Why are mainstream marathon runners slower today?

academictech wrote:

Close but believe me, I know exactly what I wrote and stand by it. All things in nature follow a Gaussian curve or Normal Distribution.

That is to say, if we took all marathon runner's time in every marathon over the past 25 years and compared them year over year, there would be an equal distribution of the times.

If performance was static (i.e times did not improve) you would be right; there would be less 2:30 guys both in terms of absolute number and percentage of marathoners.

However, one variable has changed....we are recording faster times, thus the average time taken to complete the distance will have improved.

While there are more marathon runners today, the mean percentage of sub 2:30 runners must be greater.

You don't need to go and count them, just do the math.

I am astonished at the level of ignorance -- both mathematical and empirical -- exhibited by these statements.

ouch

Well, what did you expect with a statement like "All things in nature follow a Gaussian curve or Normal Distribution"?

Just because natural phenomena follow a Gaussian distribution doesn't mean that all phenomena do. Human processes might be skewed to some other distribution. Here's one for you: the spectral density of noise in electronic circuits does not follow a Gaussian curve with respect to frequency. In fact, it tends to rise at lower frequencies. Here's another: Failure rates of electronic components exhibit a "bathtub" curve with respect to time. There's lots of examples...

I doubt there is anything natural about the spectral density of noise in an electronic circuit.

What don't you understand????????

A few guys running 2:06/07 compared to the tops guys running 2:09/10 isn't going to make much difference to average marathon times. But when you get loads of guys running outside 3 hours that used to run inside 3 hours that is going to make a hige difference - and the difference is that the average finishing time in marathons is getting slower NOT faster. Just because the facts dont fit your stupid model don't mean they ain't right. The statistics don't lie. WE ARE GETTING SLOWER.

academictech wrote:

I doubt there is anything natural about the spectral density of noise in an electronic circuit.

Two comments:

1. You apparently don't know much about electronic circuits then. Ever heard of Boltzmann's Constant?

2. Are you saying that people competing in marathons is somehow a "natural event" and not a contrived human event?

Many things do follow normal distributions. It is folly, however, to assume that all metrics of all things (natural or contrived) must follow normal distributions. Does the population density of the USA follow a normal distribution based on location? If it did, wouldn't there be an awful lot of people in Nebraska and nobody on the coasts? Does the spatial distribution of stars in a galaxy follow a normal distribution around its center? If they did, wouldn't all galaxies have to be globular clusters (instead of so many being barred spirals like our own Milky Way)?

Oops, not a "barred" spiral (although they're certainly out there).

academictech wrote:

Close but believe me, I know exactly what I wrote and stand by it. All things in nature follow a Gaussian curve or Normal Distribution.

You don't need to go and count them, just do the math.

Tell that to the geniuses at Long-Term Capital Management.

LOL

academictech wrote:

Close but believe me, I know exactly what I wrote and stand by it. All things in nature follow a Gaussian curve or Normal Distribution.

That is to say, if we took all marathon runner's time in every marathon over the past 25 years and compared them year over year, there would be an equal distribution of the times.

If performance was static (i.e times did not improve) you would be right; there would be less 2:30 guys both in terms of absolute number and percentage of marathoners.

However, one variable has changed....we are recording faster times, thus the average time taken to complete the distance will have improved.

While there are more marathon runners today, the mean percentage of sub 2:30 runners must be greater.

You don't need to go and count them, just do the math.

uh, i think you are the one who needs to do the math.

just because the outlier performances are higher doesn't mean that the overall mean increases as well (i.e., number of 2:30 guys doesn't necessarily go up).

if you were just looking at the sub-category of elite marathon runners, the fact that the mean has improved over time (including new drugs, etc) would imply better outlier performances, assuming the variance stayed the same. not the other way around.

"I am astonished at the level of ignorance -- both mathematical and empirical -- exhibited by these statements."

oldguy: Congratulations, the above statement is easily the most intelligent and well crafted putdown I've seen on these boards. Keep up the fine work....