Paul Tergat agrees with LetsRun - Sub-2 marathon is impossible

I think it's likely that competitive marathons will be around long enough for progression to take place down to 2 hrs. In most discussions it's taken as a given that marathoning will continue to be a financially rewarding activity for elites, but I doubt it. Especially when you start talking hundreds of years in the future.

Plus, I just don't think the human body is capable of a sub-2. I know all the familiar arguments against that, but there has to be a limit somewhere. It seems obvious that no human resembling the humans of today will ever run a 10-minute marathon, because that would exceed the physical abilities of the human body. Right? Well, if you accept that, then you accept that there ARE limits. It then just becomes a matter of estimating what those limits are.

Octa wrote:

I am weary of what a young super elite runner will do.

Why would wondering about what a young runner will do make you tired?

To get to sub 2, based on rough analysis, Kipsang would need to either increase his step length by 2", increase his stride rate by 5 bpm, or increase step length by 1" while increasing step rate by 2.5 bpm. Looking at it like that, (especially since he's in the low 180's for step rate), a sub 2 marathon is not impossible - even for Kipsang.

for the record ,

cant compare record with a doped up bolt

without igf-1 lr3 his tall legs would never

be able to move as fast .

how do you know that kipsang is not on same stuff ,

as definitely in kenya at moment starting with rudisha

then kiprup ,a certain javelin thrower and who knows next .

the boston times were due to gw50 ,

and no progression in mile for obvious reason .

bekele in prime training continuosly with aid of epo and maximum adaption to training it entails.

then crazy hct for race , gw50 , alittle testosterone which he always used ,some drafting and berlin course ,

have sub 2.01 already .

get your longer legged athlete similar to bekele and enhanced nervous system from igf-1 lr3 and is close to been

done.

sub 2 will happen within 10 years if hearing that new

better endurance drugs in the making .

not including doping here which is how it will be done .

Nothing is impossible unless you believe it so. The Sub - 2 will be done someday by someone. There may be a little boy in Kenya that will do it in 30 years. Man could not break the sound barrier or climb Mt. Everest until someone did it. A sub-4 mile was considered impossible in 1950. But after Bannister, it became commonplace. The sub-2 is a psychological barrier as much as it is a physical barrier. I just hope I live to see it.

There is no reason to say that we are at the human limit right now in the marathon. Anyone who says that might guess correctly. However, if records are broken on a regular basis (once every two to three years), we are very likely no where near the human limit. I would propose that only if a record is stale for 50 years do we have some minimal reason to say that the record is the human limit.

Some of the women's records, on the other hand, are getting near the point at which we can say that they are beyond the human limit by any non-doped women. Women competing at the elite level today are further from the distance records than any men or women of any era in distance running were to the records of their era. Sucks for women.

I'm weary of reading idiotic logic on both sides of this idiotic debate.

I run in college wrote:

I guess what I meant was not to ignore the science of physiology, but to ignore anyone who cites physiology to make sweeping definitive statements like 2:00 is not possible, or 4:00 is not possible. There are so many variables that it is not possible to account for everything. I did not mean to come across as anti-science, just to say never blindly accept any "scientific" statement that is not backed up by irrefutable proof.

And to those who are questioning regressions. I used decaying functions to run regressions over a variety of time intervals to account for the fact that the faster you get, the harder it is to drop time. It is imperfect, but the best that I have.

I don't trust regressions for the purpose of calculating future records. The progression can speed up as well as slow down for a number of reasons. A larger number of runners chasing the same time, will lead to better pacing for the winner. Shoe technology needs a technological leap because it has relied on EVA foam for the past 40 years, and this is just for easier manufacturing and shorter life span so we spend more money on shoes.

There is still room for adaption in training, using old fashioned methods in new ways(there are no new training methods in the last 60 years in my opinion) to develop more speed endurance. Lots of things I have missed out, which other people can remind me of?

SB wrote:

What if this discussion were about the sub-1:45 marathon? There are either limits or there are not (and, clearly, there must be). If there are limits, then it is reasonable to suggest that they reside this side of 2:00 (more than reasonable, when real, substantive factors-- i.e. rather than simple numbers-- are introduced).

No, absolutely not. The limits lie far more towards 1.45 than this side of two hours. The problem runners have in visualizing this is their own insecurity in recognizing their own insignificance when comparing their own performance to that of others.

This.

jono wrote:

I don't trust regressions for the purpose of calculating future records. The progression can speed up as well as slow down for a number of reasons. A larger number of runners chasing the same time, will lead to better pacing for the winner. Shoe technology needs a technological leap because it has relied on EVA foam for the past 40 years, and this is just for easier manufacturing and shorter life span so we spend more money on shoes.

There is still room for adaption in training, using old fashioned methods in new ways(there are no new training methods in the last 60 years in my opinion) to develop more speed endurance. Lots of things I have missed out, which other people can remind me of?

I would argue the reverse. What a regression shows is the aggregate of effects. Thinking about this thread has gotten me curious and looking into WR projections in other events as well. In several events with enough data (some events only have 8 or so WRs in the progression) There are startlingly accurate regressions. I have seen trends modeled with R^2 of .98 - strong enough that I would be willing to publish it if it came from my lab work.

And yes, there are some exception. The best example is Usain Bolt. There was a trend to the 100m record before he arrived, but now there is no logical regression that could predict the future. So far, there are no such exceptions in the marathon.

rekrunner wrote:

That's not much.

i remember when wrote:

Not too long after his 2:04:55, Tergat also said that he didn't think it would be possible for anyone to improve on that time too much. He said he put so much into training for that and completely expended himself doing it that he didn't think it would be possible for anyone to run much faster. Of course now, three times since someone has run a full minute faster.

Actually rekrunner, the record has come down by 10 seconds per year. Where have you been? Gibraltar?

Surely a regression is only as good as the inputed data?

But have you noticed that on those special runs where you seem to be floating effortlessly and breathing easily, that there is a a part of your brain suggesting that the faster you run, the faster you can run?

As I've said on the Future World Records without drugs thread, it's not about extra oxygen uptake, but about better energy return, which means better running economy. And this is where the improvement in World Record progression has come from and will continue to do so.

What I really want is for scientists in different fields to talk to each other, because PhD's tend to be mostly tunnel visionary, believing their own research to be the most important field of study.

As a result, ex phys is a very poorly taught subject.

Yes, the regression is only as good as the data, but there is nothing saying that the data we have is inaccurate. The fact that we are having this discussion suggests that the current world record is very, very good for our time. This suggests that it is a good data point for the date it was set. I am too new to the sport to comment on what people thought of the world records 20, 40, 60 years ago, but I would imagine that those WRs were considered extremely good performances then.

I have nothing to say about what is driving the trend. As you say, exercise physiology is a poor (and infrequently) taught subject. All I can say is that the trends are there and that there is no sign of deviating from them.

Also, one thing I meant to mention earlier: Some of the more statistically accurate models for the marathon I came up with push the 2:00 barrier out closer to 2065.

The recent improvement of 10 seconds per year is the most statistically valid one to me currently. I see no reason not to suppose a similar progreesion, suggesting sub two in little more than 20 years?

On this topic I'd say: Say never "never".

And why do some people think, you'd have to have this crazy 5 K/10 K speed?

A TRUE marathoner could do it with a 5/10 K potential of 12:40/26:20 I think.

12:40 > 26:20 > 57:30 > 1:59:59 isn't completely out of question some day for some freakish endurance talent.

It's a question of talent, training and conditions. I really think, it WILL happen. It might be decades down the roas. But it WILL happen IMHO.

Pikachu

WR from 2:08 back in 1984 to 2:03 a little less than 30 years later.

If the human society doesn't get largely destroyed any time soon, I think we will see sub 2:00 by 2040

the 10s/year is a tight fit, but there is validity in knowing that it gets harder to lose time as the time gets faster, which is why I have been trying to fit curves. It is a very close fitting regression, but not quite as tight as the ones that predict early 2060s. It also uses less data to obtain the result. I think that the linear 10s/year will be a good model for the next couple WRs, but probably not all the way out to 2:00.

Michael Johnson didn't believe that bolt could break his record at the olympics and he watched in disbelief. nothing is impossible