Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Looking for Race Prediction Chart

1 view
Skip to first unread message

Mario

unread,
Apr 22, 2004, 10:30:48 PM4/22/04
to
I am looking for a race prediction chart that contains the most common race
distances from 5K to the marathon. Using such a chart, a person could look
up their best 10K time and then scanning across the row, find all the times
they could run at the other distances given they were exactly as good at
those other distances as the 10K.

I know there are a lot of calculators on the net that can be used to
calculate the predicted time for a given distance. I need all the
information in a chart format to be able to present how the concept works at
a marathon training seminar.

Alternatively, can someone recommend the best calculator option that can be
run standalone (that is, while not connected to the net) that is not a
browser/java based program.

Thanks,
Mario


Bethowmuch

unread,
Apr 23, 2004, 12:12:37 AM4/23/04
to
I hate to be a smart ass but this sounds like about fifth grade math. Your
common distances are 5K 10K half-marathon and marathon. You said assuming the
same pace for all events. Using that assumption you would multiply the 10K
time by .5 2.09677 and 4.19355 to predict the person's time for the 5K
half-marathon and marathon respectively. This basic math runs standalone, you
don't need to be connected to the net, you don't even need a computer or any
other fancy shit, a $1.95 calculator from Wal-Mart will work fine and you could
really get by with a #2 pencil and a piece of paper.

Dan Stumpus

unread,
Apr 23, 2004, 12:22:30 AM4/23/04
to
Mario:

Here's a little piece of the Gardner/Purdy tables. If you can get the
book, it also rates interval workouts by point level, which is useful...you
can even use your interval workouts to predict race performance.

This was remarkably accurate for me and my friends. I was at the 730 point
level, and my times were within a few seconds of the predictions.

http://www.runnersweb.com/running/purdytbl.html

--Dan

"Mario" <nos...@myhome.com> wrote in message
news:qd%hc.35267$CO3.1...@news20.bellglobal.com...

Dan Stumpus

unread,
Apr 23, 2004, 12:36:58 AM4/23/04
to
No, it's a bit more complicated...the problem is what is how fast a marathon
can you run if your 10k time is, say 42:30 (answer: about 3:21). There is
a formula, but it's a logarithmic function, too messy to do on your normal
calculator.

Purdy figured this out as part of his Ph. D. thesis, and wrote a great
little book with Gardner. The table I link to in my earlier reply shows
some of these relationships.

--Dan

"Bethowmuch" <betho...@aol.com> wrote in message
news:20040423001237...@mb-m17.aol.com...

Phil M.

unread,
Apr 23, 2004, 5:33:36 AM4/23/04
to
"Mario" <nos...@myhome.com> wrote in
news:qd%hc.35267$CO3.1...@news20.bellglobal.com:

http://tinyurl.com/34haz

Phil M.

Donovan Rebbechi

unread,
Apr 23, 2004, 7:33:55 AM4/23/04
to

I don't understand what your aversion to java is. I have a program that
computes Daniels vdot scores in java and python. The formula for vdot is
VO2 (distance*1000.0 / t ) / percent_max(t)
where
VO2(v) = J1 + J2 * v + J3 * v*v and

percent_max(t)=K1 + K2 * exp(K3 * t) + K4 * exp(K5 * t)

K1 = 0.8 K2 = 0.1894393 K3 = -0.012778 K4 = 0.2989558 K5 = -0.1932605
J1 = -4.60 J2 = 0.182258 J3 = 0.000104

To use this as a predictor, you need to either build a table or invert the
formula. My online version is http://www.panix.com/~elflord/predict.html

I'd also suggest the WAVA tables. You can download them as a
spreadsheet here:

http://www.personal.rdg.ac.uk/~snsgrubb/athletics/wavalookup.html

Cheers,
--
Donovan Rebbechi
http://pegasus.rutgers.edu/~elflord/

N_A_J

unread,
Apr 23, 2004, 10:17:13 AM4/23/04
to

pa...@atom.sbrk.co.uk

unread,
Apr 23, 2004, 11:28:12 AM4/23/04
to
In article <561d629b.04042...@posting.google.com>, N_A_J wrote:
> "Mario" <nos...@myhome.com> wrote in message news:<qd%hc.35267$CO3.1...@news20.bellglobal.com>...
>> I am looking for a race prediction chart that contains the most common race
>> distances from 5K to the marathon. Using such a chart, a person could look
>> up their best 10K time and then scanning across the row, find all the times
>> they could run at the other distances given they were exactly as good at
>> those other distances as the 10K.

> Try this:
>
> http://www.cararuns.org/pace_calculators/pacechart.html

If you try that, prepare to be disappointed if you don't make your
"predicted time".

5k 30:00 ... 30k 3:00:00

Paul

Donovan Rebbechi

unread,
Apr 23, 2004, 12:21:48 PM4/23/04
to
In article <20040423001237...@mb-m17.aol.com>, Bethowmuch wrote:
> I hate to be a smart ass

Good, because you're not being a "smart" anything.

> but this sounds like about fifth grade math. Your

Or fifth grade reading comprehension.

> common distances are 5K 10K half-marathon and marathon. You said assuming the
> same pace for all events.

No, he said "given they were exactly as good at those other distances as the
10K."

Cheers,

Donovan Rebbechi

unread,
Apr 23, 2004, 12:22:44 PM4/23/04
to
In article <561d629b.04042...@posting.google.com>, N_A_J wrote:

> http://www.cararuns.org/pace_calculators/pacechart.html

That's not a prediction chart (despite the label), it's a pace chart. The rows
correspond to different paces, not different levels of performance.

Bethowmuch

unread,
Apr 23, 2004, 5:05:45 PM4/23/04
to
> I hate to be a smart ass

Good, because you're not being a "smart" anything.

> but this sounds like about fifth grade math

Or fifth grade reading comprehension.

> common distances are 5K 10K half-marathon and marathon. You said assuming
the same pace for all events.

No, he said "given they were exactly as good at those other distances as the
10K."

Cheers,
--
Donovan Rebbechi


Another dipshit ivy leaguer heard from. Some academics can turn taking a shit
into rocket science.

Donovan Rebbechi

unread,
Apr 23, 2004, 5:47:16 PM4/23/04
to
In article <20040423170545...@mb-m23.aol.com>, Bethowmuch wrote:
> Another dipshit ivy leaguer heard from. Some academics can turn taking a shit
> into rocket science.

Look, there is more to predicting your marathon performance than multiplying
your 1 mile time by 26.2. If you don't understand this, you are a moron.

Larry McMahan

unread,
Apr 23, 2004, 10:04:03 PM4/23/04
to
Bethowmuch <betho...@aol.com> writes:

: You said assuming the


: same pace for all events.

No, dip. :-) He said assuming you were equally good at each race.
That does NOT mean the same page. Common wisdom is that when you
double your distance, you pace falls off by 10%. So, for example,
given your 5K time, your 10K time is 2.2 times the 5K time, not
twice, as your erroneous calculations assume.

But that what you get for being a smart ass!

Larry

Mario

unread,
Apr 23, 2004, 10:09:04 PM4/23/04
to
"Dan Stumpus" <dstu...@mindspring.com> wrote in message
news:aS0ic.5856$eZ5....@newsread1.news.pas.earthlink.net...

> Mario:
>
> Here's a little piece of the Gardner/Purdy tables. If you can get the
> book, it also rates interval workouts by point level, which is
useful...you
> can even use your interval workouts to predict race performance.
>
> This was remarkably accurate for me and my friends. I was at the 730
point
> level, and my times were within a few seconds of the predictions.
>
> http://www.runnersweb.com/running/purdytbl.html
>
> --Dan
>

Thanks Dan,

That is exactly the type of chart I was looking for. Now to be really picky,
does anyone know where I can find that type of chart with more common race
distances like the 15K and half marathon?

Also, thanks to all the rest for their answers.

In case anyone in the audience will be in the neighborhood, the marathon
training seminar will presented on April 28 in Sault Ste. Marie, Ontario at
the local YMCA starting at 7 PM.

http://www.saultstryders.com/

Thanks,
Mario

Donovan Rebbechi

unread,
Apr 23, 2004, 10:32:23 PM4/23/04
to

At risk of repeating myself, the WAVA tables do include these distances.

The Daniels charts are based on curves, so you can use them for any distance
but I wouldn't recommend using the Daniels tables to predict marathon times, as
Daniels' numbers don't work very well beyond a 2hr duration.

Mario

unread,
Apr 23, 2004, 11:28:06 PM4/23/04
to

"Donovan Rebbechi" <ab...@aol.com> wrote in message
news:slrnc8jkdn...@panix2.panix.com...

