How Getting Older Affects Performance

We have been interested in how getting older affects race performance.
Personal experience tells us we're getting slower as we get older,
even though we train better, we eat better, we maintain our weight,
and we know more about racing and our bodies. Is it just us, or does
it happen to others as well? How much should we expect our
performance to decline as we "age up?" How do our "age-adjusted"
results this year compare with those of ten, fifteen, twenty years ago?

To answer these questions, we've been doing a bit of research. We dug
out the results of the ITU World Championships and the USATriathlon
National Championships (international distance) we had available and
looked at how winning times in each of the five-year age categories
changed as age increases. Here are a chart that presents what we
found and a table that shows raw numbers:
Data on Worlds and Nationals

Data from the USA Nationals and the World Championships for selected
years since 1994, updated for the 2000 ITU World Championships in
Perth, Western Australia:

Age Group 00Worlds 99 Worlds 99 Natl 98 Worlds 98 Natl 97
Worlds 97 Natl 96 Natl 94 Worlds 94 Natl Av Time Ave Wgt
F20-24 1.003 1.000 1.048 1.000 1.000 1.029 1.013 1.051
1.000 1.047 2:13:29 1.019
F25-29 1.000 1.004 1.036 1.022 1.035 1.000 1.007 1.014
1.016 1.019 2:13:01 1.015
F30-34 1.016 1.014 1.036 1.007 1.003 1.038 1.000 1.000
1.017 1.000 2:12:41 1.013
F35-39 1.040 1.021 1.000 1.048 1.037 1.036 1.064 1.054
1.035 1.042 2:15:55 1.038
F40-44 1.063 1.062 1.079 1.049 1.061 1.055 1.054 1.094
1.074 1.067 2:19:37 1.066
F45-49 1.107 1.099 1.095 1.098 1.066 1.127 1.102 1.120
1.088 1.126 2:24:24 1.103
F50-54 1.113 1.142 1.226 1.149 1.123 1.228 1.215 1.186
1.171 1.179 2:33:38 1.173
F55-59 1.271 1.243 1.285 1.213 1.231 1.278 1.238 1.190
1.215 1.208 2:41:59 1.237
F60-64 1.279 1.230 1.256 1.257 1.267 1.262 1.318 1.410
1.466 1.321 2:51:09 1.307
F65-69 1.433 1.364 1.377 1.575 1.469 1.432 1.493 1.374
1.406 1.473 3:08:43 1.440
F70-74 1.524 1.611 1.728 1.611 1.847 1.568 1.647
3:35:48 1.648

Last Updated on 5/20/2000 by Sue Falsey


So what does all this mean?

The chart tells us that slowing down starts in the 30's, increases at
an accelerating rate in the 40's and 50's, and then accelerates even
faster in the 60's and 70's. Now the chart is based on the fastest
woman in each age group and doesn't reflect the average or the average
of the top ten placing in the age group. The fastest woman is
undoubtedly a highly trained and gifted athlete, not a "middle of the
packer." But still, it shows what's humanly possible. And since we'd
all like to be the winner of our age group, it shows what we'd need to
achieve.

Sue is a PhD whose program included a heavy dose of statistics, while
I trundle along with one undergraduate course in statistics and a love
of numbers. So let me try to explain what some of those lines on the
chart mean.

* The faint colored lines connect the winning times of the age
groups for the 10 races we plotted.
* The heavy red line is the average for all 10 races.
* The black heavy line plots the simple regression (or average) as
a straight line. It's not terribly useful because the data really
indicate performance changes follow a curve, not a straight line. If
we could show the other side of the chart -- times for the 0-4, 5-9,
10-14, and 15-19 age categories (to balance things out), we'd have
faster times as one moved up in age category, and the regression line
would probably be almost straight across, like the water line in a
bathtub.
* The blue line plots the average for all 10 races as a complex
polynomial equation, a kind of "best fit" curve conforming to the race
data. This is probably the best graphic of what happens to
performance as we "age up" (we triathletes/duathletes don't get older;
we just enter new age classes).

The table following the chart presents the race results, with the best
time as 1.000 (100%) and slower times as more than 1.000 or 100%. In
other words, if your time was 2 hours 30 minutes (or 150 minutes for
purposes of calculation) and the winning time was 2 hours (or 120)
minutes, your score on this table would be 1.250. That means your
time was 25% slower than the winning time.

You'll notice that the winning time jumped around among the age
classes between 20-24 and 35-39, with the 20-24-year-old the overall
winner in 4 cases, the 30-34-year-old the winner in 3 cases, and the
25-29-year-old and 35-39-year-old the winners in one case each. Going
over to the right, you'll see that the 30-34 age group had the fastest
average time, followed closely by the 25-29 and 20-24 age groups. The
"Average Weight" column expresses these time differences as scores (or
percentages).

