Age-Grading your performances

[patybobady]

I like this. I messed around a bit, and when i typed in 4.11 meters and switched the age to 12, a male is only in the 94%...there are 12 year old boys going 13'6"???

[master]

I have written emails to a few people about the anticipated update of the Age Grading Tables. I recommend reading this link from the blog archives of Ken Stone of MastersTrack.com. The story behind the delay is pretty bizarre.
http://masterstrack.com/blog/archives/000266.html
Enjoy,
- master

[nambi]

Some of the age graded tables seem to us to be too generous, especially for the women's field events. We have developed some tables that we believe are a more accurate prediction of age graded performances.

Our investigation was prompted by the almost absurd age grade assigned to my last mark: 114% according to the WAVA table. According to our table it should be 92%.

Any event involves 4 factors: Strength, speed, endurance and skill, in varying amounts. The speed factor appears to decline at .55% of maximum speed per year at about age 30 up to 70, where it appears to decline at an increasing rate. Both strength and endurance appear to follow a similar pattern, except that they don't appear to start to decline until age 35, at about .6% per year. The skill factor, which is more important for field events such as the pole vault and javelin throw, appear to begin to decline at age 30 at about .6% per year.

Our tables are based on determining the various combinations of speed, strength, endurance and skill for the various events. The pole vault is the most skill dependent event and therefore has the highest yearly rate of decline for that factor. But, even with this highest rate of decline, our tables suggest the following samples of 100% performances:

40 year old male: 5.55
50 year old male: 4.92
60 year old male: 4.33
70 year old male: 3.77
60 year old woman: 3.47
50 year old woman: 3.94
40 year old woman: 4.44

We welcome any comments.

[master]

Hi nambi,

It looks like you (and others?) have done some homework. It is too bad the new tables were "lost" see this article. It puts everyone in the position of having to conjecture about values. For what its worth, I looked at your numbers and would agree your proposal is more appropriate than the 1994 WAVA values, at least for my age. I'm not as knowledgable with other age groups. But if your method works for M60 it probably is good for the nearby age groups.

The table below just puts your numbers in with WAVA numbers and the World Outdoor records as currently shown on MastersRankings.com.

Code: Select all

Men     Open = 6.15 (from WAVA)
-40-   -50-   -60-   -70-   
5.35   4.65   4.04   3.52   WAVA age graded
5.55   4.92   4.33   3.77   suggested
5.50   4.73   3.90   3.31   world records from MastersRankings

Women    Open = 4.45 (from WAVA)
-40-   -50-   -60-
3.79   3.26   2.81      WAVA age graded
4.44   3.94   3.47      suggested
3.45   3.25   3.05      world records from MastersRankings


By any chance have you shown your information to Howard Grubb? (see his web pages at http://www.howardgrubb.co.uk/athletics/home.html)

- master

[nambi]

Hi Master,
Thank you for putting the examples in a table for comparison and for your comments.

Bud Held has done all of the work. I thought it would be good to get his ideas out for some feedback.

We are curious as to how others react to our projected numbers. Pole vaulters decline at a faster rate than sprinters and distance runners. We believe this is because of the "skill factor." But, note that our suggested performance numbers are considerably higher than the Wava or actual world records. Bud has contacted Howard Grubb, who informed him that he will up-date his calculator as soon as he gets new numbers from WAVA. I think they use actual world performances to draw curves based on those data points.

Our numbers are based on an estimated potential for masters, rather than on actual performances. We would especially appreciate any suggestions or comments regarding our suggested rates of decline.

[master]

The more I think about the idea of age grading, the more difficult the task appears. Allow me to think out loud a little and you can comment as you see fit. First there is a difference between the best that anyone has done (ie WR) and the best that we could imagine anyone could do. Imagine the best coach training the most physically fit, fastest and most gymnastically inclined athlete in an age group that could be found. This may be coming at the task from the opposite direction as you describe, that of de-rating from the "open" performance for age. (This may be getting confusing , but bear with me please.)

Following is what went through my mind as I looked at the numbers you suggested. In less than a year I will be 60. You provide a number for that age so I will work from there. I have jumped over 3.95 recently in practice. I expect to be able to do that when I turn 60. And because I presently can not do many aspects of my jumps as well as I think I can learn to do them, and I think I can increase my speed a little and I think I will be able to get on slightly longer poles, it may be possible to do better. So I look at the current WAVA value of 4.04 as being possible for me to jump. Now, as much as I would like to be performing at 100%, I can easily imagine a vaulter my age that is a little better than me at almost every aspect of jumping and conditioning. So that leads me to say 4.04 is not accurate, but how will I determine if 4.33 (or any other number) is the correct value? I don't know. The calculation of 4.04/4.33 results in a 93% rating which sounds plausible.

