Are Boston qualifying standards unfair?

Posted by Dave on September 1, 2011 | 8 Comments

I’ve heard it mentioned several times that “women have it easier than men” when qualifying for the Boston Marathon. The argument usually goes something like this: Boston should be accepting the same proportion of talented male runers as female runners. But women get a 30-minute advantage in every age group, when the men’s world record in the marathon is only about 11 and a half minutes faster than the women’s record. It seems like a fairly compelling argument. If men are only 11 minutes faster, why should women get a 30-minute advantage?

As you probably know, the Boston Marathon is one of the few marathons that sets a minimum qualifying standard for most of the runners in the race. The only exceptions are people running for charity or who have affiliations with the sponsors of the race. These standards vary both by gender and age group, the idea being not just that men shouldn’t directly compete with women, but also that 50-year-old men shouldn’t be competing with 28-year-olds for starting spots.

But I’m not sure that the world-record argument makes sense when we’re talking about Boston qualifying standards. For the 2012 race, the fastest qualifying time, 3 hours and 10 minutes, is more than 50 percent slower than the world record. Clearly the typical qualifier doesn’t have much in common with a world-class athlete.

So I decided to do a different calculation. I took a sample of 8 popular marathons run in the past year, and then looked at what portion of men and women qualified for Boston in selected age groups. If women really have an unfair advantage, then a larger percentage of the women competing would qualify, right?

Below is the basic data:

I looked at three age groups: 20-24 (or 18-24 in some races), 35-39, and 50-54, and as you can see, for each race, there were more male qualifiers than female qualifiers. But, of course, there were also more male participants. What portion of males qualified compared to the portion of females qualifying? As it turns out, the percentage was identical for males and females: a 9.1 percent qualification rate.

Let’s take a closer look at the data:

While men outqualified women in the 20-24 and 50-54 age groups, women outqualified men in the 35-39 group. Overall, the difference between sexes is a wash. You might actually make a better case that the rules favor older runners, since there’s a distinct trend toward higher qualification rates for older runners.

But both of these observations are dwarfed by another one: The particular marathon a runner chooses to qualify in. Take a look at this graph of qualifying rates, combining the results for both sexes:

The qualifying rates for these marathons range from just 4.1 percent for Disney, up to 31.5 percent for Steamtown. It’s possible that the numbers are inflated by “easy” races like Steamtown and California International attracting better-qualified runners, but it’s clear that the difference between the sexes, and even between age groups, may not be the most important factor in determining who does and does not qualify for Boston.

Comments

8 Responses to “Are Boston qualifying standards unfair?”

  1. Roberto
    September 2nd, 2011 @ 12:18 am

    This post would have been better, I think, with an explanation of how Boston decided its qualifying procedure (recently amended), and what they think of criticisms that the standards are too soft for women. The WSJ ran a piece on the same subject around 11 months ago, but also neglected to get an explanation from the horse’s mouth.

    http://online.wsj.com/article/SB10001424052748703673604575550133914934718.html

  2. Dave
    September 2nd, 2011 @ 5:15 pm

    Hi Roberto,

    You’re probably right that this post could have been better, but I don’t think this blog yet has the clout to get an explanation of the BAA’s decision-making process directly from them.

    To respond to the WSJ article, I still dispute the notion that an 11-minute disparity in the world records means anything for the recreational runners who are trying to qualify. They are taking 50 to 100 percent longer than the elites to complete a marathon, so the 11-minute difference wouldn’t be the same over that longer time span. If I run 11 minutes slower for double the time, it’s really a 22-minute gap.

    What’s more, recently men have been showing more improvement in elite times compared to women. This year’s male Boston winner finished in 2:03, while the female winner finished in 2:22, nearly 20 minutes slower. Add 50 percent to those times to get to the fastest qualifying times, and the gap is once again 30 minutes.

  3. Roberto
    September 2nd, 2011 @ 5:45 pm

    Absolutely true. I had a piece in the Wall Street Journal this week, in celebration of the World Athletics Championships, comparing the relative strength of world records (i.e. showing the margin of the record over the next best performer’s mark).

    Everyone agreed that standard deviation from mean is the best way to do that, but the data set of performances is open-ended, and it’s difficult to get a mean (I did a straight percentage comparison). I compared the records to 10th on the list (there wasn’t space in the paper), and the data is available at least 100 places out, but as you note, that’s all elite data.

    Likewise, the WSJ’s inclusion of Western States entry standards fails to note that at ultra distances, the difference between good men and good women is small; in those events, given a certain fitness baseline, success or failure mostly boils down to determination. [That’s not true for elite marathon runners, of course, who are (men) running 3:00 kilometers 42 times in a row.

