Quitting and the “World’s Toughest Bicycle Race” are an unfortunate but inevitable pair. 3000 miles between start and finish ensures that some of the athletes who roll up to the start line won’t be able to finish. Each athlete’s effort, successful or not, is recorded on the Race Across America’s website. What can we learn from these records?
Race Across America (RAAM) is a non-stop race starting in Oceanside, CA and ending in Annapolis, MA. Solo racers have 12 days to finish and team racers need to finish in 9. This course includes 12 states, over 170,000 feet of climbing, and is broken up into ~54 time stations.
Race Across the West (RAW) is a similar non-stop race that starts in Oceanside, CA and ends in Durango, CO along the same route. It is 930 miles long, has shorter time cutoffs, and is the first 15 times stations of the RAAM route.
This race is truly unique in its transparency. The results page on raamrace.org provides comprehensive metrics like total mileage completed, arrival date, arrival time, and average speed. This data is all thanks to a GPS tracker the racers wear during the race. Most importantly, they provide the Did Not Finish (DNF) data for each RAAM and RAW team and solo racer.
To get the full scope of DNF’s I needed to pull data from as far back as possible. That pull provided over 1300 teams and solo athletes to sort through.
I used Google Sheets, the “=importhtml” function, some conditional formatting, and filtered the DNF results for each year starting from 2009. The end result was a chart including the total RAAM & RAW DNFs for each time station broken down by year.
My hope is that through parsing this data, we can start to find patterns and correlations that will help guide more research, and comprehend the problem areas faced by athletes and race organizers.
Staring at a wall of numbers is not ideal, but creating a way to visualize the data allows for easier comprehension. While Google Sheets has some charts to visualize data, they lacked the complexity needed to show this dataset. Therefore I learned how to use Tableau, and here’s what I came up with for a first draft:
Yearly racer DNF count for each time station (TS).
The bar chart above gives a good deep dive on DNF frequency based on time station, but it does not give an easy-to-comprehend view. To fix that issue, I transformed the time stations into latitude and longitude coordinates, then mapped out the time station DNF’s on the actual RAAM route. This new map below now makes it easier to visualize over a decade worth of DNFs along the RAAM route.
The total DNF count from 2009-2022 along the RAAM route.
While this quantitative data doesn’t give us the “Why” behind racer and team DNFs, it does open the door to explore.
What I Learned:
Questions to explore: