NOW do you not think a sub-2:00:00 marathon will happen?

I think if the Brojos offered a LRC t-shirt, it could happen in Chicago next month.

DavidSHM wrote:

p.s You can offer someone 100 million dollars, but that type of enticement is a moot offer considering the task at hand. It's not going to happen for a very, very long time.

The "demographic" shift started after Geb, in the last 6 years or so.

Just the same, this subject seems decades premature. Not too long ago, we broke the 2:04 barrier, and some years before that, the 2:05 barrier. Now that we are at 2:03, it remains to break 2:02, then 2:01. Then the subject becomes ripe for discussion.

If we assume some linear progression, my simple math says 16.5 years, or in the spring of 2031. This is some kind of optimistic upper bound. If it happens sooner, it will come from a healthy male adult who weighs 103 pounds, or 48 kilos.

But, there has been a recent change in the athlete profile, transitioning from the aged, coming into retirement track athlete, to the younger, in their prime, athlete. It's not hard to imagine that Geb, in his prime, would have run 1 minute faster.

Furthermore, at some point we will start to approach human limits, where linear progressions no longer hold. It's hard to imagine that we are not yet approaching that limit.

These factors suggest that 16.5 years might be optimistic, and I hope I live long enough to see that it might happen in 50 years.

As pointed out by several people on this board already, I think it is illogical to assume a linear regression for the world record. It would be wiser to assume some sort of exponential or logarithmic function, with smaller chunks of time being taken off until 1:59:59 is reached.

Someone on twitter posted a pretty good prediction based on a logarithmic trend line:

https://twitter.com/isthatsol/status/516663935949611008

always been that way wrote:

I see sub-2:00 marathon eventually. The bonus for a WR means it will go down a few seconds at a time. There's still over 120 seconds left so it could be over 60 years (assuming 1-3 seconds improvement a year) to get under 2 hours..

Agree with that.

The marathon is not the kind of race you can just take a few seconds off here and there. By the time you see the clock you might be 30 seconds ahead of schedule and at that point you can't just stop.

Not to mention, the marathon is a race. If you try to run an exact time then you might get into a final 3k-5k "sprint" finish that will end up making your time 20 seconds faster anyways.

Its going to take some sort of genetic outlier human like Paul Radcliff or Usain Bolt. A freak of nature.

Here's another way to look at it. What is the distance of the "equivalent" 2:02:57 marathon performance, at 2:00:00 marathon pace?

We can calculate by 400m laps:

2:02:57 for the marathon is 69.93s/400m for 105.4875 laps

2:00:00 for the marathon is 68.25s/400m for 105.4875 laps

Using "Purdy", a 2:02:57 marathon is worth 1128.16 Purdy points.

A 1:26:27 performance for 30400m (76 laps) is worth 1128.25 Purdy points, at the desired pace of 68.25s/400m lap.

In other words, Dennis Kimetto, today, if equally trained for ~30K distance, can only maintain a 2:00:00 marathon pace for 1:26:27, or 30.4K. Then he has to stop, stoop over hands on knees, spit, think about not vomiting, gather the strength to hold it back, grab a towel, wipe his face, grab a Kenyan flag, and smile for the camera, with thumbs up.

Although 30K and 25K records tend to be weak, no one has yet run these shorter distances faster than this 2:00:00 marathon pace:

- In June 2011, Moses Mosop smashed the 30K world record on a track, in 1:26:47.4 (merely 69.43s/400m). The equivalent "2:00:00 (road) marathon pace" for 30K would be 1:25:37.5. In his performance on the track, he would have been lapped, and lost another 20 seconds, to this "equivalent" road runner.

- Even Dennis Kimetto's Berlin May 2012 25K performance is "only" 1:11:18, or merely 68.45s/400m. The equivalent 2:00:00 marathon pace is 1:11:05.6.

It seems to me, no matter how you slice it, there are too many barriers to entertaining any kind of discussion about the 2:00:00 marathon. We should let this thread die, and resurrect it after another 11 years, only if we break the 2:01 barrier.

rekrunner wrote:

But, there has been a recent change in the athlete profile, transitioning from the aged, coming into retirement track athlete, to the younger, in their prime, athlete.

Someone like this:

Not sure it's wiser to curve fit using data starting in 1907. A large part of curve fitting is the data set you pick. Stop at 1948, and you see that reality departed greatly from the theory over the subsequent 20 years, requiring a new theory that seems to show stagnation (understandably during the war), followed by a period of "making up for lost time". Seems to me that the data since 1980 is more coherent with modern trends, but then we have a lot fewer points to make a decent curve fit. I would also need the rationale for choosing a logarithmic model over a function with an eventual asymptote representing a human limit.Just the same, my gut agrees that 2046 is more likely than 2030 (but that 2065 is even more probable).

NotLinear wrote:

As pointed out by several people on this board already, I think it is illogical to assume a linear regression for the world record. It would be wiser to assume some sort of exponential or logarithmic function, with smaller chunks of time being taken off until 1:59:59 is reached.

Someone on twitter posted a pretty good prediction based on a logarithmic trend line:

https://twitter.com/isthatsol/status/516663935949611008

rupp-certified saladbar wrote:

2003 Berlin: 2:04:55

2014 Berlin: 2:02:57

Simple arithmetic could let you to believe 1:59:59 will happen in 15 years or so.

But then, the last decade has seen a sea change in the marathon talent pool, with runners "moving to" the distance from an early age, even their teen years. (Causing the inverse sea change in the now-stagnant 10,000.) We probably won't see a minute coming off the WR every five years once we settle into this "new normal."

But we WILL see sub-2. Before 2040.

You letsrun geniuses do realise that taking a minute off the WR gets progressively harder

A question for those who say that sub 2 will never be run. If sub 2 is not possible, then what time is the lowest limit that can or will be run? It is now not 2:02:57, so it has to be somewhere between 1:59:59 and the 2:02:57. If you say, for example, 1:01:18, how did you arrive at that time?

in the year 2020

Dingler wrote:

I don't think anyone has split sub-61:00 yet (first or second half). Yesterday Kimetto's second half was like 61:15. You guys are talking about sub-60:00 back-to-back. I don't see anyone doing that until they are running close to 57:00-flat for a half. It will happen one day but we are still a long, long way off from that. I don't really see anyone running now (including Bekele) going much faster than 2:02-flat.

Well put. Let's start talking about someone breaking 2:00:00 for the half when someone runs a sub 1:00:00 for a half marathon in a marathon.

Has 10k in a marathon ever been run at 2:00:00 pace? I don't think but could be wrong.

Barring some radical change in training we're a long way away.

Crazy to think Renaldo DaCosta took the record to 2:06:05 in 1998. So since then we're halfway to 2:00:00.

wejo wrote:

Dingler wrote:

I don't think anyone has split sub-61:00 yet (first or second half). Yesterday Kimetto's second half was like 61:15. You guys are talking about sub-60:00 back-to-back. I don't see anyone doing that until they are running close to 57:00-flat for a half. It will happen one day but we are still a long, long way off from that. I don't really see anyone running now (including Bekele) going much faster than 2:02-flat.

Well put. Let's start talking about someone breaking 2:00:00 for the half when someone runs a sub 1:00:00 for a half marathon in a marathon.

Has 10k in a marathon ever been run at 2:00:00 pace? I don't think but could be wrong.

Barring some radical change in training we're a long way away.

Crazy to think Renaldo DaCosta took the record to 2:06:05 in 1998. So since then we're halfway to 2:00:00.

Kimettos 10k split of 28:39 between 25-35k is the fastest ever 10k split in a marathon. It's just over 1 second per kilometer slower than sub-2 pace. Sub-2 pace is 28:26 per 10k.

Wow. Seeing that makes me feel good. Supports what I just said yesterday on a run. That the past decade has been people getting the marathon where it should be.

wejo wrote:

Dingler wrote:

I don't think anyone has split sub-61:00 yet (first or second half). Yesterday Kimetto's second half was like 61:15. You guys are talking about sub-60:00 back-to-back. I don't see anyone doing that until they are running close to 57:00-flat for a half. It will happen one day but we are still a long, long way off from that. I don't really see anyone running now (including Bekele) going much faster than 2:02-flat.

Well put. Let's start talking about someone breaking 2:00:00 for the half when someone runs a sub 1:00:00 for a half marathon in a marathon.

Has 10k in a marathon ever been run at 2:00:00 pace? I don't think but could be wrong.

Barring some radical change in training we're a long way away.

Crazy to think Renaldo DaCosta took the record to 2:06:05 in 1998. So since then we're halfway to 2:00:00.

This was already mentioned, several times.

The Tergat WR was 2:04:55 and was 11 yrs ago. This is 1:58 faster than the Tergat WR. To go 2:58 faster (sub-2:00:00) in a certified marathon, clean, I think will take some evolution that will take more than 15 yrs. Prob 100 years or 200.

If Derek Clayton or Frank Shorter or Carlos Lopes had the same drugs and the same doped rabbits and the $300-500,000 paydays and the flat-fast courses that we have now ... they would have run 2:02-2:04 just like these guys are running. Keep in mind Lopes ran 2:07:12 twenty-nine yrs ago. This WR is "only" 4:14 faster.

What are they going to do with current physiology to go 2:58 faster? The ability of man will have to improve beyond what is possible right now. The sub-2 hr marathoner has not been born yet ... in my opinion.

There were some fast courses for popular marathons before the EPO era (Rotterdam, London, Chicago) but they were not as ultra-fast as the best ones today.

I don't think the 2:07-2:10 marathoners from the 60's, 70's or 80's were clean either, they just did not have what they do in 2014 and what has been being used (in combination) for the last 12 yrs.

posted in Kimettoooooooooo!

In fact, a Sub 2 marathon has already been run if using the original distance from Marathon to Athens; it was about 40km or 25 miles. How many have already broken 2?anyone who has broken 2:06 for 26.2 has broken 2:00 for 40km. My point is that we are using an abstract standard set in the 1920s and equating it to a set time of 2 hours! illogical?

Let's congratulate Kimetto for his world record and not be petty about a mythical barrier (you can thank the English for the extra 385 yards above 26 miles!)

wejo wrote:

Barring some radical change in training we're a long way away.

But it may not be just a change in training. I was originally thinking roughly 40 years away at the current rate of improvement, until someone earlier in the thread said this:

That changes the physiological landscape. Sort of reminiscent of the doping of athletic minors by the east german state back in the 70's, no?

kipisfast wrote:

How close to optimal is the Berlin course? I know it is considered extremely fast, but how about a dead flat circular course of circumference 42,195 meters?

I think you're on to something, let's petition CERN to host the Large Hadron Collider Marathon 2015!

OK, so they'd have to run a lap and a half, big deal.

The corridor is about wide enough for two people to run abreast if they're careful, or for a single file line with enough room to trade off the lead occasionally.

And they do shut the thing down pretty often for upgrades and what not, shouldn't be too hard to squeeze in a few days for preparation and to hold the race.