Data Analysis of Road Races: Interpreting Segment Pacing from Garmin and Strava

Introduction

After a marathon or half marathon, your watch holds a complete “race physical report.” Many runners only glance at their finish time before putting the watch back in the drawer, missing out on the most valuable learning opportunity. Data analysis is not exclusive to elite runners—as long as you know how to read a few key charts, anyone can dig up clues for improvement from their own race data.

This article uses Garmin Connect and Strava as the primary tools, guiding you step by step through segment pacing, heart rate trends, and cadence changes to uncover what happened during your race.

Garmin Connect: Your Race Operating Table

Chart 1: Pace vs. Heart Rate Overlay

This is the most important chart. Open Garmin Connect on the web, go to the activity, click “Charts,” and check both “Pace” and “Heart Rate” to display them together.

You should pay attention to the following three patterns:

Aerobic Decoupling is an indicator of aerobic efficiency:

• < 5%: Very stable aerobic pace, solid fitness base
• 5–7%: Acceptable, still room for improvement

• 10%: Significant drift, possibly pacing too fast or insufficient training volume

Garmin’s Firstbeat algorithm automatically calculates and displays this value.

Chart 2: Per-Kilometer Split Pacing Table

On the “Laps” page in Garmin Connect, you can see pace, heart rate, and cadence for each kilometer. Copying this data into Excel or Google Sheets allows for the following analysis:

• First 5km vs. last 5km pace difference: More than 15–20 sec/km indicates pacing control issues
• Where does the slowest kilometer occur?: Usually around aid stations, on uphill sections, or at the wall after 30km
• Pacing “death valley”: Many runners enter a fatigued but not-yet-finished phase between 20–30km, with both mental and physical performance declining

Chart 3: Cadence and Stride Length Changes

When fatigued, most runners experience shorter stride length and lower cadence. If only stride length shortens while cadence holds, your running form is holding up relatively well under fatigue; if cadence also drops noticeably, neuromuscular fatigue is quite severe.

Ideal cadence range:

• Marathon pace: 170–180 spm (steps per minute)
• If cadence drops below 165 spm late in a race, it usually indicates severe fatigue

Strava: The Tool for Segment Comparison

Segment Analysis

Strava’s segment feature allows you to compare performance on the same stretch of road across different training days. For runners who frequently run fixed routes (such as park loops or riverside bike paths), this is the most intuitive tool for tracking progress.

How to use it:

1. Open the activity on Strava → click “Segments”
2. Compare your historical best on the same segment with this performance
3. Watch for cases of same heart rate but faster pace, or same pace but lower heart rate—both indicate fitness improvement

Fitness & Freshness Chart (requires Strava Premium)

This chart displays three lines:

• Fitness: Long-term training accumulation effect (CTL)
• Fatigue: Recent training load (ATL)
• Form: Fitness minus Fatigue

The ideal Form value before a race should be between +5 and +25. If your Form value is negative on race day (e.g., -20), it means you’re competing in a fatigued state, and a pace collapse is an expected outcome—not a lack of effort on your part.

Real-World Case Interpretation

Suppose a runner finishes the Taipei Marathon in 4 hours 15 minutes, with the following data:

• First 21km: average pace 5:45/km, heart rate 145bpm
• Last 21km: average pace 6:30/km, heart rate 152bpm
• Cadence: 180spm in the first half, 166spm in the second half
• Decoupling rate: 12%

Analysis conclusions:

• Pace slowed by 45 sec/km in the second half while heart rate rose → classic glycogen depletion response
• Significant cadence drop → neuromuscular fatigue, possibly insufficient long-run training volume
• Decoupling rate of 12% → room for aerobic base improvement; recommend increasing Zone 2 long-run training proportion

Next steps for improvement: Increase the number of long runs over 30km, and refine fueling strategy (recommend taking an energy gel every 45 minutes).

Practical Tips

• Within 24 hours after the race, download your data and take screenshots; write down “how you felt” during each anomalous pacing segment while the memory is still fresh
• Create a race data folder (Google Sheets or Notion) to record key metrics from every race
• Compare against the same metrics from your previous race—don’t just compare with others, compare with your past self
• If you use Garmin, regularly update your VO2max and training status assessments; these long-term trends are more meaningful than single-race data

Conclusion

Data doesn’t lie, but it needs you to listen. Every race is an experiment—your watch records not just time, but your body’s real-time response at every moment across those 42 kilometers. Learn to read this report, and you’ll get the most out of every race, giving your next effort clearer direction.

Related Reading

• Road Running Pace Watch Applications: Heart Rate vs. Pace vs. Power Meter Choices for Marathons
• Complete Guide to Strava Running Features: Segment Timing, Training Logs, and Social Interaction
• Trail Race Pacing: Why You Shouldn’t Look at Your Garmin on the Trails
• Running Pace Calculation: Finding Your Optimal Pace with Heart Rate and Cadence