A beautiful graph...

running-forum.com
Promoting running discussion.

Main

Date: 25 Oct 2006 00:28:27
From: anders
Subject: A beautiful graph...

...illustrating a curious phenomenon in marathon finish times:

http://www.schottner.de/ma2006_herren_verteilung.gif

Anders

Date: 25 Oct 2006 13:36:31
From: steve common
Subject: Re: A beautiful graph...

"anders" <hop.allez@suomi24.fi > wrote:

>
>...illustrating a curious phenomenon in marathon finish times:

Indeed. Those spikes just before 3h, 3h15, 4h etc and the troughs just
after (look at 3:45 and the Marianas Trench just after 4h)

I bet if 2:48, for example, were defined as a "real runners start here",
you'd get a major peak just under 2:48 followed by a trough leading to 3h.

Date: 25 Oct 2006 12:12:15
From: Elflord
Subject: Re: A beautiful graph...

On 2006-10-25, steve common <stevenZ.common@wanadoo.fr > wrote:
> "anders" <hop.allez@suomi24.fi> wrote:
>
>>
>>...illustrating a curious phenomenon in marathon finish times:
>
> Indeed. Those spikes just before 3h, 3h15, 4h etc and the troughs just
> after (look at 3:45 and the Marianas Trench just after 4h)
>
> I bet if 2:48, for example, were defined as a "real runners start here",
> you'd get a major peak just under 2:48 followed by a trough leading to 3h.

Would be interesting to look at female finishers in Chicago ...

Cheers,
--
Elflord

Date: 25 Oct 2006 07:28:28
From: Todd
Subject: Re: A beautiful graph...

"Elflord" <abuse@aol.com > wrote in message
news:slrnejul4v.o78.abuse@panix2.panix.com...
> On 2006-10-25, steve common <stevenZ.common@wanadoo.fr> wrote:
> >
> > I bet if 2:48, for example, were defined as a "real runners start here",
> > you'd get a major peak just under 2:48 followed by a trough leading to
3h.
>
> Would be interesting to look at female finishers in Chicago ...

It would be interesting to look at female runners in Chicago regardless.
>
> Cheers,
> --
> Elflord

Date: 25 Oct 2006 20:14:01
From: steve common
Subject: Re: A beautiful graph...

"Todd" <Todd33@hotmail.com > wrote:

>It would be interesting to look at female runners in Chicago regardless.

Ya know I noticed Donovan didn't mention tight glutes anywhere. I think he
should train with Charlie (or me) more often :oP

Date: 25 Oct 2006 20:15:27
From: steve common
Subject: Re: A beautiful graph...

Elflord <abuse@aol.com > wrote:

>Would be interesting to look at female finishers in Chicago ...

I started by looking at first two males and females. The men positive
splitted (OMG! sorry) and the ladies "evened". Didn't analyse any further.

Date: 26 Oct 2006 14:19:02
From: Steve K
Subject: Re: A beautiful graph...

Elflord <abuse@aol.com > wrote:

> On 2006-10-25, steve common <stevenZ.common@wanadoo.fr> wrote:
> > "anders" <hop.allez@suomi24.fi> wrote:
> >
> >>
> >>...illustrating a curious phenomenon in marathon finish times:
> >
> > Indeed. Those spikes just before 3h, 3h15, 4h etc and the troughs just
> > after (look at 3:45 and the Marianas Trench just after 4h)
> >
> > I bet if 2:48, for example, were defined as a "real runners start here",
> > you'd get a major peak just under 2:48 followed by a trough leading to 3h.
>
> Would be interesting to look at female finishers in Chicago ...
>
> Cheers,

Some graphs for male/female and a few male age ranges from the Chicago
preliminary results: http://tinyurl.com/sju3g

Date: 25 Oct 2006 13:26:26
From: Anthony
Subject: Re: A beautiful graph...

Date: 25 Oct 2006 02:13:20
From: PiledHIgher
Subject: Re: A beautiful graph...

Elflord wrote:
> On 2006-10-25, anders <hop.allez@suomi24.fi> wrote:
> >
> > ...illustrating a curious phenomenon in marathon finish times:
> >
> > http://www.schottner.de/ma2006_herren_verteilung.gif
>
> Interesting. Notice that the right tail is kind of fat, as one might expect.
>
> What I'd be interested in is, what would the PDF look like for an individual
> in some time range ? Or to put it another way, if one were to plot the same
> graph for a given age/gender/HM PR bucket, what would the picture look like ?
>
> Cheers,
> --
> Elflord

I assume the point was the deviation in the curve befor each of the 1/2
'barriers' ie. 3hrs, 3hrs 30 etc.

Date: 25 Oct 2006 12:10:59
From: Elflord
Subject: Re: A beautiful graph...

On 2006-10-25, PiledHIgher <PiledHigher@gmail.com > wrote:
>
> Elflord wrote:
>> On 2006-10-25, anders <hop.allez@suomi24.fi> wrote:
>> >
>> > ...illustrating a curious phenomenon in marathon finish times:
>> >
>> > http://www.schottner.de/ma2006_herren_verteilung.gif
>>
>> Interesting. Notice that the right tail is kind of fat, as one might expect.
>>
>> What I'd be interested in is, what would the PDF look like for an individual
>> in some time range ? Or to put it another way, if one were to plot the same
>> graph for a given age/gender/HM PR bucket, what would the picture look like ?
>>
>> Cheers,
>> --
>> Elflord
>
> I assume the point was the deviation in the curve befor each of the 1/2
> 'barriers' ie. 3hrs, 3hrs 30 etc.

Good call, I was asleep when I posted that followup. Interesting that there's
no such notch once below the 3hr mark too ... (but several at 10 minute boundaries
in the 3-4hr range)

Cheers,
--
Elflord

Date: 25 Oct 2006 08:53:57
From: Elflord
Subject: Re: A beautiful graph...

On 2006-10-25, anders <hop.allez@suomi24.fi > wrote:
>
> ...illustrating a curious phenomenon in marathon finish times:
>
> http://www.schottner.de/ma2006_herren_verteilung.gif

Interesting. Notice that the right tail is kind of fat, as one might expect.

What I'd be interested in is, what would the PDF look like for an individual
in some time range ? Or to put it another way, if one were to plot the same
graph for a given age/gender/HM PR bucket, what would the picture look like ?

Cheers,
--
Elflord

Date: 25 Oct 2006 19:10:04
From: PiledHIgher
Subject: Re: A beautiful graph...

steve common wrote:
> "Tony S." <email_tonys@yahoo.com> wrote:
>
> >but I also think it indicates that many of
> >them could go faster even, since it appears relatively easy for many to move
> >to one side of those time barriers.
>
> 100% agreement. Particularly as the marathon distance is totally
> arbitrary. It was 40k (and IMHO should've stayed so - I love the Queen and
> all her family but WTF, 42 1950 meters? What is this?)
>
> It had to be a bunch of Lords (as drunk as a) to be so dickwitted as to
> accept 26 miles and WTF yards, the first time, instead of some handy
> approximation of 40k.
>
> Makes me hate being a Brit sometimes, when you think...
>
> > The relative flatness from 4 to 4:30
>
> are we looking at the same graph? Or am I misunderstanding "flatness"
> (probable cause :-) ?
>
> >represents the "jogger" group. Beyond that, it's hard to run that slowly,
>
> Either you're trolling or you are projecting. I can do a sub 3h marathon
> but 80% of my run time is currently done at <11kph => about a 4h marathon.
> And I can run a LOT slower. Some of my "jog components" will come out at
> 6:00/k or slower = 9'40/k = 4:13 marathon.
>
> By concentrating, or by getting into a fellow runner's groove, I can run
> 6:30/k's = 4:43 marathon, with no difficulty whatever.

I think that the 'flatness between 4 and 4:30 is a funtion of the push
to get just ubder 4, 4:15 and 4:30 respectively.

Interesting to note that even 5 hour pace there is still a clear push
to beat the time barrier.

Date: 26 Oct 2006 00:41:42
From: Tony S.
Subject: Re: A beautiful graph...

"anders" <hop.allez@suomi24.fi > wrote in message
news:1161761307.755613.177690@h48g2000cwc.googlegroups.com...
>
> ...illustrating a curious phenomenon in marathon finish times:
>
> http://www.schottner.de/ma2006_herren_verteilung.gif
>
>
> Anders

Yes, most interesting. The smaller spike at sub-3 indicates it's a much
harder achievement. But the pronounced spikes at sub-3:30 and especially
sub-4:00 indicate that those are psychological goals for many runners who
otherwise would end up slower, but I also think it indicates that many of
them could go faster even, since it appears relatively easy for many to move
to one side of those time barriers. The relative flatness from 4 to 4:30
represents the "jogger" group. Beyond that, it's hard to run that slowly,
and more walking is being mixed in the further you go out.

