Crossing the finish line ends the race but should begin the analysis. Most athletes glance at their overall time, compare it to their goal, and move on. They miss the richest data source available: the split-by-split story of how their race unfolded. A 2022 pacing study of 14,000 marathon finishers found that athletes who conducted structured post-race split analysis improved their subsequent marathon time by an average of 3.2% compared to 0.8% for those who did not analyse their pacing data — a difference of 6–8 minutes at the 3:30 marathon level.
Split analysis is not about finding blame for a bad race. It is about identifying the specific kilometre or mile where execution diverged from plan, diagnosing why, and embedding that knowledge into your next race strategy. Even a perfect race contains information: what worked, and how closely can you replicate those conditions?
Collecting Your Split Data
Before you can analyse, you need accurate data. Most GPS watches record per-kilometre or per-mile splits automatically. However, GPS accuracy degrades in urban canyons (tall buildings), dense tree cover, and crowded starts where multipath errors can inflate early splits by 2–5%. Cross-reference your GPS data with the official timing mat splits from the race organiser, which are typically available within 24–48 hours at checkpoint intervals of 5K, 10K, half, and finish.
For ultra and trail events, timing mat splits at aid stations provide the most reliable segment data. Record your aid station arrival and departure times separately (most tracking systems only capture arrival). The difference reveals dwell time — and dwell time analysis is often more revealing than running pace analysis in ultras, where athletes lose 15–45 minutes to excessive aid station stops without realising it.
Positive/Negative Split Ratio
The split ratio compares the time for the first half of the race to the second half. A perfectly even-paced race has a ratio of 1.00. A negative split (second half faster than first) produces a ratio below 1.00. The optimal ratio varies by distance:
- 10K: Optimal ratio 0.98–1.02. The short distance tolerates near-even pacing with a slight positive split in the final 2K as anaerobic contribution rises
- Half-marathon: Optimal ratio 0.99–1.03. A slight positive split (3–5 seconds/km slower in the second half) is expected and physiologically normal
- Marathon: Optimal ratio 1.00–1.05. Research shows the fastest marathon performances cluster around a 1.01–1.03 ratio. A ratio above 1.08 indicates significant pacing failure — the athlete went out too fast
- Ultramarathon: Ratios of 1.10–1.25 are normal due to terrain variation, fatigue accumulation, and day/night transitions. Focus on segment-by-segment trends rather than half/half ratio
Fade Rate Analysis
Fade rate measures the percentage slowdown from your fastest split to your slowest split (excluding the first and last kilometre, which are typically noisy). For a well-paced marathon, the fade rate should be under 10% — meaning if your fastest split was 4:50/km, your slowest should be no worse than 5:19/km. A fade rate above 15% points to one of three causes: inadequate aerobic base (the pace exceeded lactate threshold sustainability), insufficient fueling (glycogen depletion caused the classic "wall"), or inappropriate starting pace (exceeded critical velocity in the first 10K).
Plot your splits on a chart with kilometre on the x-axis and pace on the y-axis. A gradually ascending line indicates progressive fatigue (normal and manageable). A sudden step change — 3 kilometres that are 20–30 seconds slower than the preceding ones — indicates an acute event: missed fuel, heat onset, or injury. Identifying where the step change occurred and correlating it with your fueling log or conditions reveals the specific intervention needed.
Terrain and Condition Adjustments
Raw pace data on a hilly course is misleading without gradient adjustment. A 5:00/km split on a 5% uphill is physiologically equivalent to approximately 4:20/km on flat ground — the athlete was actually working harder despite the slower split. Most analysis platforms (Strava, TrainingPeaks, Garmin Connect) provide grade-adjusted pace (GAP) that normalises for elevation change. Always review GAP rather than raw pace on courses with significant elevation variation.
Temperature and wind also require adjustment. Performance declines approximately 1–2% per degree Celsius above 15°C in marathon-distance events. If you raced in 25°C conditions and your fade rate was 12%, adjusting for the 10°C above neutral brings the heat-adjusted fade to approximately 2–4% — which may indicate excellent pacing rather than a problem. Context matters.
Building the Post-Race Action Plan
Within 48 hours of your race (while memories of sensations and decisions are fresh), complete a structured review: what was your A-goal pace, what was your actual pace through each 5K segment, where did you first deviate from plan, what caused the deviation, and what specific change would prevent that deviation next time? Write this down — athletes who document their review retain and apply the insights at 3x the rate of those who merely think about it.
Cross-reference your split data with your pre-race plan built in the NorthLine Running Pace Calculator. Overlay the planned splits against actual splits to visualise exactly where execution diverged. This comparison — plan versus reality, segment by segment — is the single most valuable analytical exercise in competitive endurance sport.