The comparison of men and women performances in athletics is a long-standing question. The standard approach is to compare the current world records and get it done with. However, as I have argued in an article, co-authored with Y. Charon and published in New Studies in Athletics, this is a bad choice. The world record can be due to an exceptional combination of talent and circumstances. It might also be due to lower standards as far as control of the competition is concerned (wind-speed limit measurement, application of anti-doping controls, etc.). And the same applies to the olympic record and to any major championships record (to say nothing of the fact that, as far as middle distances are concerned, the races of major championships may be tactical ones and thus not representative of the real values of the athletes).
A simple solution would have been to neglect the top-most performances and compare, say, the records of the all-time 50th or 100th performer. In the figure below I show the evolution of the performance in long jump with the rank of the athlete for men and women. Already from this graphic we can see that a problem does exist. The women's performances decrease faster than those of men. (The possible explanation for this is that, despite a century of women's involvement in athletics, the recruitment of women athletes, in particular at high level, is still trailing behind that of men).
This becomes even clearer when one computes the ratio of performances as shown in the next graphic. In the article mentioned above we had proposed solutions to the problem. One was to fit the ratio points by a simple mathematical expression (the best fit is shown by the light, continuous curve). An even simpler solution is to use the data from the 100th performer onwards and extrapolate linearly towards 0. This is shown by the light, dashed, line in the graphic. Once this is done, one can use the value obtained at the intersection point as the reference value for the ratio. This is clearly an arbitrary choice but one which has the advantage to be systematic and easily implemented.
Once the methodology is fixed I can now proceed to a comparison of men and women performances over a wide range of events. The results are summarised in the table below.
We remark that concerning track events the ratio is close to 0.9 while for jumps the value is closer to 0.85. Thus women are good at running but less so at jumping.
This is in accordance with physio-anatomical studies which show that where woman are closer to men is in the case of lower-body force. Jumps put in contribution not only the legs but the trunk muscles as well and this results to a larger difference between men and women.
At this point one can wonder what is the situation concerning throws. First, due to the fact that implements of different weights are used for the two sexes, it does not make much sense to proceed to explicit comparisons. The choice of the implement weights, with a factor of roughly 2 between men and women (excluding the javelin, for which a special article will be written sometime), appears to be optimal, essentially due to the fact that the world records for shot, discus and hammer are comparable. This appears to be corroborated by the fact that some studies resulted to a factor of almost 2 between the upper-body force of men and women. However I believe this factor of 2 to be coincidental and in any case these observations do not give a handle for the comparison of men and women performances. So the way I decided to proceed to a comparison was by estimating what would be the performances of women were they to throw with men's implements. (We would have reached the same conclusions by estimating what would be the performances of men were they to throw with women's implements). I will limit the study to the discus and the shot, since hammer throw is a rather recent discipline and it has not yet reached the necessary maturity. (This is in fact the reason that triple jump and pole vault were excluded from the jump comparison).
First, using the data from the 100th up to the 500th performer and extrapolating linearly towards 0, I obtain a reference performance. These turn out to be 21.99 m and 20.07 m for men's and women's shot put respectively. For discus throw we have 68.08 m and 67.30 m respectively. Next starting from the women's performance, obtained with implements of weights 4 kg for the shot and 1 kg for the discus, I use the method presented in a previous post. Namely I use the expression
where m is the mass of the implement and f a term related to the arm inertia, with values 6 kg for shot put and 1.5 kg for discus throw. Knowing the length L of the throw one can compute the value of a, and using the latter, obtain the length of the throw for a different value of the implement mass m. I find thus that if women were to throw with the men's implement their reference performance would have been 15 m for the shot put and 48 m for discus throw. From these values we obtain a ratio of 0.69 for shot put and 0.71 for discus throw.
So, from 0.90 for running events, the ratio of women to men performances goes to 0.85 for jumps and all the way down to 0.70 for throws. This confirms what one would have intuitively expected, i.e. that women are not as good in throws as they are in the remaining track and field events. It would have been great to have direct confirmation of this by actual performances of women throwers using men's implements. Unfortunately this is something that is completely lacking (apart from Spotakova's throw, I referred to in the post mentioned in the previous paragraph). Till more data emerge we'll have to do with estimates and extrapolations, like the ones presented in this article. Still the image is clear, as summarised in the title of the post.