-Tony

Date: 26 Oct 2006 03:06:04
From: steve common
Subject: Re: A beautiful graph...

"Tony S." <email_tonys@yahoo.com > wrote:

>but I also think it indicates that many of
>them could go faster even, since it appears relatively easy for many to move
>to one side of those time barriers.

100% agreement. Particularly as the marathon distance is totally
arbitrary. It was 40k (and IMHO should've stayed so - I love the Queen and
all her family but WTF, 42 1950 meters? What is this?)

It had to be a bunch of Lords (as drunk as a) to be so dickwitted as to
accept 26 miles and WTF yards, the first time, instead of some handy
approximation of 40k.

Makes me hate being a Brit sometimes, when you think...

> The relative flatness from 4 to 4:30

are we looking at the same graph? Or am I misunderstanding "flatness"
(probable cause :-) ?

>represents the "jogger" group. Beyond that, it's hard to run that slowly,

Either you're trolling or you are projecting. I can do a sub 3h marathon
but 80% of my run time is currently done at <11kph = > about a 4h marathon.
And I can run a LOT slower. Some of my "jog components" will come out at
6:00/k or slower = 9'40/k = 4:13 marathon.

By concentrating, or by getting into a fellow runner's groove, I can run
6:30/k's = 4:43 marathon, with no difficulty whatever.

Date: 26 Oct 2006 13:49:43
From: Tony S.
Subject: Re: A beautiful graph...

"steve common" <steven.common@wanadoo.fr > wrote in message
news:ab10k2p9dvb0fno7qtosg4ihgnt5p4t2to@4ax.com...
> "Tony S." <email_tonys@yahoo.com> wrote:
>
> >but I also think it indicates that many of
> >them could go faster even, since it appears relatively easy for many to
move
> >to one side of those time barriers.
>
> 100% agreement. Particularly as the marathon distance is totally
> arbitrary. It was 40k (and IMHO should've stayed so - I love the Queen and
> all her family but WTF, 42 1950 meters? What is this?)
>
> It had to be a bunch of Lords (as drunk as a) to be so dickwitted as to
> accept 26 miles and WTF yards, the first time, instead of some handy
> approximation of 40k.
>
> Makes me hate being a Brit sometimes, when you think...

> > The relative flatness from 4 to 4:30
>
> are we looking at the same graph? Or am I misunderstanding "flatness"
> (probable cause :-) ?
> >represents the "jogger" group. Beyond that, it's hard to run that slowly,
>
> Either you're trolling or you are projecting. I can do a sub 3h marathon
> but 80% of my run time is currently done at <11kph => about a 4h marathon.
> And I can run a LOT slower. Some of my "jog components" will come out at
> 6:00/k or slower = 9'40/k = 4:13 marathon.

That's true, one can jog a lot slower.

> By concentrating, or by getting into a fellow runner's groove, I can run
> 6:30/k's = 4:43 marathon, with no difficulty whatever.

As could I, yet that's just beyond 4:30. I'm just suggesting that most
people who train would have a natural slow pace somewhere in the 9-11mpm
range. But that's projecting, as you said, and 11mpm would put one closer to
5 hours.

-Tony

Date: 30 Oct 2006 16:08:25
From: Robert Grumbine
Subject: Re: A beautiful graph...

In article <avT%g.8146$hK.914@trnddc02 >, Tony S. <email_tonys@yahoo.com> wrote:
>"anders" <hop.allez@suomi24.fi> wrote in message
>news:1161761307.755613.177690@h48g2000cwc.googlegroups.com...
>>
>> ...illustrating a curious phenomenon in marathon finish times:
>>
>> http://www.schottner.de/ma2006_herren_verteilung.gif
>
>Yes, most interesting. The smaller spike at sub-3 indicates it's a much
>harder achievement.

But is it smaller? The spike is from about 200 runners at
just over 3:00 to about 320 runners at 2:58. spike = 60% of the
'just over'

Around 3:30, the spike is from 650 to 960, adding 48% of the
'just over. So, if anything, the spike is _more_ pronounced at
3:00 than 3:30.

What's going on from 3:40 to 4:35 ... well, first show me what
the 'smooth' curve would be, and why it should be that.

