Big Shoulders stats: Stormy, Husky, Brawling

Less than 3 weeks ‘til Big Shoulders! This race has a special place in my heart: It was at Big Shoulders ‘09 where I caught the open-water bug. Without which, this summer wouldn’t have been nearly as awesome. Little did I realize that Big Shoulders would soon be my hometown race.


In its 20th year, it reached the maximum registration of 800 swimmers for the first time. That’s an eightfold increase since 1998, the first year for which results are available on the web.

To facilitate analysis across years, I aggregated these 12 years of results (1998-2009) into a single CSV file. This is what you might call a picture of success:



Historically, more men than women have taken the plunge, but the gap has narrowed in recent years. In 2009, women were 43% of the total participants.


Masters swimming is traditionally dominated by people in their 40’s and 50’s - is the same true here?

It seems the modal age is actually a bit younger in Big Shoulders - lots of people in their 30’s. But the “50’s” have been mounting a furious comeback (see the blue line) - perhaps a baby boomer effect.

Local vs Out-of-State

Here’s the proportion of Big Shoulders participants hailing from Illinois, Indiana, and “other” - i.e., anyplace besides IL and IN.

Clearly, Illinois locals still predominate, but recent years have seen a greater influx of out-of-state swimmers. In 2009, almost 30% came from outside of Illinois and Indiana - an all-time high.

Finish Times

People say times don’t matter in open water - or at least that you don’t always know what they mean. And perhaps that’s part of its attraction. While in the pool “the clock never lies,” in open water it’s not much more than a ranking device.

Even so, I’ve been surprised by how closely most of my open-water pace times have approximated my pool speed at various distances - from 1:15 at 1 mile (Huntersville), to 1:17 at 1.5 miles (Livermore), to 1:19 at 2 miles (H’ville again) up to 6K (Windsor), and 1:22 at 10K (Noblesville).

When an event has been staged for many years, though - at the same location, on the same course layout - comparing times makes a little more sense. Big Shoulders is one such event.

In that spirit, here are the finish times in Big Shoulders across the 12 years of available data, starting with the 5K race:

5K times

That chart is a little busy, so let’s unpack it:

Make sense? Now, here are the 2.5K swims over the years:

2.5K times

What does it all mean? While the slowest and fastest swims each year will depend on “who shows up,” I think we can interpret the median swim as a broad measurement of “conditions.” In Lake Michigan, that generally means water temperature and/or surface chop (but usually not current).

For a swim in the same location, with the same course layout, which draws a reasonably large sample from the same population (people who live within a few hours’ drive of Chicago), we wouldn’t expect the median finish time to vary much over time. To the extent that it does vary, we can probably attribute it to “conditions.”

One probable exception is 2003, in which both the median and fastest times were substantially faster than usual. Not surprisingly, on an anecdotal level, it was widely assumed among those who participated in 2003 that the course was shorter than 2.5K.

Posted 24 August 2010 in: analysis Tags: Big Shoulders , data