DIFFERENCES BETWEEN DISCIPLINES




Analysis

World records of masters in a certain discipline are not all just as good, as became clear in the graphs. It is not difficult to sort out the weakest records, but how to find out whether a certain discipline is weak or deviating in all age groups? Or in other words: can something be learned by comparing a discipline with other disciplines?

First analysis: tangents

In the graphs with records a tangent line has been drawn as first aproximation of how performances decline with age. You would expect that in all disciplines decline would go in about the same manner, or at least that resembling disciplines show resembling decline. So 100 resembles 200, 200 resembles 400, shot is like discus, triple and long jump are about the same.
A tangent line starts at the upper left at a certain age where decline of performances sets in, the tangent line then goes with a certain slope to the right. Slope is percentage decline per year. These two things (start and slope of decline) are shown in the diagrams below:

[grafiek]


[grafiek]


[grafiek]


[grafiek]


That doesn't look too uniform... 400 of both men and women do not resemble neighbouring distances, discus women is differing from the other throws, etectera.
Age on which decline sets in differs severely, it goes from 24 to 43, which seems biologically unlikely.

Second analysis: all tangent lines

Showing all tangent lines in one graph also does not give much information, differences between disciplines are quite big:

[grafiek]


[grafiek]


It becomes better when related disciplines are compared -- runs, jumps and throws:

[grafiek]


[grafiek]


[grafiek]


[grafiek]


[grafiek]


[grafiek]


When a tangent line is steeper than another one this can have several causes:
A High age at which decline starts: records of younger masters are very strong
B Records of older athletes are weak
C Low age of decline onset: records of younger masters are weak
D Records of older athletes are strong
E The absolute record (OC) is very strong
F Or that record has been influence by the use of drugs
G Or combinations of these causes...

Third analysis: no tangents

The tangent lines do not give much information, let's look at the records themselves. Up to now in the graphs a big part of the sheet is empty, everything is happening in a relatively narrow diagonal strip. Better is not to use percentages but only how percentages are away of the thick black line in the graphs below. It is an arbitrary line just above all records. It's only function is to show small differences a bit enlarged.

[grafiek]


[grafiek]


Now it can be calculated how much all masters records are below this baseline. The numbers are negative, the upper limit of the graph is y=0, the lower limit is chosen to give a well filled sheet:

[grafiek]


[grafiek]


It becomes a rather chaotic picture but it makes clear that decline is not of the same type in all disciplines. Therefore I give runs, jumps and throws apart:

[grafiek]


[grafiek]


[grafiek]


[grafiek]


[grafiek]


[grafiek]


Some trends

Now differences become visible between disciplines and between men and women. Because athletics is been done more by men than by women (this is true for all ages and all countries) first the men's graph is looked at, followed by the women's graph. The baseline is the same for both sexes (thick black line some pictures above), so different trends between men and women should become clear. When I speak about 'significant' no official statistic significance calculation has been done.

Runs - men

Shorter distances show less decline than longer distances. From about 800 meters all runs follow more or less the same pattern. At first decline seems to be about linear, in higher age groups decline accelerates. This sets in at about 70 or 80 years.

Runs - women

Decline is more severe than decline of the men, again the sprints show less decline than longer distances. Accelerated decline seems to be less pronounced than in men and maybe sets in at age 75 or 80. Significant? Interesting is that no menopausal effect seems to be present.

Jumps - men

Excel has chosen light colors, I have kept them. High jump shows less decline in performances than the other jumps, all are close to linear. Accelerated decline happens later than in the runs, at age 85 or even 95!

Jumps - women

Pole vault has only two strong records, W35 and W65, all others being weak. It is a new discipline for women and the older women will never have done pole vault in their youth. Triple jump also is a new discipline, but it does not show such a pronounced effect. As with men high jump shows less decline. Accelerated decline seems to take place earlier than for men.

Throws - men

Overall pattern: linear. Javelin is positioned low in the graph which could mean that the OC record of Jan Zelezny is extremely strong, or that decline sets in earlier than is the case with other throws, or both. The curve of weight throw is a bit chaotic: an underdeveloped discipline in younger age groups? Or the square root formulae does not work well for the weight throw? (See below 'WANTED'.) Discus in fact has an extraordinary record at age 59, but this men has been caught cheating, so it is put aside in thi graph and in the rest of this study. Question is of course whether more records are affected by doping.

Throws - women

Clearly another pattern than throws of the men and other disciplines of women. Is it a concave pattern or is it a special effect in the youngest age groups after which a linear decline like that of the men sets in? Possible explanation: here in these disciplines where power is important you see a menopausal effect, from age 55 decline is like the men. Or: in the youngest masters categories records are affected by drugs. Weight throw W35 is very underdeveloped.

Next step

On the following page a mathematical model will be developed based on all these patterns.

WANTED

The graphs of the weight throw are a bit chaotic. To test how different weights compare experienced weight throwers (men and women) are asked to experiment with different implements. Use a workout to throw with two, three or even more implements and measure thrown distances carefully, as if it were a competition. I would be very glad to recieve those data!

See also:
Backgrounds of age gradings
Masters world records
Three systems of age grading compared
A mathematical model of gradings
The model in use
Tests, fine tuning, problems



Weia Reinboud (weiatletiek (at-symbol) xmsnet (dot) nl)

Also see my page on athletics, in case it isn't open yet.