While perusing the Evolution of Scoring Tables, as summarised in the World Athletics Scoring Tables for combined events, one encounters the following statement:
... the "Portuguese Tables" of 1949/1954/1962, acquired an excellent world reputation ...
So naturally the question arises what are these Portuguese Tables which earned such a flattering mention.
In my post on the note I found in Quercetani's book, I wrote about how I came across the reference to Amado, and how I managed to obtain a copy of his tables. In that post I promised to write about his work in the series on performance scoring. Now is the time to keep the promise.
Fernando Alberto da Silva Amado is known to most people as a playwright. He started writing for the theatre at the age of 17 and during his career authored more than 30 plays. Some people, when referring to his work on the tables, present him as a mathematician. Nothing is further from the truth. Amado had a degree in Historical-Geographic Sciences from the Faculty of Arts of Lisbon. Although not an athlete himself, he had a keen interest in athletics, probably kindled by his assisting to the 1924, Paris, Olympic Games. Soon after that he started working on his scoring tables. The first version was published after the war in french. (Don't forget that at that time french was the "lingua franca" in the western world: english having just established the first foothold towards total domination, thanks the US role in WWII). The single volume version of the tables, edited by the Portuguese Federation, appeared in 1956, but by then the tables of Amado had been circulating for quite a few years. The final, two volume, version came out in 1962, after the 1960, Rome, Olympics. (Amado passed away in 1968 at the age of 69, and no further work on the tables was published after 1962).
For the high end of his tables, Amado was based, not on the world record but on what he called the "lower limit of international class" performances. That was an excellent choice. In my work, in collaboration with Y. Charon, published in New Studies in Athletics 29 (2014) page 37, we argued that "the current world record cannot be used as a reference performance" and recommended an analysis based on the distribution of the performances versus the rank of the performer. Amado was in some sense a precursor in this domain. For the low end of his table Amado decided to attribute zero points to performances of young children. But he points out that zero does not constitute a limit since his tables can be extrapolated to negative numbers of points.
Amado devoted quite a few pages of his books to a critique of the "scandinavian" tables which had been used, under various guises, in athletics since the introduction of the decathlon and up to 1962. This may be due to a certain resentment on behalf of Amado of the fact that his tables, which were definitely better than the official ones, were never adopted by the IAAF (that's what World Athletics was called at the time).
Amado approach is an insightful one. He recognises that each event has its specificities and so the scoring has to be tailored to each particular event, taking into account the specific effort required by the athlete for each performance. He postulates that the difference in quality between two performances separated by a given number of points should be the same between any two other performances separated by the same number of points. Ensuring this throughout the 1000 point tables is a tall order but Amado did manage to get as close as possible to that.
What I found disconcerting is Amado's dismissal of the velocity as a useful quality for the construction of the tables. He claims that the quantity that should be used for track events is the time, in the same way as length is used for field events. While time is the quantity obtained at the end of a race, the physical quantity that can be used in order to quantify the effort is the velocity. From this remark alone, one sees that Amado has had a literary education and not a scientific one.
Although the Portuguese tables were the best one could do in their time they had one basic flaw. Amado was convinced that, for field events, a gain of a fixed amount for a low performance should bring in more points that the gain of the same amount for a high performance. To put it simply, progressing by 50 cm between 4 and 4.5 m in long jump should correspond to more points gained than progressing by 50 cm between 8 and 8.5 m. The graphic above shows the consequence of this assumption on the scoring. Unfortunately this is serious mistake (and Amado is not the only one doing it). Had the decathlon tables been based on such a principle they would have resulted in a stagnation of performances in field events. Why should one strive to improve his performances when the reward in points is ever diminishing? In fact the 1962 decathlon tables were plagued by such a feature: they were regressive for all field events. It was later claimed that this was due to a sign mistake, something one has trouble believing. And it took the IAAF more than 20 years before correcting this. The current tables are slightly progressive in all events.
Amado's, was a monumental work. In an era where calculation had to be done essentially by hand he produced tables covering all possible disciplines, including races for imperial distances, distinguishing performances for classical javelins and the Held-type ones, and even, in the 1962 version, providing scoring for fibreglass poles. He went as far as introducing the evolution coefficient, which would have made possible to compare performances registered at different periods. But being a realist he acknowledges that the problem of evolution is fundamentally intractable. Still his readjustment of the performances of the past olympic winners and/or record holders is an interesting read and a stroll down the athletics history lane.
No comments:
Post a Comment