Predicting Marthon times from 10K times

(fjwilson) said:

i heard of a formula years ago that one could use to predict marathon times from 10K time
performance. I don't know how accurate it was, but it was quite simple. Can anyone help?
Thanks. fjwilson

There are many predictors out there. Do a yahoo search on running calculators.

Accuracy depends mostly on how similar you will run and train for a 10k vs. a marathon.

Andy

Hi,

A rule said : 10km time X 2.25 = Semi marathon time 10km time X 4.96 = Marathon time

Chris

"fjwilson" <[email hidden]> a écrit dans le message de news:
[email hidden]...

Quoted message said:

i heard of a formula years ago that one could use to predict marathon times from 10K time
performance. I don't know how accurate it was, but it was quite simple. Can anyone help? Thanks.
fjwilson

In article said:

i heard of a formula years ago that one could use to predict marathon times from 10K time
performance. I don't know how accurate it was, but it was quite simple.

There are a bucket load of predictors out there. Keeping it simple: First marathon = 5 * 10k time
Experienced marathon = 4.75 * 10k time

Sam and I once talked about collecting some data to try testing the predictors with,
especially the first marathon vs. experienced marathon. Never quite carried it out, but maybe
one of these years.

--
Robert Grumbine radix.net~bobgOpen ↗ 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

The best way to predict marathon time is the time it takes a person to run 800 meters. If a person
runs an 800 in 3:15 minutes then their marathon time with be 3:15 hours. Runner's World has some
articles at this idea: runnersworld.com0,1300,1 51 54 624,00.htmlOpen ↗

"fjwilson" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

i heard of a formula years ago that one could use to predict marathon times from 10K time
performance. I don't know how accurate it was, but it was quite simple. Can anyone help? Thanks.
fjwilson

Robert Grumbine said:

There are a bucket load of predictors out there. Keeping it simple: First marathon = 5 * 10k
time Experienced marathon = 4.75 * 10k time

I'm so disappointed. After all the math revealed regarding work etc., I would have thought a formula
with Pi, logs, and 7 or 8 unknown variables would have fit. 😉

Quoted message said:

Sam and I once talked about collecting some data to try testing the predictors with, especially
the first marathon vs. experienced marathon. Never quite carried it out, but maybe one of these
years.

Frankly I don't think it would be worth the effort.

--
Caveat Lector "the further you go outside, the further you go inside" - B. McKibben Doug Freese
[email hidden]

My experience: At my first and only marathon, I was running 10km in 43min and 5k in 21min. I ran the
marathon in 4h11. I think I had the capacity to run a bit faster but my objective was to finish it.
Faster, I believe I would have run it in 3h50 ? No better than this. It took me 1,5 months to
recover. I ran for 1,5 years before the marathon, had a bad form and had some problems with my
training. I was still fit enough to do the race. Lots of conclusions to pull from this... In my
mind, the formulas are similar in accuracy than calories reported by Polar watchs, the formula of
max HR by age, etc... And if it was so proportionnal, there would be only one formula.

Regards, Daniel.

"fjwilson" <[email hidden]> a écrit dans le message de news:
[email hidden]...

Quoted message said:

i heard of a formula years ago that one could use to predict marathon times from 10K time
performance. I don't know how accurate it was, but it was quite simple. Can anyone help? Thanks.
fjwilson

Alan Williamson said:

The best way to predict marathon time is the time it takes a person to run 800 meters. If a person
runs an 800 in 3:15 minutes then their marathon time with be 3:15 hours. Runner's World has some
articles at this idea:

Please not Yasso's 800's again. You can get the same accuracy from 3.972 times the height of the
average 22.789 year old aardvark snout.

--
Caveat Lector "the further you go outside, the further you go inside" - B. McKibben Doug Freese
[email hidden]

"Doug Freese" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

Please not Yasso's 800's again. You can get the same accuracy from 3.972 times the height of the
average 22.789 year old aardvark snout.

Without doubt, that's the most accurate statement I've ever seen you post.

Roger.

Doug Freese said:

Please not Yasso's 800's again. You can get the same accuracy from 3.972 times the height of the
average 22.789 year old aardvark snout.

Yikes, Doug's using decimal points 🙂 🙂

Dot

--
"Success is different things to different people" -Bernd Heinrich in Racing the Antelope

"Doug Freese" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

Alan Williamson said:

The best way to predict marathon time is the time it takes a person to

run

Quoted message said:

Quoted message said:

800 meters. If a person runs an 800 in 3:15 minutes then their marathon time with be 3:15 hours.
Runner's World has some articles at this idea:

Please not Yasso's 800's again. You can get the same accuracy from 3.972 times the height of the
average 22.789 year old aardvark snout.

What about 9 times your hand span equals your height? Totally unrelated to the above, but I was
shocked when I found it was 100% accurate.

