There’s definitely an allure to running the Boston Marathon if you are a runner. Personally, I don’t find it my supreme goal to race there, but I would by lying if I said it isn’t on my radar. I’ve passed up registering for four years now, and I’m not really sure why, other than I don’t really have the drive to race a big race like that just yet.
That being said, in my 50×50 list, I do have it laid out that I plan to do Boston as my Massachusetts marathon. I’ve qualified (so far) at every open marathon I’ve run (and even at one iron-distance race, if that counts, which I do not think it does). Last fall, when registration opened for the 2011 Boston Marathon , it sold out in lightning speed. So this brought the folks at BAA to reconsider the qualifying times. After months of deliberating, the time has come.
Yesterday, the BAA released the new standards. First, let me start by saying there is no real easy or perfect way to figure out these types of things. It’s not simple math; there is no real straight forward approach to saying one time is fast enough and another isn’t. There’s statistics involved, means and standard deviations, and the data might not follow a Gaussian distribution (in other words, the spread might not be normal like a bell curve).
If we look at the world’s best for guidance in determining these standards, for example, we see this: The fastest marathon time for a woman in 2010 was nearly five minutes slower than the world record, set in 2003 by Paula Radcliffe. For men, it was a different story. The fastest marathon time last year for men was less than a minute slower than the all-time best set just two years ago. So, the men’s times are a little closer packed together. This makes sense, in a way, because the faster you get, the more those seconds and milliseconds start to matter (ask any sprinter what their best 200m time is… you’ll likely get a different resolution than if you as a marathoner their PR).
OK… So let’s look further at the data. More recent years, men are getting faster and being more competitive with the world record in the marathon. Women, however, are having a harder time. In the following figure, I’ve compiled the top performances in history for men and women in the marathon. In the data, which I acquired on the website for The Association of Road Racing Statisticians, I made a list of the top 30 times in history for both men and women, as well as each time’s respective year of accomplishment. In the plot below is a graphical representation of the data, where women’s times are in blue triangles and men’s times are in red boxes. There’s an obvious difference in performance times; men are approximately 15 minutes faster (on average) than women. Also on the plot you will see two lines. The location of the colored lines illustrate the mean year from the top 30 times for both males (red line) and females (blue line). It is clear that the blue line for females occurs much earlier than the males’. Now, if you’re a statistics geek like I am, the data shows some interesting trends. Take for example the horizontal spread of the blue triangles compared to the red squares. I performed an F-test for variance comparing, first, the years of the top 30 performances between men and women. This gives somewhat of an idea about when the performances were made, whether or not the times saw any dramatic drops over history (if so, the dots should be clustered together) or if there has been any stagnation in performance. Interestingly (and somewhat obvious), there is a significant difference in the variance, or the spread, of years in which the top 30 performances were made, when comparing men and women. And, the F-test demonstrated significant differences in variances, underlining that there’s more than a 99.9% likelihood that the temporal spread between these two groups differs. This could mean a lot of things, but I see it as possibly suggesting that the men continue to beat down the door of the marathon world record, whereas the women had a few great years (in the early 2000s) and haven’t really got back there since. This is a difficult one to translate to the average marathoner; if we were to extrapolate these findings (from data of professional world-champion level runners) to the average marathoner, this might suggest that women’s personal bests aren’t generally improving year after year, whereas men’s (likely) are. In other words, men keep ticking off seconds from their best times, whereas women are just as fast now as they were about ten years ago. [I don’t like to extrapolate data, and I wouldn’t put any support behind that statement. I am simply being facetious).
It’s also interesting to look at the spread in the data for performance time as well. If you look at the vertical spread – or in other words, the variability- in the top 30 performances, for both men and women, you can see that the red squares are pretty tightly packed, whereas the blue triangles are spread out a little more in this direction as well. The F-test of variance on this data is also convincingly significant, Again, this could mean that there were a few standout times (notably in the early 2000s) for women. If we take out the women’s current-standing world record time (which is the blue triangle that rests in the 2:15:22 line), the vertical spread for women is a bit more similar to the mens. What does this mean? Well, again, the standout times are one thing. But also, men are at a point where they are having a hard time getting a lot faster. Women, on the other hand, have demonstrated their potential to go fast but don’t have the depth in performance history. It could be simply that women have not been running marathons competitively nearly as long as men have. The first time a woman broke 3 hrs in the marathon was in 1971, whereas men have been under 3hrs since at least 1908.
So what does this all mean? Well, it’s hard to say. In general, the time differential between the world’s best male and female marathons is about 15-16minutes.
But if we look at results from last year’s Boston Marathon, there’s a different story depending on what age group you look at. For the 18-39 age group, top men and women times differ by more than 20 minutes. In the 45-49 age group, for example, the discrepency between top male and female times is a little more than 30 minutes. How does one come up with a standard that is easily translatable and most fair to everyone? The BAA made a decision, based on knowledge and data, to keep men and women’s qualifying times separated by 30 min in each age group. Sure, this might make things a little more difficult for the 18-35 yr olds men, but how many men do you know play the “I’ll wait til I’m 40 to qualify” card so they get an extra 5 minutes? And, how would they get away with saying: “You 45-49yr olds get a gender gap of 30 minutes, but you younger kids in the 18-35 range only get 15.” I don’t think that would go over too well either.
As far as the rolling qualifying times go, it makes sense – at least to me – and I don’t think anyone who is confident that they can beat the qualifying by a good margin has to worry about getting into BAA in 2012. Besides, its I think it ups the bar a bit. The race this year sold out in 8hrs, but how many of those people were “squeakers”, people who JUST got under the qualifying time and were waiting by their computers hoping above hope to get in? I would imagine, a lot. In 2013, things will get a little more difficult, but I think the storm will calm when everyone sees how the 2012 registration goes. And for women to run 3:15 is not uncommon, per se, but it is an atypical goal, and it’s not easy. As a friend of mine pointed out, the “just making it” qualifying pace groups at some easy-to-qualify-at races are generally packed. Will these athletes try to get faster, or will they just run for the fun of it?
And think about how many women and men run the first round qualifying times? Is the race filling up really something to worry about? Now move down to the second qualifying time (which is 3:25 for women and 2:55 for men). Not as hard, but still- not easy. I think I finished around 40th when I ran a 3:22 at Columbus, and that was a relatively large and fast course. So the odds of getting in with a 3:25? Still pretty good. To be honest, had I really wanted to race Boston this year, I would not have got in (I think the registration date was when I was traveling for interviews). And with a 3:22, I would have been quite disappointed that I didn’t get in, knowing that some others in my same age group most likely got in with 3:40:59s. For me, I want to have wanted Boston be a competitive race, but not even giving me the chance to race takes me out of being competitive whatsoever.
In the end, I think the BAA did the best they could to make the playing field even. The new standards act as a sort of filter, and the time periods for registering under each qualifying time are still restricting. It’s not like they are allowing the 2:45 male marathons an entire month to register and fill up the race. And… maybe it is a little elitist, but hasn’t that always been the Boston way?
As an aside: No one complains about the NYC Marathon qualifying times, and maybe that is because of their amount of lottery and charity entries. Boston is making their early qualifying standards somewhat comparable to NYC Marathon’s guaranteed entry times, only not having those charity and lottery spots. In that way, it is remaining different than NYC and not falling under the peer pressure of making everyone happy (which is an impossible thing to do).
These thoughts and opinions are strictly my own with the help of data from online sources (hyperlinked).