So from this we can present how much slower, as a percent, the winner
might be as one moves to older age classes:

* 35-39 winner of the age group is 3.8% slower than the
overall winner
* 40-44 6.6% slower
* 45-49 10.3% slower
* 50-54 17.3% slower
* 55-59 23.7% slower
* 60-64 30.7% slower
* 65-69 44% slower
* 70-74 64.8% slower

There are a bunch of caveats that need to be mentioned before you go
looking at your old times and calculating what's happened to your
performance over the years:

* Many if not most triathletes/duathletes improve during the first
few years of participation in the sport. I've read that it takes 7
years of training to reach one's peak as a runner and longer,
particularly for those not in swim racing as kids, to do the same in
swimming. In short, your performance due to mastery of the discipline
may more than offset any decline due to aging during the same period.
I ran faster per mile at the 5 kilometer distance at age 50 than I
did at age 45 -- and than I did at age 18 as a college freshman on the
crosscountry team. But the table above does show you what "drag" due
to aging you need to overcome to get faster absolute times through
achieving mastery in the disciplines or training better or...
* Times in the older age categories are getting better as more
people stay in the sport longer, maximizing their mastery of the
disciplines, experience, and training. Back in the 1980's, the oldest
contested age group for women was 50-54; now that age group is a
competitive hotbed overflowing with talent. Of course, times are
improving also in the overall women's winner category, but more slowly
than the improvement in the older age categories.
* This table sets its baseline on the fastest overall age group
woman. The elites do go faster yet, and the bulk of them are in the
20-24, 25-29, and 30-34 age groups. But the elites don't compete in
the national age group championships and have different courses (and
rules, like drafting, that effect times) in the world championships,
so we can't really construct a comparative table with an elite woman
as the overall winner with the 1.000 score.
* Triathlons don't really have standardized, equalized distances
and conditions the way some other sports, such as track and field,
have. Your performance on any given day is affected by the hills, the
wind, the weather, how you cut the tangents, and a host of other
factors, assuming the course is accurately measured to the mouse's
eyelash (and how many swim courses do you think are measured
accurately down to the last meter?). Everyone did the same course on
a given day, right? So we compare performance based on a percentage
time slower than the overall winner that day -- as does the
USATriathlon ranking system. But other factors enter in when
comparing across races.

So why should we care about all this?

Well, probably no reason at all if you don't want to. But it has some
applications, if you take it with a grain of salt and realize it's not
perfect:

* Probably one use is to give yourself some expectation as to the
range of "slowing down" you can expect as you age up. "As one gets
over the hill, one accelerates as one goes down the other side."
* You can extrapolate what you might have done as a 22-year-old,
or how you might do against a son or daughter if you were the same
age. "Why, my time would have been 40% better if I were 35 rather
than 66..."
* If we were trying to judge between outstanding performances on
the part of a 60-year-old and a 75-year-old for the annual Grand
Master award, we could do some comparisons on their performances vs.
what one might expect at their respective ages.
* We could have an age-graded race to see who wins. The Dipsea in
California does this by modifying the start time for each competitor
based on his/her age/sex. US Masters Track and Field Championship is
going to have an age-graded 100 meter dash this year, where the
starting lines for each competitor are shortened according to an age
adjustment factor. (On the other hand, I'd hate to be in a triathlon
where all 500 competitors hit the finish line at the same time!)
* We could calculate and pick out outstanding age-graded
performances. If you read National Masters News, you'll find they do
this quite frequently in track and field and road races. They even
make their All-American awards based on age-graded performances.

What would be interesting and perhaps helpful to a lot of other
athletes, would be some case histories of outstanding athletes -- men
and women -- and how their performance changed as they "aged up" to
new age groups. But even such performance case histories would have
been influenced by the effects of continuous and improving training,
better mastery of the disciplines, and improvement in the performance
of the overall winner "rabbits" who set the par (1.000) scores. We
live in an imperfect world!

For another take on age grading, please visit Lew Kidder's work on
CoolTri. Lew gives some applications to recent triathlons and
duathlons. We agree on most things, but do have a "nerdy" difference
of opinion over whether the fall-off in performance accelerates as an
athlete passes 40, 50, 60, 70...

What about the men?

Yeah, what about the men? If you want to see a comparable chart and
table for men, please click here -- men's age-graded results.

Reactions, comments, and suggestions are most welcome! E-mail us!

Back to top of page Back to Age Group Commission page
Date Updated: 05/15/06