Sinking my statistical life raft
A factual error from “Born to Run” helped me get through my first Leadville Trail 100
The 2013 Leadville Trail 100 Run starts at 4 a.m. tomorrow. A year ago, I was getting ready for this race. My scary, mountainous, high-altitude, first ever 100-mile race. And yet, with less than 24-hours to go – as I assembled a stack of peanut butter, banana and honey sandwiches and made sure my crew knew where to find my spare batteries and chafe-fighting lube – I was uncharacteristically calm while teetering on the edge of the unknown. Surrounded by friends and family, the anxiety about running non-stop for more than 24 hours at two-miles above sea level (anxiety that had plagued me for the past eight months) evaporated.
But my friends and family hadn’t been there to chill me out when, six weeks before the race, on what was supposed to be a mellow training run in the mountains, I started hyperventilating from the high altitude and broke down crying, because how in the hell was I going to do this thing?
During moments like this – which happened at least once a month – I kept going back to a passage I read in Christopher McDougall’s Born to Run:
Take this equation: how come nearly all the women finish Leadville and fewer than half the men do? Every year, more than 90 percent of the female runners come home with a buckle, while 50 percent of the men come up with an excuse. Not even [race co-founder] Ken Chlouber can explain the sky-high female finishing rate, but he can damn well exploit it: “All my pacers are women,” Chlouber says. “They get the job done.” (page 79)
I clung to this factoid – that nearly every woman who starts the LT100 finishes it – like a life raft. As a data geek, this statistic gave me hope in the face of uncertainty. If nine out of ten women in the past had successfully completed this notoriously punishing race, I could do it, too.
And I did. :)
Right after the race – well, sleeping, eating an entire chocolate cake, and standing in a mountain stream sipping a Budweiser & Clamato Chelada came first – I checked up on my cohort of women runners. I hoped to see that ninety percent of them finished, too.
But that’s not what I found. Not even close.
Instead, the race results showed that 55 out of 137 women who started the race actually finished the race. That’s a finisher rate of 40 percent, not 90. The men didn’t fare much better with a 46 percent finisher rate. Overall, with both genders combined, 45 percent of starters finished within the 30-hour cut off.
Was 2012 an anomaly for women? What happened to 90 percent?
So I started digging, and compiled the finisher rates for every year I could find, going back to 1995. Here’s what I found:
For the years where finisher rates by gender are available, 2002 to 2012, the overall finisher rate was 49 percent. The men’s finisher rate was 50 percent and the women’s finisher rate was 47 percent, a difference that wasn’t statistically significant.
Men and women finish the race at pretty much the same rate.
So what’s going on here? How could McDougall get it so wrong?
I haven’t spoken with McDougall (yet). But my first guess is that his source, Ken Chlouber, was exaggerating or simply sharing fuzzy anecdotal evidence. And because this information was coming from a trustworthy source – the race director – McDougall didn’t go back and check the facts.
Or, a more forgiving explanation: Perhaps the finisher rate drastically changed as the race evolved from a small, fringe race into a huge event that attracts nearly one thousand runners a year. I was only able to find data on finisher rate by gender going back to 2002 (and without gender going back to 1996). The first LT100 was run in 1983. I’m not sure how many women ran that day, but it couldn’t have been very many — it wasn’t until 1984 the women’s marathon was introduced to the Olympics. Thirty years later, when I ran the LT100,one out of five racers was a woman.Women are a minority in ultra-running, but a growing one. In the 1980s, there weren’t a lot of women running Leadville — and I bet the ones who did were darn tough. Perhaps the data from earlier years tells a different story.
I’m going to keep digging into this mystery by tracking down the historical data. But even if the women’s finisher rate was sky-high early on, there simply weren’t enough women running in the 80s and 90s to offset the finisher rate of the entrants in the past ten years. The historical, all-time finisher rate won’t be anywhere close to 90 percent.
My statistical life raft has been punctured. So where does that leave me?
Since Leadville, I’ve successfully completed another 100-miler and am gearing up for my third, the Cascade Crest 100, next weekend. At this point, I’ve learned a lot about this crazy distance – I didn’t make it through a single one of those peanut butter and banana sandwiches, and now know that I lose the ability to chew and swallow real food about 20 miles into race – and don’t need my statistical life raft any more. I’m focusing on my personal statistical truths.
McDougall’s snippet seeped its way into my subconscious and helped me achieve something incredible. And I know I’m not the only woman who carried this quote around in my back pocket and shared it with all my female friends. If it inspires more women to break into the male-dominated sport of ultra-running, does it really matter if it isn’t true?
The data scientist and journalist in me can’t say yes to that. But the runner in me can say that I’m glad I did this digging after, not before, my first 100-miler.