    This year’s Boston data was extraordinary, I think, though, and elite marathon data in general suffers from an abundance of top-level competitors on the men’s side, and relatively fewer on the women’s side. Many more men than women think they can make a living at marathon running, I think, and that means the level of competition is very high.

    Lots of factors, but making your point, I think, that elite data has very little to do with the man/woman in the street.

  4. Alex Hutchinson
    September 5th, 2011 @ 4:33 am

    Great post, Dave! Very interesting to see the results of your data-crunching.

    “It’s possible that the numbers are inflated by “easy” races like Steamtown and California International attracting better-qualified runners…”

    I think it’s more than possible. The differences in “culture” between marathons are enormous. Marine Corps is branded as “The People’s Marathon,” and doesn’t even offer prize money, and Disney is also focused on the masses. Races like Steamtown and Cal aren’t just “easy” courses — they’re also marketed very aggressively to would-be Boston qualifiers. The resulting differences in proportion of BQs isn’t at all surprising, and I’d argue is primarily due to which runners show up. The proof that it’s not just about the “easiness” of the courses is Chicago’s low ranking: Chicago is a very fast course (multiple world records set there) but also a very popular marathon (second- or third-largest in world), so it has a relatively low proportion of qualifiers due to the profile of runners it attracts.

    “…it’s clear that the difference between the sexes, and even between age groups, may not be the most important factor in determining who does and does not qualify for Boston.”

    Definitely! Your analysis makes a strong case that the key factor is… how fast you are.

  5. marty
    September 6th, 2011 @ 1:02 pm

    I don’t know if this analysis necessarily does what you want it to… If the question is: “is it easier for a woman to hit the boston qualifier than a man” then the world records are not a bad place to start, being that those are currently the best someone who was really trying could do. But maybe look at their relative position…If the men’s wr is 10% faster than the women’s then BQ time should be…

    The problem with just comparing the porportion of people who made it out of various (and kind of oddly chosen–Steamtown? Rock and Roll AZ?) marathons, is that men may just be more competitive runners on average, therefore they SHOULD be over-represented porportion wise. If women were less competitive (which btw i don’t necessarily think they are), then this analysis would show that the Women’s boston times ARE easier.

  6. Patrick Makau
    September 27th, 2011 @ 3:17 pm

    Considering women have been running marathons for decades less, I think it likely that we will see more and more women running qualifying times as the notion of uterine displacement goes the way of the dodo :)

    The reasons are so diverse – everything can be thown into the mix from ideas that men prefer team sports to the notion that women are better equipped to deal with endurance races. In the end though, it seems the Boston qualifying times are no more arbitrary than any other numbers which commentators pick out of their – ahem – sneakers. Some good recent articles on this:

    http://runningtimes.com/Article.aspx?ArticleID=22159

    http://runningtimes.com/Article.aspx?ArticleID=21003

    Also I’d agree with Alex (sycophant? moi?) but the profile of the event is all important. London and NYC for instance are huge charity events (in London it is getting harder and harder to run simply as an individual runner) so the proportions of qualifiers will be minimal. CIM’s downhill route has people clamouring to run there in some sort of self-fulfilling prophetic scenario, wheras Big Sur runners are more about the scenery than the time (this isn’t to say that CIM isn’t scenic…)

    Not that your blog was saying this – but if a male has an issue not qualifying for Boston, blaming it on filling up wth females and veterans, then they just need to train harder and/or smarter, and remember they likely had more sporting headstarts than the females who they are now ‘competing’ with for places at Boston.

  7. jim
    December 27th, 2011 @ 11:28 pm

    A potentially significant factor that is not reflected in the qualifying rate data above is the number of individuals who post multiple qualifying times in one year. I suspect, given the nature of many marathoners, that the typical Boston Qualifier individual runs two or more qualifying times per year. That would suggest that the actual qualifying rate is closer to 5% than the generally accepted 10%.

  8. Birgit
    May 20th, 2012 @ 11:35 am

    I find it interesting that the higher rate of qualifiers among women are in the 30-35 age group. I have just started running, but have practiced martial arts for quite a long time, and find that women often don’t start until their 30’s, when they have finished reproducing. It was certainly that way for me. If you do tend to start late (and until the last couple of decades women were discouraged from high exertion athletics for “health” reasons), it will take you a while to ramp up to reasonable speeds and endurance.

    Also, I find the time it takes to train for any race is immense, and the longer the race, the more all-consuming. My sense is that this will impact women somewhat more than men as well, due to their social conditioning regarding family.