Lausanne Marathon
This page focuses on the statistical analysis done on data from Lausanne Marathon 2016.
The data used was parsed from this Datasport page. For more details about how the analysis is done, please refer to this jupyter notebook hosted on Github.
Table of contents
- Number of runners by sex and by category
- Age distributions
- Towns of residence of the runners
- Times distribution by sex and by category
- Paces distribution by category
For simplicity and for a more natural visualization, we only consider the three main races (or distances) of Lausanne marathon : 10km, half marathon and full marathon.
Number of runners by sex and by category
The following stacked bar chart represents the number of runners for each category and for each sex. There are three categories : 10km, 21km and 42 km (the last two respectively correspond to the half and the full marathon).
The first observation is that there is globally more men than women. Most of the runners, males and females, participate to the 10km race. This corresponds to almost half of the total number of runners for these three categories. The less popular race is the full marathon, especially for females. This result is expectable since it is physically the hardest one. The big difference between the number of male and female runners consequently comes mostly from the marathon.
Age distributions
Overall age distribution
The following graph represents the fraction of the total number of runners with respect to their age at the time of the race.
The younger runner was 5 years old, the older one was 82 years old. Most of the runners are between 25 and 45 years old. There is an interesting minimum between 10 and 16 years old that can be socially interesting to analyse further. Finally, very few people still run such distances after 60 years old, for physical reasons.
Age distribution by sex and by category
Women
The following graph represents the fraction of the total number of female runners with respect to their age at the time of the race, for the three different chosen categories.
On the legend, you can click or unclick some of the categories to only see the ones you want to see.
After performing 2-samples Kolmogorov-Smirnov statistical tests, we conclude that only women who run the 10km race are significantly younger than the ones running the full marathon or the half marathon.
The average age for each race is:
-
10km: 36.11 years old;
-
21km: 37.48 years old;
-
42km: 39.16 years old.
Men
The following graph represents the fraction of the total number of male runners with respect to their age at the time of the race, for the three different chosen categories.
On the legend, you can click or unclick some of the categories to only see the ones you want to see.
After performing 2-samples Kolmogorov-Smirnov statistical tests, we conclude that only men who run the full marathon are significantly older than the ones running the half marathon or the 10km race.
The average age for each race is:
-
10km: 40.26 years old;
-
21km: 39.88 years old;
-
42km: 42.15 years old.
Women vs Men
After performing 2-samples Kolmogorov-Smirnov statistical tests, we conclude that women are significantly younger than men in all the three chosen competitions.
Towns of residence of the runners
The following graph represents the number of towns from which come from the runners with respect to the number of runners coming from this town and that participate to the Lausanne marathon.
It is obvious that the one city from where come more than 1000 runners is Lausanne. Moreover, it was also expectable that there are many towns (around 1000) from which comes only one runner. These towns either correspond to places very far from Lausanne, or to very small towns. Between those two extremes, the behaviour is a power law.
Times distribution by sex and by category
Marathon
The following graph represents the fraction of runners with respect to the time it took them to complete the full Lausanne marathon.
On the legend, you can click or unclick some of the categories to only see the ones you want to see.
After performing a 2-samples Kolmogorov-Smirnov test, we conclude that men are significantly faster than women at the full marathon.
Mean time to complete the full marathon for:
-
Women: 4:16.47;
-
Men: 3:54.59.
Half-marathon
The following graph represents the fraction of runners with respect to the time it took them to complete the half Lausanne marathon.
On the legend, you can click or unclick some of the categories to only see the ones you want to see.
After performing a 2-samples Kolmogorov-Smirnov test, we conclude that men are significantly faster than women at the half-marathon.
Mean time to complete the half marathon for:
-
Women: 2:03.28;
-
Men: 01:50.10.
10 km
The following graph represents the fraction of runners with respect to the time it took them to complete the 10km race.
On the legend, you can click or unclick some of the categories to only see the ones you want to see.
After performing a 2-samples Kolmogorov-Smirnov test, we conclude that men are significantly faster than women at the 10km race.
Mean time to complete the 10km race for:
-
Women: 00:59.03;
-
Men: 00:51.18.
Paces distribution by category
Men
The following graph represents the fraction of male runners with respect to their paces during each type of race (10km, half-marathon, full marathon).
On the legend, you can click or unclick some of the categories to only see the ones you want to see.
After performing a 2-samples Kolmogorov-Smirnov test, we conclude that men pace significantly increases with race distance in all categories.
Women
The following graph represents the fraction of female runners with respect to their paces during each type of race (10km, half-marathon, full marathon).
On the legend, you can click or unclick some of the categories to only see the ones you want to see.
After performing a 2-samples Kolmogorov-Smirnov test, we conclude that women pace is significantly higher only for the marathon.