Bonking in Boston

A Data Analysis of the 2017 Boston Marathon. Part 3

barrysmyth
Running with Data

--

TLDR;

  • Hitting the wall of bonking is the ‘big-bad’ of marathon running and in this article we explore the frequency and effect of hitting the wall among runners of the 2017 Boston Marathon.
  • This data analysis is based on more than 26,000 individual race records and is one of a series of posts that looks various aspects of the Marathon, from participants and performance to pacing.

Introduction

This is the third in a series of blog posts on a data analysis of this year’s Boston Marathon. Our dataset includes 26,263 standard runner race-records; a small fraction of records have been excluded (<1%) because they contained missing timing data or other anomalies. Each remaining race-record includes information such as the gender and age of the runner, their country, state, and/or city of origin, their overall finish-time Importantly, each race-record also includes timing and pacing information at 5km (3.1 mile) intervals during the race, effectuvely dividing the course up into 9 different segments: 0–5k, 5k-10km, …, 35km-40km, and 40km-finish. In what follows we will refer to these race segments by their end-distance; thus the 5km segment refers to the segment from the start of the race to the 5km mark.

--

--

barrysmyth
Running with Data

Professor of Computer Science at University College Dublin. Focus on AI/ML and data science with applications in e-commerce, media, and health.