Maybe we take the approach of de-rating a performance on some kind of a trend line that is similar to those generated for running, but modify the slope of the curve based on a few data points like I just described in my thought process. In fact, that may be exactly what Bud Held and others did to come up with these numbers. Hopefully other vaulters will go through the mental exercise that I did and offer their opinions about the suggested numbers.

Although this is already longer than intended, I wanted to ask about the women's Open number of 4.45 that is used in the 1994 tables. At the very least, the performance comparison should be made to near the current WR (around 4.90). Or am I mistaken about how that should work?

- master

[master]

In an email from Rex Harvey he comments

... the full tables were ready for release when questions arose about women's running on the Track (questionable Chinese World Records) and that is now being sorted out.

This apparently is what is now holding up the release of the upgraded WAVA age grading tables. I have written to Rex to ask if the pole vault data could be released "unofficially". I will post what I learn. It would certainly be understandable if he felt that was not appropriate.

[master]

I have some real information for everyone. I have been in email contact with Rex Harvey (who has been very responsive and helpful) and have learned the new age factors for pole vaulting were officially released sometime in 2002. They were released as part of the Decathlon age factors. They can be found on the World Masters Athletics web site. Here are links directly to the pdf files for the women's decathlon and the men's decathlon.

Additionally, noted in Rex's emails, are the Open Class standard for men of 6.14m and for women 5.00m (an estimate of what was expected at the time of the release.) The age factors are applied to these numbers to determine all age-related measures. Also worth noting is the philosophy that

...the WAVA age grading curves were based more on "what people should be doing" rather than what they have done in actual performances as (in) this 2002 update.

I will do some looking to see if there is a web based calculator with these newer factors available. If not, I might be able to build one. Until then, here is the simple arithmetic used to calculate the
1) age-graded result = your result * your age factor
2) age standard = Open Class standard / age factor
3) age performance % = age-graded result / OC standard

Below are the factors and computed age-standards.

Code: Select all

       ---Men---       ---Women---
OC std    6.14            5.00

Age  factor  age-std factor  age-std
---  ------  ------- ------  --------
30   1.0000   6.14   1.0000   5.00
35   1.0390   5.91   1.0459   4.78
40   1.1046   5.56   1.1225   4.45
45   1.1791   5.21   1.2111   4.13
50   1.2643   4.86   1.3150   3.80
55   1.3628   4.51   1.4384   3.48
60   1.4780   4.15   1.5873   3.15
65   1.6144   3.80   1.7707   2.82
70   1.7786   3.45   2.0019   2.50
75   1.9800   3.10   2.3025   2.17
80   2.2327   2.75   2.7095   1.85
85   2.5595   2.40   3.2911   1.52
90   3.0700   2.00   3.8772   1.29
95   4.0933   1.50   5.3357   0.94
100+ 6.1400   1.00   8.5536   0.58


- master

[vaultwest]

Master

Thanks for all your hard work on this. I have a couple of questions in your formulas in number 1. age-graded result = Your result*Your age factor does the * mean times. and secondly so at age 50 if I jump 4.12m is my result as follows; age graded result = 5.21 so My percentage equals 1.07% that seems high to me have I done something wrong.
Thanks

[souleman]

Not to be snotty but.........."Yeah, like I'm going to ever see 16'5" in my lifetime"or in the next 7 years for that matter. Frankly, it's rather depressing. Later............Mike

[master]

Master

Thanks for all your hard work on this. I have a couple of questions in your formulas in number 1. age-graded result = Your result*Your age factor does the * mean times. and secondly so at age 50 if I jump 4.12m is my result as follows; age graded result = 5.21 so My percentage equals 1.07% that seems high to me have I done something wrong.
Thanks

Thanks vaultwest. I'm glad I have keen eyes keeping me honest. I made a mistake which I will correct in the previous post so as not to let it be displayed incorrectly anymore.

The answer to your first question is Yes. "*" means times. The correction that I will make in the formula is
3) age performance % = age-graded result / OC standard

With that correction, your 5.21 age-graded result computes to an age performance % = 5.21 / 6.14 = 84.9% Does that seem more appropriate?

- master

[master]

Not to be snotty but.........."Yeah, like I'm going to ever see 16'5" in my lifetime"or in the next 7 years for that matter. Frankly, it's rather depressing. Later............Mike

Hey souleman, don't let the absolute numbers make you sad . Figure out how you are doing now and compare that to what you actually do as the years progress. Also, if you are comparing from one year to the next, remember you can/should interpolate the age factors for individual years. I could post those numbers if anyone was interested. Or that refinement could be included in a web based calculator if it were developed.

Smile, and keep up the good fight. The cold Minnesota weather should break soon and you will be working out in the warm sunshine . That always helps this old body feel better.
- master