>
> At risk of repeating myself, the WAVA tables do include these distances.
>
> The Daniels charts are based on curves, so you can use them for any
distance
> but I wouldn't recommend using the Daniels tables to predict marathon
times, as
> Daniels' numbers don't work very well beyond a 2hr duration.
>
> Cheers,
> --
> Donovan Rebbechi
> http://pegasus.rutgers.edu/~elflord/

Donovan,

I checked the WAVA tables and they were for age and gender handicapping, not
race prediction.

I will check again. I may have scanned too quickly.

Thanks,
Mario


Larry McMahan

unread,
Apr 24, 2004, 1:00:17 AM4/24/04
to
Down boy! Why are you having a battle of wits with an unarmed person?

Larry

Donovan Rebbechi <ab...@aol.com> writes:

Dan Stumpus

unread,
Apr 24, 2004, 12:35:16 AM4/24/04
to
Mario:

If you can get ahold of one of Purdy's books (it was reviewed back in '97)
that would have the whole shebang, including interval workouts and all
standard metric and running distances.

Do a search for Gardner Purdy running tables and you'll find it out there.

--Dan

PS: Out here in California, I've never heard of a 15k! It's
5k/10k/half/whole/50k/50mile/100mile and various odd distances.

"Mario" <nos...@myhome.com> wrote in message

news:3%jic.42787$CO3.1...@news20.bellglobal.com...

Donovan Rebbechi

unread,
Apr 24, 2004, 7:47:27 AM4/24/04
to

You can use them for race prediction as well.

I have a WAVA based predictory online here:
http://www.panix.com/~elflord/wava.html

It's simple enough that I can explain how the predictor works, so here goes:

It works as follows: for each distance, there's a "gold standard" time, T. For
each distance, there's an age correction factor C. So if your time is t, your
rating is

r = T/t * C

one can invert this formula -- if your desired rating is r, then your required
time is

t = T*C/r

You can predict a time, by computing the rating for one performance and using
the inversion formula above to get the time at that rating for the other
distance. So you time predicted from a performance t1, at a distance with
"gold standard" time T1 correction factor C1, when the race is to be run over
distance with gold standard T2, correction factor C2 is:

t2 = T2*C2/r1 = T2*C2/(T1/t1 * C1)

Bethowmuch

unread,
Apr 24, 2004, 8:13:34 AM4/24/04
to
Exactly as good as, or assuming the same pace, a battle of semantics. I'm
capable of using polysyllabic words too. I stand by my guns about academics
and intelligence. Experience has taught me the relationship between academic
achievement and intelligence appears to be inverse rather than direct. I once
had a person with a degree in electronics ask for help fixing their car's turn
signals. I asked if they checked the fuse. They went into some long-winded
(albeit empty headed) dissertation on why that could not be the problem. Why
they babbled on I put in a new fuse and voila, the turn signals worked. As for
predicting a time, you can do a little simple mate or turn it into a project
for a super computer.

Donovan Rebbechi

unread,
Apr 24, 2004, 10:33:03 AM4/24/04
to
In article <20040424081334...@mb-m26.aol.com>, Bethowmuch wrote:
> predicting a time, you can do a little simple mate or turn it into a project
> for a super computer.

Using the WAVA formulas requires junior high level math ability at most, which
is well within the capabilities of most of the posters here (if not yourself)

tenkbabe

unread,
Apr 24, 2004, 1:17:45 PM4/24/04
to
betho...@aol.com (Bethowmuch) wrote in message news:<20040423001237...@mb-m17.aol.com>...

> I hate to be a smart ass but this sounds like about fifth grade math. Your
> common distances are 5K 10K half-marathon and marathon.
> You said assuming the same pace for all events.

Where did he say that?

Don't worry moron. You're not a smart ass. Not as smart as I.

tkb

ralphnade...@hotmail.com

unread,
Apr 24, 2004, 4:10:22 PM4/24/04
to
"Mario" <nos...@myhome.com> wrote in message news:<qd%hc.35267$CO3.1...@news20.bellglobal.com>...

> I am looking for a race prediction chart

You'll lose.

PaulM1125

unread,
Apr 25, 2004, 6:24:47 AM4/25/04
to
>That is exactly the type of chart I was looking for. Now to be really picky,
>does anyone know where I can find that type of chart with more common race
>distances like the 15K and half marathon?

You might try this one.
http://www.cs.uml.edu/~phoffman/ex1.html

Paul

0 new messages