Home / Lab Notes / Running
London Marathon Full Field Analysis

65,294
RUNNERS.
ONE FINISH
LINE.

A deep-dive into pace, age, nationality, positive splits, and what it actually takes to reach the top 1% of one of the world's most iconic marathons.

59,814
Finishers
4:28:44
Median finish
93.8%
Positive split rate
168
Nations
01 Overview

Key findings
at a glance

Six headline numbers from 65,000 runners. Each one hides a deeper story worth telling.

āš”
Sub-3h runners
5.5%
Only 3,298 of 59,814 finishers broke the 3-hour barrier. This elite tier paces on a completely different strategy from everyone else in the field.
šŸƒ
Club runner advantage
49 MIN
Affiliated club runners finish with a median of 3:46 while solo runners clock 4:35. Training structure and group accountability compound over time in measurable ways.
šŸ“‰
Positive split rate
93.8%
Nine in ten runners ran their second half slower than their first. The marathon wall is not a myth. It is structural, and even with a 5-minute grace threshold, 63% of the field still faded.
šŸŽ‚
Surprising age winner
40ā€“44
The 40ā€“44 group runs a median of 4:20 while the 18 to 39 group clocks 4:24. Experience and disciplined pacing outperform raw youth by a measurable margin.
šŸŒ
Nations represented
168
London Marathon is genuinely a world event. Israel leads the speed table with an average of 3:51 among nations with at least 50 finishers in the field.
šŸ†
80+ fastest finisher
4:07:16
John F. (GBR) ran 4:07 at age 80 with a positive split of just 2 minutes 34 seconds. He finished nearly an hour ahead of the second-place runner in his age group.
02 Finish Time Distribution

Where does the
field cluster?

59,814 finish times broken into 15-minute windows. The shape reveals how runners self-select across the course and where round-number targets create artificial bunching in the results.

Finish Time Distribution
15-minute bins across all finishers, colour-coded by finish tier
Sub-4h (34%)
4h to 6h (56%)
6h and over (10%)
The 4-hour dip is real. There is a measurable drop in finishers right at 4:00 followed by a surge at 4:15. Runners push hard to beat the benchmark, blow up trying to hold pace, and end up finishing just after it anyway.
Right-skewed distribution. The mean of 4:32:55 sits noticeably to the right of the median at 4:24:55. A long tail of slower runners pulls the average up, so the median is a far better representation of the typical runner's experience.
Finish Tier Breakdown
Percentage of the field that reached each benchmark time
03 Age Group Analysis

Does age slow
you down?

Not until 50. The data reveals a counterintuitive pattern in the early age brackets, followed by a steady and consistent decline from 55 onward.

Median Finish Time by Age Group
IQR range shown as Q1 and Q3 reference lines, alongside median fade time per group
The 40ā€“44 paradox: Their Q1 of 3:39:51 is also faster than the 18 to 39 Q1 of 3:43:24. This is not just a median effect. The top performers in this age group are genuinely quicker across the board, which points to two decades of accumulated race experience and disciplined pacing wisdom compounding into a real performance edge.
Median Fade by Age Group
Minutes lost in the second half per group, with positive split rate overlaid
Minutes lost in 2nd half
Percentage who positive split
Age GroupRunnersFastestQ1MedianQ3Median Fade% Pos Split
04 Pacing Strategy

Sub-3h vs
the field

Elite runners do not just run faster. They operate on a completely different pacing strategy. Five metrics reveal exactly where the gap comes from.

Sub-3 Hour
Runners
Median fade
+2:30
3.1% slower in the second half
Negative splits
16.5%
Even splits (Ā±1 min)
22.7%
Split variability (std dev)
Ā±4.8%
Even pacers (49 to 51% ratio)
58.9%
Average Field
(4ā€“5 Hours)
Median fade
+16:07
12.7% slower in the second half, 6.4 times worse than sub-3h
Negative splits
4.0%
Even splits (Ā±1 min)
2.6%
Split variability (std dev)
Ā±11.2%
Even pacers (49 to 51% ratio)
13.9%
Pacing Split Distribution, Sub-3h vs Average Field
Normalised to percentage of each group, showing how much slower each runner's second half was
Pacing Discipline Buckets
First half as a proportion of total finish time
Split Variability Curves
Normal distribution curves showing how tightly each group clusters around its mean fade
05 The Wall

Biggest fades
in the field

These are the runners who went out hardest and paid for it most. Ranked by absolute time lost between their projected finish based on half split and their actual finish time.

Fade Distribution Across All Finishers
How many minutes runners lost in the second half, grouped into seven severity bands
#NameAgeHalfFinishTime Lost% Slower
A clear pattern emerges: Nearly all of the top faders are 18 to 39 GBR runners who set out on sub-1:55 half splits and completely collapsed in the second half. Going out too fast is the single strongest predictor of a catastrophic blow-up.
06 Nations

168 countries.
One course.

Fastest nations by average finish time, with a minimum of 50 runners required to qualify. GBR is excluded from the chart since their 45,000-runner volume would collapse the scale for everyone else.

Fastest Nations by Average Finish Time
Minimum 50 finishers per nation, GBR excluded for scale readability
Israel leads with an average of 3:51 across 73 runners. This likely reflects a self-selecting group of highly dedicated competitive runners who travel internationally specifically to race.
Brazil sends 622 runners, the largest non-GBR contingent in the field, averaging 4:04. A thriving running culture and deep marathon infrastructure at home clearly produces competitive travellers.
07 Percentile Guide

Where do you
stand? (18 to 39)

Based on 31,507 runners in the 18 to 39 age group. The mean finish is 4:32:55 with a standard deviation of 68 minutes. The distribution is right-skewed, so the normal curve serves as an approximation rather than a perfect fit.

18 to 39 Finish Time Distribution with Percentile Bands
Colour bands represent the top 1%, 10%, 25%, 50%, 75%, and 90% thresholds