Attrition and Resilience

I have been claiming for some time that 80% of runners are no longer racing by age 60. I reasoned this by looking at results of various races, such as the BAA Distance Medley races that draw 10,000+ runners.  I’ve also looked at large datasets, such as RunSignup’s annual RaceTrends report.  Their 2024 report assessed 10.8 million registrations in the U.S., which they estimate represents 45-50% of total race registrations. This report indicates 7.4% of registrants were in their 60s and only 2.5% 70 and older.  On the surface this might suggest an even greater than 80% drop-off, but that could be wrong for several reasons.  One, this blends race distances from 5K to ultra and even includes triathlons, which probably tend to draw younger athletes. Two, the RunSignup dataset notes nearly 20% of participants are under 18.  I suspect this includes a lot of high school events, also evident by the preponderance of 5K races representing 65% of registrations.  Also, it’s possible, if not likely, that younger people run more races each year. Let’s assume the average 60+ runner runs three races a year (yes, some of us run much more than that!) and those younger run an average of five races a year, that would get us in the range of 16% of unique runners being 60+.  This is a ballpark but it does not seem too far-fetched to estimate ~20% of these runners are 60+ (16/84% = 19%). The next step to consider is whether this infers an 80%  drop-off.  An issue is that the dropout rate accelerates with age. Meaning, it might be plausible that by age 65, 80% have dropped out from racing, but not by age 60.

Data from specific races avoids such assumptions.  With that in mind, I looked at the results from the September 7th North Shore Runfest 10K in Salem MA, which served as the USATF New England 10K Championship. The Second Wind timers provided results from the 484 Grand Prix participants as well as the full field of 882 runners (which shows how valuable it is for a race to host a Grand Prix event!)

What I like about data from Grand Prix races is by and large the club runners are “serious” runners, meaning they regularly train and make racing a priority.  These are the runners I am interested in understanding! Data from a broad range of runners include those who might run once a week and jump into a race because someone at work or a family member is doing it.  Having a lot of these “occasional runners” embedded in results makes it really hard to draw valid conclusions.  The other thing I liked about this race is it was a 10K.  According to RunSignup, only 9% of races are 10K down to 5K+ (i.e., this includes 5M, 8K, and 4M races), whereas 53% of events are 5Ks or less, attracting 65% of total runners. I surmise a greater percentage of less serious and younger runners show for 5Ks than 10Ks.

As shown below, of the 484 runners at the Runfest, just 92 or 19.0% were age 60 or older.   Since 81% of the runners were younger, maybe this supports the notion that 80% of former runners have dropped out.  Yet, even more stark was that only 20 — 4.1% — were 70 or older.  This is clear evidence of attrition! 

Age	Male	Female	NSR 10K Total	% of Total	% of RunSignup Data Set	% of U.S. Population
70+	15	5	20	4.1	2.5	12.9
60 – <70	39	33	72	14.9	7.4	12.0
50 – <60	57	46	103	21.3	12.8	12.7
40 – <50	62	36	98	20.2	18.0	12.4
Under 40	111	80	191	39.5	59.3 *	50.0
Total	284 (59%)	200 (41%)	484	100.0	100.0	100.0

           * Included 19.4 % under age 18

I noticed 59% of the Runfest field was male. Yet, the RunSignup data indicate 54% of 10.8 million registrants are female. Maybe for this 10K, some clubs were male dominated (e.g., the home-team Notch Run Club had 81 runners and 57 were men.)  However, the Bobby Doyle 5M in August also showed 59% of the 470 USATF runners were male (with 18.5% of the total runners 60 and older.)  So, while there are exceptions (e.g., Liberty Athletic Club and Sole Sisters, both all women) it seems USATF clubs send more men to Grand Prix races.

I also found it interesting to compare the RunSignup data to the U.S. population.  The 70+ population is only about one-fifth as likely to be racing, whereas those in their 60s are about 60% as likely – it seems fewer in that cohort are choosing the couch compared to the 70s! Those in their 50s are nearly even with the general population while runners under age 50 are significantly more likely to be racing, due in part perhaps to youth running.

Digging into the North Shore data implied two other things.  One, while the oldest cohort (70+) is somewhat less inclined (no doubt for various reasons) to enter road races, those younger than 70 appear significantly more involved in racing at a “serious level” (i.e., they run the Grand Prix!) than both the RunSignup participants and the general U.S. population all the way down to age 40. Two, there appears to be a groundswell of youth runners coming into the ranks.  The unknown is whether they will continue to run in their 20s and beyond.

Maybe this analysis adds little (or nothing!) to the body of knowledge favoring running as an activity that helps keep us resilient – active and functioning at a reasonably high level into our later years. While it is encouraging to see the percentage of runners over 60 still engaging in competition, it does not mask the reality there is a big dropout rate and that this dramatically increases with age.