I think, as I look at the 3:00 and 3:30 spikes (and 3:15 and
3:45, though lesser degree) that it's wrong to think of it
as runners pushing to get under the bar, so depleting the slightly
slower group and enhancing the slightly faster. As I eyeball the
curve, there seems to be a fair smooth curve that can be drawn
from 2:30 through 3:40. The spikes are, very largely, _additions_
to the smooth curve.

i.e., people who are pursuing these magic numbers came out to this
marathon for the purpose of getting them. It isn't a matter of
people who would be running the marathon anyhow pushing to
get under that limit (in which case the slightly slower times
should be severely depleted relative to the background).

For the matter of nudging under a 'bar', the 5:00 population looks
like it might be a candidate. Or maybe not even that. From
4:00 to 5:00 there's one more or less smooth curve. From 5:00 to
6:00 there's another. But it's discontinuous across 5:00, whether
from people pushing across the faster side (and by large margins,
down to 4:00) or by reduced participation by over-5 hour marathoners.

The 3:40-4:30 realm, is actually no surprise for not having
a well-defined continuum curve if the results are for all finishers
(which seems likely the case given the numbers of finishers). This
is the realm of including mid-pack open class men (who might be
engaging in the 'taking it too easy', if one could define what
'too easy' was) to competitive age group runners (certainly not taking
it 'too easy')

--
Robert Grumbine http://www.radix.net/~bobg/ Science faqs and amateur activities notes and links.
Sagredo (Galileo Galilei) "You present these recondite matters with too much
evidence and ease; this great facility makes them less appreciated than they
would be had they been presented in a more abstruse manner." Two New Sciences

Date: 26 Oct 2006 06:56:47
From: tfactor
Subject: Re: A beautiful graph...

Steve K wrote:

> Some graphs for male/female and a few male age ranges from the Chicago
> preliminary results: http://tinyurl.com/sju3g

Nice. It looks like some of the spikes that aren't at obvious round
number times correspond to Boston qualifying times, e.g. 3:10 and :20
for men, 3:40 for women.

Date: 26 Oct 2006 00:53:55
From: lth
Subject: Re: A beautiful graph...

Looks like a typical normal distribution to me.... they are found quite
often in nature, actually, by definition all the time, I think. It
would be more surprising if the graph did not look like a normal
distribution.... Still, it is a beautiful demonstration of the
distribution.

http://en.wikipedia.org/wiki/Normal_distribution

L.

anders wrote:
> ...illustrating a curious phenomenon in marathon finish times:
>
> http://www.schottner.de/ma2006_herren_verteilung.gif
>
>
> Anders

Date: 26 Oct 2006 11:50:19
From: Elflord
Subject: Re: A beautiful graph...

On 2006-10-26, lth <luanhuynh@gmail.com > wrote:
> Looks like a typical normal distribution to me.... they are found quite
> often in nature, actually, by definition all the time, I think. It
> would be more surprising if the graph did not look like a normal
> distribution.... Still, it is a beautiful demonstration of the
> distribution.

No. Besides the already observed phenomena, the right tail is relatively
fat.

I wouldn't expect it to be symmetric.

I would expect the right tail to be fat, especially in adverse conditions.

Cheers,
--
Elflord

Date: 26 Oct 2006 00:19:21
From: anders
Subject: Re: A beautiful graph...

Anthony kirjoitti:

> What is the basis for this graph? Berlin?

The results of all marathons run in Germany this year (up to ober
8th, I believe).

Anders (who wonders whether a graph of female marathoners' times would
show an equally strong "sub-effect")

Date: 27 Oct 2006 15:05:57
From:
Subject: Re: A beautiful graph...

True-- you would expect a positive skew any time you have a variable
with a real lower limit (2:05 or whatever) and no real upper limit.
That's what you are seeing here.

Elflord wrote:
> On 2006-10-26, lth <luanhuynh@gmail.com> wrote:
> > Looks like a typical normal distribution to me.... they are found quite
> > often in nature, actually, by definition all the time, I think. It
> > would be more surprising if the graph did not look like a normal
> > distribution.... Still, it is a beautiful demonstration of the
> > distribution.
>
> No. Besides the already observed phenomena, the right tail is relatively
> fat.
>
> I wouldn't expect it to be symmetric.
>
> I would expect the right tail to be fat, especially in adverse conditions.
>
> Cheers,
> --
> Elflord