New study shows how much slower marathoners get as they age

This study has a design flaw. Median performance doesn't reflect age-related decline as much as age-related dedication to training, which is more cultural in nature.

Looking at winners is better, but it also is flawed because outliers like Meb mess up the results.

A better test of age related decline would be to look at the top 5 finishers in each age group. This is especially true in older age groups, where one person can really mess up the results. Take Ed Whitlock for example. His performance is not an indication of age-related decline for a person his age.

amazes me how people are quick to critique, but don't read... read the methodology, dude - they eliminated the outliers from their model, which makes your point moot:

"Since regression is very sensitive to extreme cases, outliers were removed. Data were screened by identifying outliers. Any outlier, defined as ≥ 3.0 standard deviation of the residuals were eliminated. A plot was created between the standardized residuals (y-axis) and standardized predicted values (x-axis) to see if the values were consistently spread out, which would indicate normality and homoscedasticity. "

looks like they put the dedication to training stuff as a limitation

"It is accepted that human physical performance declines with advancing age and there are multiple theories available to explain this mechanism besides the obvious reduction in aerobic capacity. Some of these theories include decreased motivation [26, 38], the decline in the magnitude of the training stimulus [39], and orthopedic issues [40]. In champion runners, the former two may occur but are unlikely to explain performance declines in the elite population, in particular for those professional runners whose financial income is dependent on fast finishing times."

Another problem is that pro runners often are not necessarily running for time. For example, someone who is in the top five in a big marathon after mi 20 may lose contact with the leaders after a big surge. But that runner is often well ahead of the 6-10 place runners and can safely jog it in to get decent prize money. Eliud Kiptanui ran 2:07 in Berlin to lose by 4 minutes, but was 2 min ahead of 6th place. Why would he try squeeze out an extra minute on his overall time when he could not catch the 4th place runner?

Um ever heard of age grade tables?

i think recreational runners slow down for a lot of non-physiological reasons (like having a demanding job or having young kids) and, since these aren't really bad physical limits, can reverse them at will by training more. I know I personally have a 5k time that represents acceptable fitness for me and if I get much slower than that, I get more serious about training until I can hit it again.

omitting "outliers" is a good way to blur over the fact that so-called age related decline is largely due to people not training as hard or believing in themselves. if there were a necessary linear decline due to reduced capacity, then people like Meb and Geb would not exist. As it stands, they were able to run at or near their personal bests far beyond age 35. If you keep training hard, you should be able to hold pretty close to or even at your lifetime bests at most distances until you're almost fifty.

yeah, but they still decline like the paper says. near one's personal best is not the same as at one's personal best. Geb can still be world class, but not good enough to be winning, because he has gotten slower with age.

Using races with time-based qualifying standards will/does skew the results. The qualifying standards end up influencing the outcome more than personal performance does.

Wiki:

"Derek Turnbull (born 5 December 1926 in Waikaka, New Zealand - 2 November 2006) was a New Zealand runner. A farmer, he took up the sport of long distance running in his early 40's.

As a pioneer in masters athletics he achieved worldwide acclaim by setting numerous Masters Athletics World Records. Several of those records, set in the 1990s,[1] remain the current world record in those age divisions.

Three of Derek Turnbull's notable marathons include the New York Marathon in 1990 placing 1st for his age group 60-69 in a time of 2:41.21. Derek also placed first for the London Marathon, 2:41.57 in 1992 in the 65-69 age category. However his fastest marathon time in the 60+ age group was at the Alaska marathon in 2:38.46 in 1987."

Exactly like the researchers said in the paper. Seriously people you look like fools if you post criticisms that the researchers have already discussed - it makes it clear that you aren't actually reading it. Not everything can be included in the title or abstract.

I have a theory too wrote:

Using races with time-based qualifying standards will/does skew the results. The qualifying standards end up influencing the outcome more than personal performance does.

captain obvious (or oblivious) wrote:

Exactly like the researchers said in the paper.

I have a theory too wrote:

Using races with time-based qualifying standards will/does skew the results. The qualifying standards end up influencing the outcome more than personal performance does.

I know, they presented it as a limitation, it skews the results.

You may want to brush up on your reading skills.

3hr marathoner wrote:

i think recreational runners slow down for a lot of non-physiological reasons (like having a demanding job or having young kids) and, since these aren't really bad physical limits, can reverse them at will by training more. I know I personally have a 5k time that represents acceptable fitness for me and if I get much slower than that, I get more serious about training until I can hit it again.

