Wednesday, October 20, 2010

Boston Registration

Although I wasn't involved this year in the spectacle which has become the registration for the Boston Marathon, it definitely got me to thinking.  Boston has always been the crown jewel of marathons because it was the only one that you had to earn a spot in; no lottery, no luck involved.  If you were good enough to qualify you were in.  Now that has changed.  By selling out in a day we are now presented with a situation in which a person could spend years training to achieve a BQ only to find themselves shut out on registration day.  Suddenly it is no longer good enough to be good enough to run; luck, and a high-speed internet connection, has entered into the mix.

B.A.A. Executive Director Guy Morse released a statement about the recent registration process in which he not-so-subtly hints that qualifying times will be getting tighter. Bad news for those of us who have had a specific time on our radar.  Are there other ways to make changes though?  Many people bemoan the inclusion of charity runners but as far as I can tell they are only a small portion of the field, maybe 3000 or so.  Some of those numbers could be taken back for qualifiers but I don't see that happening.  The other complaint I hear a lot is that the qualifying times for men and women are radically different and that it is much easier for women to qualify.  This makes sense since there is a 30 minute difference between qualifying times in the age groups, but is it really much easier for women to qualify?

There were 468,000 marathon finishing times recorded in the USA in 2009; here is a breakdown of the numbers (click to expand).
You can see that the difference in average times and qualifying times is pretty consistent.  The total average percent difference for men and women is only different by about .6% which is amazingly close.  This means that across all age groups both men and womens' qualifying times are an average 23% faster than the national average times for all runners.

If we drill into this more and break it down by age group we can get some further info.
Let's focus on the 35-39 age group.  There is a difference of 28:25 in the average times for male and female as opposed to the BQ spread of a half hour.  That one minute 35 seconds represents a 5.57% difference in allowed and actual times which seems to be a sizable discrepancy.  But if we then look solely at the differences between average times and BQ times we see that the difference shrinks to only 2.97%; showing that the assigned BQ times are very close to being the same percentage of the avg times for both men and women.  Statistically there is very little difference between the male and female numbers and so I could probably conclude that it actually isn't easier for women to qualify.

These numbers don't necessarily tell the whole story though.  For instance, the avg times include the nearly 42,000 people who finished over 6 hours.  Further breakdown would be needed in order to determine what age group these runners fall in to.  Obviously if a large portion of them fall into one age range this would skew the avg time much more than an age group with very few 6+ hour finishers.

Seeing the numbers it would appear that women do not have a statistical advantage over men in qualifying so there is no need to tighten the requirements on the women alone.  It will be interesting to see what he BAA does to try and mitigate the rush of registrations we saw this Monday.  Was it a self-perpetuating fluke fueled by unfounded rumor?  Or is this indicative of the marathons future?  Only time will tell but if this continues I fear it may become a harsh reality for those of us who may be good enough to qualify today to see our chances of qualifying become increasingly out of reach.


  1. "Seeing the numbers it would appear that women do not have a statistical advantage over men in qualifying..."

    I disagree! I think a more helpful statistic is to compare the gender breakdown of one of the big urban marathons with the gender breakdown of Boston, for the Open division. New York is about 60% Male. Boston is about 40% Male. What explains that discrepancy, if not higher relative qualifying difficulty in the Open division?

    Put another way, you compared average times in the general population (4:45 vs 4:18) and concluded that the difference was about the same as the BQ qualifying times for the same 25-29 age group. But, the difference in relative difficulty between 4:45 and 4:18 is way lower than the difference between 3:40 and 3:10!

    I do think the BAA should just lower qualifying times across the board, though. Even if it means I might not be running it any time soon. Boston should be a goal to be honorably achieved in competition, not a lottery.

    Glad to hear your foot felt better!

  2. David,
    Here are the stats for Boston Finishers in 2009:
    Finishers: 22849, Males - 13547 (59%), Females - 9302 (41%)
    This is right in line with total percentage of male to female marathoners at large.

    While it seems that however they come up with the BQ times is consistent I agree with you that it does not take in to account the actual effort it takes to achieve some of these times. A 3:10 is exponentially harder than a 3:40.

  3. right! I was talking specifically about "open" stats (i.e., ages 18-39). In that group, men are 5455 / 11397, or 48% (which I admit is a lot closer than the number I remembered and quoted above--perhaps things changed since 2009). Still, an interesting discrepancy compared to the normal 60/40 split.

  4. Great post Jeff. Very interesting. I'm back blogging and Yes, I have left a million and 3 comments on other blogs. :) Hope to see you soon.