Thursday, February 24, 2011
BQ Standard... Revisited
So, I decided to do my own analysis. Sean had mentioned that it would be interesting to look at the difference in finishing times for the top athletes in each age group. So, I did. I chose the 2010 New York City Marathon to analyze (there is a qualifying standard in place, but there is also a lottery system). I averaged the finishing times for the top 25 athletes (elites included) in each age group (well, that's a lie... if you're 70+ and still running, then you deserve to run a marathon). This first graph shows the average finishing times for the top 25 athletes in each age group.
Not too exciting... no surprise... men are faster than women. However, this appears more pronounced at either end of the "age spectrum" (this may result from the fact that there were more men running in these age groups compared to women... the standard deviation was much larger in the case of the later). To get a closer look at the difference, I graphed the difference in the average finishing times between the two genders for each age group (again the top 25 athletes were used).
So, we were right. In fact, the times are not so different between males in females ages 25-29... and that half hour BQ difference is looking a bit suspect until, well, 50+. And, just for kicks, I plotted the difference between the 2013 BQ standards and the average finishing times (again the top 25 athletes were used).
So, my analysis confirmed my gut instinct... the BQ standards for males aged 20-34 is proportionately more difficult than for females in that same age bracket. In fact, the 30min difference really shouldn't be applied until the 50+ age bracket. And, well, I really think 3:05 is a bit crazy for males aged 18-34... so I feel that knocking this back up to 3:10 would be appropriate. In fact, leaving the "old" standard in place for males aged 18-44 would work well to "even up" the time difference between the BQ standard and the average finishing times, as graphed below.
I hope you have enjoyed my analysis... take it for what you will... my worst grade in grad school was in statistics!
Subscribe to:
Post Comments (Atom)
This comment has been removed by the author.
ReplyDeleteJust a couple quick averages for counter argument's sake:
ReplyDelete(From the 2nd graph of difference in avg. finishing times for each gender, red bar graph):
The average of all of the differences in finishing times for all AG
= approx. 33 min
The average of the differences in finishing times for AG 20 to 64 (drop first and last AG, based on demographics the 1-19 and 65+ AG had the fewest participants)
= approx 29 min
Also just calculated a weighted average of avg difference between genders for AG 1 to 69 based on the percent of total finishers that each AG was comprised of, i.e. AG 40 to 44 had the most finishers (19% of total).
ReplyDeleteWeighted difference is 28.5 minutes
Yeah... the average is around 30min... I just feel that a tougher standard should be applied to the "younger" females, specifically in the 20-40 year range... but I guess if they were just going to do something "across the board", what they did was fair... my main problem I guess is that I didn't agree with the old standard!
ReplyDelete