I have to agree somewhat to the above statement as well as motivation factors come into play. I feel like the last couple years has taken a toll on me w/life changes, etc. and therefore I don't feel like I'm giving my all or the best ability to train. I hope to change that in couple months when I need to train seriously for a full marathon in Oct....😬

zx.vm.sadfljk wrote:

omitting "outliers" is a good way to blur over the fact that so-called age related decline is largely due to people not training as hard or believing in themselves. if there were a necessary linear decline due to reduced capacity, then people like Meb and Geb would not exist. As it stands, they were able to run at or near their personal bests far beyond age 35. If you keep training hard, you should be able to hold pretty close to or even at your lifetime bests at most distances until you're almost fifty.

"If you keep training hard, you should be able to hold pretty close to or even at your lifetime bests at most distances until you're almost fifty.[/quote]"

Ummmm.....and the Grammy for Blurry Brain goes to..........

zx.vm.sadfljk wrote:

omitting "outliers" is a good way to blur over the fact that so-called age related decline is largely due to people not training as hard or believing in themselves. if there were a necessary linear decline due to reduced capacity, then people like Meb and Geb would not exist. As it stands, they were able to run at or near their personal bests far beyond age 35. If you keep training hard, you should be able to hold pretty close to or even at your lifetime bests at most distances until you're almost fifty.

At first I wanted to disagree with you. Then I thought harder about what you were saying and wanted to agree with you. Now I am going to take issue with just one part. (I am going to ignore another part I disagree with because it's a can of worms.)

As you suggest, it makes sense for age-related decline to resemble a hockey stick rather than a straight line. The three parameters are the slope of the long shallow decline, the age at the inflection point, and the slope of the short sharp decline.

But I don't think Meb and Geb and other outliers prove that just *anyone* can postpone decline until "almost fifty". These guys are outliers not in age but in their long shallow slope -- virtually flat in these exceptional cases. Most runners though will decline *somewhat* as they age beyond 30 or 35. The question is how much.

The article is taking a whole bunch of hockey sticks of different lengths and angles, tying them together with a string, and finding the best fit line through the mass of wood. It would be much more interesting, and much more difficult, to do longitudinal studies to calculate each runner's hockey stick. If we did that, we might find the age at which decline becomes more rapid to be, on average, around 50. It's a plausible number to me. But at the same time there could still be a big standard deviation in that age, as well as wide variation in the slopes before and after the inflection point.

zx.vm.sadfljk wrote:

omitting "outliers" is a good way to blur over the fact that so-called age related decline is largely due to people not training as hard or believing in themselves. if there were a necessary linear decline due to reduced capacity, then people like Meb and Geb would not exist. As it stands, they were able to run at or near their personal bests far beyond age 35. If you keep training hard, you should be able to hold pretty close to or even at your lifetime bests at most distances until you're almost fifty.

People like Meb clearly show that it is possible to remain fast as we age, but the decline is due to far more than confidence.

What I have come to learn as I've entered my later 30's is that the big challenges with age are (1) you don't improve as quickly as you used to and (2) you don't recover as quickly. For many of us, this means that the speed lost during an injury or gap in running is much harder to make up, even with comparable effort and "confidence". The ability to be fast and the ability to become fast are very different

fisky wrote:

This study has a design flaw. Median performance doesn't reflect age-related decline as much as age-related dedication to training, which is more cultural in nature.

Looking at winners is better, but it also is flawed because outliers like Meb mess up the results.

Look at the graphs. They denote the 90th percentile by hash mark and the individual data points above that. It seems you could draw conclusions that reflect your (more accurate) assessment of the study's design flaws.

Age-related decline is the summation of two compnents 1) actual physical decline and 2) commitment decline.

There is huge flaw any predicating any assessment in distance running based upon age. The reason for this is East African age-changing. I happen to know, for a fact, that Wilson Kipsang is turning 39 this year, and that Eliud Kipchoge is over 40. Geb and Tergat were much older than reported as well. In Kenya, especially in the marathon, average age is often nearer to 30 than 20.

Meb is not an outlier, if anything he is the norm.

The real problem in Western Countries (at least in the USA) is this belief that physical performance "peaks" during your university years. This is a culturally contrived misconception based upon an obsession with agility-based sports, i.e. Football, basketball, baseball, soccer.

In aerobic sports, the development time is much longer, and science has yet to prove where a human beings aerobic ceiling stops.

I think one of the principle reasons as to why East Africans dominate the distance events is due to a socio-economic phenomenon that makes it possible for men and women to train at a high level well into their 30's. The reasons for this are too many to list on a letsrun post, but let's just say economic reasons are key (Sustainable agriculture, lack of job opportunity, social acceptance of "running" as a profession, etc).

We in the west need to stop buying into this fantasy of freak-genetic teenagers running world record times and accept that there is a system over in East Africa that allows talented athletes to mature far beyond our own.

Stop the ethnocentric arrogance, wise up!

That's just a dumb study. Probably cost millions for that information.

Listen, a 20 year old person can run faster than a 60 year old person... common knowledge.