A simple solution would have been to neglect the top-most performances and compare, say, the records of the all-time 50th or 100th performer. In the figure below I show the evolution of the performance in long jump with the rank of the athlete for men and women. Already from this graphic we can see that a problem does exist. The women's performances decrease faster than those of men. (The possible explanation for this is that, despite a century of women's involvement in athletics, the recruitment of women athletes, in particular at high level, is still trailing behind that of men).
This becomes even clearer when one computes the ratio of performances as shown in the next graphic. In the article mentioned above we had proposed solutions to the problem. One was to fit the ratio points by a simple mathematical expression (the best fit is shown by the light, continuous curve). An even simpler solution is to use the data from the 100th performer onwards and extrapolate linearly towards 0. This is shown by the light, dashed, line in the graphic. Once this is done, one can use the value obtained at the intersection point as the reference value for the ratio. This is clearly an arbitrary choice but one which has the advantage to be systematic and easily implemented.
Once the methodology is fixed I can now proceed to a comparison of men and women performances over a wide range of events. The results are summarised in the table below.
| Event | Ratio |
|---|---|
| 100 m | 0.91 |
| 400 m | 0.89 |
| 1500 m | 0.88 |
| 5000 m | 0.88 |
| Marathon | 0.89 |
| High Jump | 0.85 |
| Long Jump | 0.83 |
We remark that concerning track events the ratio is close to 0.9 while for jumps the value is closer to 0.85. Thus women are good at running but less so at jumping.
M. Ahouré, the 2018 world champion over 60 m
This is in accordance with physio-anatomical studies which show that where woman are closer to men is in the case of lower-body force. Jumps put in contribution not only the legs but the trunk muscles as well and this results to a larger difference between men and women.
I. Spanovic, 2018 long jump world champion
At this point one can wonder what is the situation concerning throws. First, due to the fact that implements of different weights are used for the two sexes, it does not make much sense to proceed to explicit comparisons. The choice of the implement weights, with a factor of roughly 2 between men and women (excluding the javelin, for which a special article will be written sometime), appears to be optimal, essentially due to the fact that the world records for shot, discus and hammer are comparable. This appears to be corroborated by the fact that some studies resulted to a factor of almost 2 between the upper-body force of men and women. However I believe this factor of 2 to be coincidental and in any case these observations do not give a handle for the comparison of men and women performances. So the way I decided to proceed to a comparison was by estimating what would be the performances of women were they to throw with men's implements. (We would have reached the same conclusions by estimating what would be the performances of men were they to throw with women's implements). I will limit the study to the discus and the shot, since hammer throw is a rather recent discipline and it has not yet reached the necessary maturity. (This is in fact the reason that triple jump and pole vault were excluded from the jump comparison).
M. Abakumova who had, to my eyes, the best style in javelin throw
First, using the data from the 100th up to the 500th performer and extrapolating linearly towards 0, I obtain a reference performance. These turn out to be 21.99 m and 20.07 m for men's and women's shot put respectively. For discus throw we have 68.08 m and 67.30 m respectively. Next starting from the women's performance, obtained with implements of weights 4 kg for the shot and 1 kg for the discus, I use the method presented in a previous post. Namely I use the expression
where m is the mass of the implement and f a term related to the arm inertia, with values 6 kg for shot put and 1.5 kg for discus throw. Knowing the length L of the throw one can compute the value of a, and using the latter, obtain the length of the throw for a different value of the implement mass m. I find thus that if women were to throw with the men's implement their reference performance would have been 15 m for the shot put and 48 m for discus throw. From these values we obtain a ratio of 0.69 for shot put and 0.71 for discus throw.
So, from 0.90 for running events, the ratio of women to men performances goes to 0.85 for jumps and all the way down to 0.70 for throws. This confirms what one would have intuitively expected, i.e. that women are not as good in throws as they are in the remaining track and field events. It would have been great to have direct confirmation of this by actual performances of women throwers using men's implements. Unfortunately this is something that is completely lacking (apart from Spotakova's throw, I referred to in the post mentioned in the previous paragraph). Till more data emerge we'll have to do with estimates and extrapolations, like the ones presented in this article. Still the image is clear, as summarised in the title of the post.
No comments:
Post a Comment