosen pacing strategy. Plus a Race Conditions card showing exactly how heat, hills, and wind modify your goal pace — with WBGT heat stress per hour.
Find your race →Research Sources
Smyth (2018) — An analysis of pacing profiles of 1.7 million marathon finishers. British Journal of Sports Medicine 52(8):549–556.
Even pacing → best outcomes. 87% positive split. Only 1-8% negative split.
Hanley (2016) — Pacing profiles and pack running at the IAAF World Half Marathon Championships. Journal of Sports Sciences 34(17):1637–1645.
Elite non-fallers: 2.9% speed variation.
Díaz et al. (2018) — Pacing and Performance in the 6 World Marathon Majors. Frontiers in Sports and Active Living.
Elite world record negative splits are modest (0.5-0.8% between halves).
Deaner et al. (2015) — Men are more likely than women to slow in the marathon. Medicine & Science in Sports & Exercise 47(3):607–616.
Men fade 14%, women 9% in the second half.
Wall et al. (2015) — Physiological responses to marathons and ultramarathons. Comprehensive Physiology 5(4):1611–1639.
Cardiac drift, glycogen depletion, and thermoregulation across marathon distance.
Minetti et al. (2002) — Energy cost of walking and running at extreme uphill and downhill slopes. Journal of Applied Physiology 93(3):1039–1046.
Grade-cost polynomial: uphill costs more than downhill saves.
Racinais et al. (2019) — Heat acclimation and athletic performance. Sports Medicine 49(Suppl 1):S97–S101.
WBGT-based heat stress model for endurance performance.
PLOS ONE (2021) — Prevalence and predictors of hitting the wall in marathon running. PLOS ONE 16(5).
43-56% bonk; 73% after mile 19.