"ZB" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

"Doug Freese" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

Alan Williamson said:

The best way to predict marathon time is the time it takes a person to

run

Quoted message said:

Quoted message said:

800 meters. If a person runs an 800 in 3:15 minutes then their

marathon

Quoted message said:

Quoted message said:

Quoted message said:

time with be 3:15 hours. Runner's World has some articles at this

idea:

Quoted message said:

Quoted message said:

Please not Yasso's 800's again. You can get the same accuracy from 3.972 times the height of the
average 22.789 year old aardvark snout.

What about 9 times your hand span equals your height? Totally unrelated to the above, but I was
shocked when I found it was 100% accurate.

'fraid not...

me: 8.5" hand span,...* 9 = 76.5 inches, or 6'4.5",...exactly 4" taller than I actually am,...so
this little formula is not only wrong for me, it's not even close.

"Jake Lee" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

"ZB" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

"Doug Freese" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

Alan Williamson wrote:
> The best way to predict marathon time is the time it takes a person

to

Quoted message said:

Quoted message said:

run

Quoted message said:

> 800 meters. If a person runs an 800 in 3:15 minutes then their

marathon

Quoted message said:

Quoted message said:

> time with be 3:15 hours. Runner's World has some articles at this

idea:

Quoted message said:

Quoted message said:

Please not Yasso's 800's again. You can get the same accuracy from 3.972 times the height of
the average 22.789 year old aardvark snout.

What about 9 times your hand span equals your height? Totally unrelated

to

Quoted message said:

Quoted message said:

the above, but I was shocked when I found it was 100% accurate.

'fraid not...

me: 8.5" hand span,...* 9 = 76.5 inches, or 6'4.5",...exactly 4" taller

than

Quoted message said:

I actually am,...so this little formula is not only wrong for me, it's not even close.

Ach you're obviously doing it wrong ;-)

ZB said:

"Jake Lee" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

"ZB" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

"Doug Freese" <[email hidden]> wrote in message "]news:[email hidden]...

>Alan Williamson wrote:
>
>
>>The best way to predict marathon time is the time it takes a person
>>
>>

to

Quoted message said:

Quoted message said:

run

>>800 meters. If a person runs an 800 in 3:15 minutes then their
>>
>>

marathon

Quoted message said:

>>time with be 3:15 hours. Runner's World has some articles at this
>>
>>

idea:

Quoted message said:

>Please not Yasso's 800's again. You can get the same accuracy from 3.972 times the height of the
>average 22.789 year old aardvark snout.
>
>
What about 9 times your hand span equals your height? Totally unrelated

to

Quoted message said:

Quoted message said:

the above, but I was shocked when I found it was 100% accurate.

'fraid not...

me: 8.5" hand span,...* 9 = 76.5 inches, or 6'4.5",...exactly 4" taller

than

Quoted message said:

I actually am,...so this little formula is not only wrong for me, it's not even close.

Ach you're obviously doing it wrong ;-)

Well, that's 2 of us then, since mine comes out exactly 4 inches too tall also.

dot

--
"Success is different things to different people" -Bernd Heinrich in Racing the Antelope

In article Tgcoa.286451$Zo.57051@sccrnsc03 said:

The best way to predict marathon time is the time it takes a person to run 800 meters. If a person
runs an 800 in 3:15 minutes then their marathon time with be 3:15 hours. Runner's World has some
articles at this idea: runnersworld.com0,1300,1 51 54 624,00.htmlOpen ↗

The runner's world notion, Yasso 800's, is _not_ how long it takes to run 800 meters. Plenty of
high schoolers can run an 800 in under
2:00, but the men's marathon world record is still a 'sluggish' 2:05 (and change).

Even if you actually did the Yasso 800's, it's a terrible predictor. It would be pretty easy for
a track person to crank out those 800's at a pace far faster than the 'equivalent' marathon. Take
a 20:50 5k runner. (Corresponds, loosely, to a 3:30 marathon). A typical workout for the 5k'er
would be to do 800's in 3:20, with about 2 minutes recovery. (5k pace, 2/3rds recovery). Give him
equal recovery, and he'll be cranking them out pretty easily at 3:15. The predictor is toast at
that point. (At least if you regard a 15 minute error at the marathon to be a bad prediction,
which I do.)

The only merit, and limited at that, is that they can serve as a negative standard. If you can't
do the 800's in 3:15, you can't do the marathon in 3:15.

--
Robert Grumbine radix.net~bobgOpen ↗ 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

In article said:

Robert Grumbine said:

There are a bucket load of predictors out there. Keeping it simple: First marathon = 5 * 10k
time Experienced marathon = 4.75 * 10k time

I'm so disappointed. After all the math revealed regarding work etc., I would have thought a
formula with Pi, logs, and 7 or 8 unknown variables would have fit. 😉

Ha. Only 7 or 8 variables, I thought you wanted something _fun_! I deal with millions, as a rule.
But I did call it simple. If it's going to be simple, why mess around with too few variables to be
interesting?

