Race time predictions and ... such

"steve common" said...

Quoted message said:

Can anyone explain how wildly different these predictions can be?

http://www.runningforfitness.org/calc/rp.php

I punched my numbers in and the purdy is the outlier for me. The others are
close enough for me...

The Purdy formula doesn't work for older men or for women -- it uses the
performance curve for young men. When I was in my 30's Purdy was dead-on
accurate, that's what it was optimized for.

Women and older men have less of a drop-off in speed as distances increase.

-- Dan

--
Posted via a free Usenet account from http://www.teranews.com

Stump said:

Women and older men have less of a drop-off in speed as distances increase.

Guess I'm an outlier then: I have more of a ramp-up in suckage as
distance decreases.

In article <[email hidden]>,

steve common said:

Can anyone explain how wildly different these predictions can be?

http://www.runningforfitness.org/calc/rp.php

Wildly different premises and verification methods. I do like,
though, that it goes out to ultramarathon distances. Like, at
least in the sense that it recognizes that such things exist.
What would actually happen on going out to such distances, and
realizing that one seldom runs ultras on courses as easy as
a 5k.

The site describes the methods they show:
http://www.runningforfitness.org/faq/rp.php#Q4

Of the list, only the VO2max one has any physiology involved.
The others are regressions (Riegel and Cameron) or ... performance
presumptions (age graded, Purdy).

If you're looking at races around VO2max, then this one ought to be
relatively good. On the other hand, you're not (you're trying to
go from your 10k to the marathon estimate, right?).

Cameron and Reigel both regress time and distance (why the
page says it's a model vs. a regression, when it's sitting there
with 4 tunable parameters ...). The Cameron as used on the
site, exponent of 1.06, is extremely optimistic. The more typically
quoted value is 1.07 (the difference becomes important for greater
extrapolations). For 200 m to 5k, for myself (and in this range
you seem similar to me) I use 1.115. 1.07 isn't too bad for
for me for over 5k.

Age graded takes the idea that if you run some percentage of
world record pace at one distance, you'll run that same percentage
at all other distances. It's obviously wrong, but for distances
only moderately different, may not be too bad. Purdy takes a
similar notion, but instead of percent of record pace, looks more
towards how you'd place in the different distances (relative
to open class men). To the extent that your age and gender are
different, and that performance at different distances depends
on that, this may not work well for you. (Well, you've got the
gender it's based on, but the calendar is starting to speak up.)

Noodling around a bit with times from when I was in good training,
Cameron is terrible -- too pessimistic for going to shorter distances,
far, far too optimistic on going to longer. Riegel tends that way
as well. The others aren't too terribly off. Hope you weren't
pinning your hopes on a Cameron prediction.

(n.b., you can get really amusing results by entering a sprint
time -- a non VO2max race, and looking at VO2max's prediction --
my 30.75 200m predicts, by that, a 2:32 marathon. Riegel and Cameron
say 2:29 and 2:36, also ludicrous. Age grading gave 3:09 and Purdy
gave 3:59, more plausible figures.)

--
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

(Robert Grumbine) said:

Wildly different premises and verification methods.

Figures.

I tended to "believe" the VO2max column, in the range I generally run (10k
to 42k) but the age graded one is more honest when I compare my 200m time
aged 45 (28.4"😉 to marathon time.

I'll have another go at this in the light of your input (notably the coeff
tweaking...) thanks for taking the time :-)

PS No, Cameron got zilched at first reading :-) Using the predictor in a
serious fashion (coupled to observed training paces vs. effort at all
paces - a kind of personal regression, back-fitted to 2002 performance
curve), I reckon I can hope for something between 2:52 and 2:56, depending
on whether my observations (and math) suck, or not ;-)

Chippie said:

Stump said:

Women and older men have less of a drop-off in speed as distances
increase.

Guess I'm an outlier then: I have more of a ramp-up in suckage as
distance decreases.

Hey, when you get to be my age, you get the suckage with decreasing
distance, *plus* sluggage at longer distances.

--
Posted via a free Usenet account from http://www.teranews.com

steve common said:

Can anyone explain how wildly different these predictions can be?

http://www.runningforfitness.org/calc/rp.php

PS I am not accepting worst-case scenario unless you kiss my bum and
promise you'll love me for ever :-P

The Cameron formula is completely bogus, the others aren't that far apart.

Daniels tables fall apart at distances less than 5km and more than half
marathon. WAVA tables are solid, and take age/gender into account (other
methods don't)

btw taking an "average prediction" is really dumb.

Cheers,
--
Elflord

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

Quoted message said:

Chippie said:

Stump said:

Women and older men have less of a drop-off in speed as distances
increase.

Guess I'm an outlier then: I have more of a ramp-up in suckage as
distance decreases.

Hey, when you get to be my age, you get the suckage with decreasing
distance, *plus* sluggage at longer distances.

You have yet to experience real suckage until the 60's and then it's
geometrically downhill on your ass. 😉

-Methuselah's Grandfather

commie said:

Can anyone explain how wildly different these predictions can be?

http://www.runningforfitness.org/calc/rp.php

You wouldn't expect to run the exact same time for every race at a
given distance, would you?

Well, these methods each predict your time at a different race! If
they're radically different, obviously they anticipate variations in
course terrain, weather, your training, injury status, shoe choice,
etc.

In article <[email hidden]>,

steve common said:

(Robert Grumbine) said:

Wildly different premises and verification methods.

Figures.

I tended to "believe" the VO2max column, in the range I generally run (10k
to 42k) but the age graded one is more honest when I compare my 200m time
aged 45 (28.4"😉 to marathon time.

I'll have another go at this in the light of your input (notably the coeff
tweaking...) thanks for taking the time :-)

PS No, Cameron got zilched at first reading :-) Using the predictor in a
serious fashion (coupled to observed training paces vs. effort at all
paces - a kind of personal regression, back-fitted to 2002 performance
curve), I reckon I can hope for something between 2:52 and 2:56, depending
on whether my observations (and math) suck, or not ;-)

Hey, you could always send me your times and then rely on whether
my math sucks or not.

A predictor that is not on that table, but which has worked well
for me -- including that it makes allowance for how many miles you're
training -- is Tom Ehrensperger's 'Runpaces'
http://members.aol.com/eburger/index.html
See also his verification study/report:
http://members.aol.com/eburger/study.html

Tom used to be a regular here.

--
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

Elflord said:

steve common said:

Can anyone explain how wildly different these predictions can be?

Quoted message said:

http://www.runningforfitness.org/calc/rp.php

Quoted message said:

PS I am not accepting worst-case scenario unless you kiss my bum and
promise you'll love me for ever :-P

The Cameron formula is completely bogus, the others aren't that far apart.

Daniels tables fall apart at distances less than 5km and more than half
marathon. WAVA tables are solid, and take age/gender into account (other
methods don't)

btw taking an "average prediction" is really dumb.

Cheers,
--
Elflord

Yeah, things vary. I just realized I could get a VO2 max from a Bruce
treadmill stress test I did two years ago....I plugged in 16 minutes =
59.04. I know I could have gone another minute, 62.74. For some
reason they made me stop at 16. Maybe they didn't want to clean up
after me.

But I plug in my best half marathon (1:31ish) into the link and I get
50.3

I'm sure it has something to do with my short legs and long muscular
torso. I'd rather believe that it just means I have much more room
for improvment over the distance, and efficiency.

gym.gravity said:

I'm sure it has something to do with my short legs and long muscular
torso.

(yeah, like a chimp, or a gorilla)