Quoted message said:

Quoted message said:

Sam and I once talked about collecting some data to try testing the predictors with, especially
the first marathon vs. experienced marathon. Never quite carried it out, but maybe one of these
years.

Frankly I don't think it would be worth the effort.

Ok, we won't make you do it. I've already received a data point in email. Anyone else willing to
volunteer some data points for 10k - marathon (first, nth, of each)?

--
Robert Grumbine radix.net~bobgOpen ↗ 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

Great! I've done 1:55.8 so look out for my 1:56 or so marathon any time soon... Cheers Tim

"Alan Williamson" <[email hidden]> wrote in message news:Tgcoa.286451$Zo.57051@sccrnsc03...

Quoted message said:

The best way to predict marathon time is the time it takes a person to run 800 meters. If a person
runs an 800 in 3:15 minutes then their marathon time with be 3:15 hours. Runner's World has some
articles at this idea: runnersworld.com0,1300,1 51 54 624,00.htmlOpen ↗

"fjwilson" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

i heard of a formula years ago that one could use to predict marathon times from 10K time
performance. I don't know how accurate it was, but it was quite simple. Can anyone help? Thanks.
fjwilson

Doug Freese said:

Quoted message said:

Sam and I once talked about collecting some data to try testing the predictors with,
especially the first marathon vs. experienced marathon. Never quite carried it out, but maybe
one of these years.

Frankly I don't think it would be worth the effort.

Actually, I think it would be worth the effort! (Here I go again, advocating for proof...)

How many times are estimations, formulas, and standards thrown around here without any clarity on
accuracy? Whether it's heart rate formulas, calorie burning formulas, marathon prediction formulas,
how many have...
a) a source at all
b) a source with a reference to a study
c) a study of real outcomes such as the relationship between 10k time and marathon time for
1000 runners?

I only point this out because (gasp) the practice of medicine is full of lore and speculation which
has become entrenched without ever knowing whether it's true. I love it when someone finally goes
out and sees what happens to 1000 real people when one thinks that screening test X or vitamin Y or
blood-pressure pill Z might be effective.

So why not prove things in the field of athletic performance? It ought to be fairly easy with a
large running group such as the NYRRC, as an example (only one I can think of off top of head), or
college track teams in same athletic conference, etc.

Just curious, as always. -- Josh

In article said:

Robert Grumbine said:

There are a bucket load of predictors out there. Keeping it simple: First marathon = 5 * 10k
time Experienced marathon = 4.75 * 10k time

I'm so disappointed. After all the math revealed regarding work etc., I would have thought a
formula with Pi, logs, and 7 or 8 unknown variables would have fit. 😉

The race time predictor I use consists of a composite of an exponential and a quadratic, and uses
Newton's mthod to invert the formula (-; (but I doubt it's that much mre accurate as a result)

--
Donovan Rebbechi pegasus.rutgers.edu~elflordOpen ↗

In article said:

Doug Freese said:

Quoted message said:

Sam and I once talked about collecting some data to try testing the predictors with,
especially the first marathon vs. experienced marathon. Never quite carried it out, but maybe
one of these years.

Frankly I don't think it would be worth the effort.

Actually, I think it would be worth the effort! (Here I go again, advocating for proof...)

How many times are estimations, formulas, and standards thrown around here without any clarity on
accuracy? Whether it's heart rate formulas, calorie burning formulas, marathon prediction
formulas, how many have...
a) a source at all
b) a source with a reference to a study
c) a study of real outcomes such as the relationship between 10k time and marathon time for 1000
runners?

The question about HR formulas has been tested, and they've computed the regression lines etc. There
are a lot of studies on this.

Calorie burning has been tested, though I don't know what the extent of this
is.

Quoted message said:

I only point this out because (gasp) the practice of medicine is full of lore and speculation
which has become entrenched without ever knowing whether it's true. I love it when someone finally
goes out and sees what happens to 1000 real people when one thinks that screening test X or
vitamin Y or blood-pressure pill Z might be effective.

So why not prove things in the field of athletic performance? It ought to be fairly easy with a
large running group such as the NYRRC, as an example (only one I can think of off top of head), or
college track teams in same athletic conference, etc.

Hey you know, that's a good idea. For example, one could download the results of NYRRC events and
analyse them.

You know, if you like, I could take a cut at doing something like this. Maybe. When I have time. It
would be reasonably easy to grab the first 500 runners from several events, look for duplicates and
compare them. The main problem would be that maybe some people don't race over a wide enough variety
of distances.

Cheers,
--
Donovan Rebbechi pegasus.rutgers.edu~elflordOpen ↗