More than a month ago, the first competition using the Colley-Gustafsson format was held in Malmö. There were 25 competitors, 19 males and 6 female fencers. In order to see to it that everyone got a significant amount of fencing, the competition was run in a gender-mixed format up until the DE rounds, in which the genders were segregated. Everything was held in salle of the MF19 fencing club, and they provided all staffing except for us three in the competition leadership that provided knowledge of the – to everyone else – completely new competition format. Without the help of MF19 this format could not have gotten off the ground, and I am very thankful indeed for their kind and enthusiastic help.

The fencers were by a large margin positive to the new format, both in verbal comments and in the questionnaire – more on the latter later on in this post. The Facebook comments were also positive, unanimously so as far as I have seen.

I was also, by and large, happy with how things went. A few points will illuminate that:

  1. Of the 153 matches that were 5-point matches, only 14 ended with the loser not scoring at all. That seems to be a lower proportion than what is the usual in the regional competitions in Southern Sweden. Given that this competition was gender-mixed and open to all age categories, that is saying something.
  2. All fencers got to fence at least 11 5-point matches (average: 12,24) and then at least one DE match, so they got their money’s worth for coming to the competition.
  3. Of the 25 competitors, 13 ended up with a win-loss differential in the range -2 to +2, so they were up against mostly fairly evenly matched opposition.
  4. The eventual winner faced the fencers who got these final placements: 3 (three times), 9, 10, 4, 7, 5 (twice), 14, 13, 17, 2, and the eventual silver medalist in the women’s event who was placed 7th overall after poules and connection matches. He won all his 14 matches, and of those 11 were against opponents who were on the upper half of the ranking after poules. The highest-ranked fencer whom he did not face ended up being placed 6th, so no one can gainsay that he earned his gold fair and square. He took on the best of what the competition had to offer, and compiled a perfect record.
  5. There were no fencers who ended up winless – the less proficient fencers fence each other in the poules, and in such a match someone, as in every match, must win.

There were also two issues that proved that there is room for improvement:

  1. The logistics of the connecting matches could be better. Obviously, this is no small extent due to unfamiliarity with the concept, but the fencers did not instantly go to the right pistes against the correct opposition. I expect this to become better with more experience.
  2. Due to a communication error on my part, the Excel file did not take into consideration all 5-point matches when calculating the ranking after the second poule round. This error has since been corrected.

The fencers were given the opportunity to give their opinions regarding 6 statements on a questionnaire, and 14 of them availed themselves of this opportunity. The statements posed were:

  1. I think that most of the matches were close and hard-fought
  2. I think that the poules were more even and demanding than in previous competitions
  3. I think that I have fenced many matches
  4. I think that the competition was more fun than the traditional competition format
  5. I think that my final placement is fair, given what I achieved on the piste
  6. All fencers got a reasonably fair final placement, in my estimation

The respondents could respond with a number from 1 to 5, where 1 denoted complete disagreement and 5 a complete agreement. All that meant that the sum of the numerical answers from a given respondent could be in the range 5 – 30, with 30 a perfect score of total agreement on all statements. Of the 14 respondents, 2 gave this perfect score, and a further 6 gave responses adding up to 28. That means that the median response sum is 28 out of a maximum possible of 30, indicated that the respondents were quite pleased with the competition experience, by and large.

I tried to glean more information from the data by separating the answers into two groups – those from fencers with above-median total agreement score, and those with below-median. This was based on the hunch that if a respondent has an above-median agreement score, that might be due to that respondent being likely to give high marks as a personal trait, and the low agreement scores given by those respondents would be especially noteworthy – and the converse for respondents with a below-median agreement score. However, no trends that seemed to be statistically significant could be seen by doing so, and therefore such an analysis must wait until I have a larger statistical sample.

The questionnaire data that I have seem to support the following conclusions:

  1. The statement with the highest average agreement score (4,71 out of a maximum of 5) is “I think that I have fenced many matches”
  2. The statement with the lowest average agreement score (3,93 out of a maximum of 5) is “I think that the competition was more fun than the traditional competition format”.
  3. The other statements had fairly similar agreement scores, ranging from 4,21 to 4,36.
  4. The by far most common response was the maximum agreement score, accounting for 46 of 84 – 55% – of all responses. Every single respondent gave this agreement score to at least one statement.

The relatively low agreement score to the statement “I think that the competition was more fun that the traditional competition format” warrants further study. It might be so that there is some subset of all fencers whose competitive experience is not influenced by the competition format, so changing the competition format will not improve their experience. One way to see if that is the case is to provide the opposite statement to agree or disagree with: “I think that the competition was less fun that the traditional competition format”. By having those two statements in the questionnaire, one can see whether those fencers actively prefer the traditional format, or if it is so that it does not matter to them either way.

Also, the informational value of a subsequent questionnaire will be quite a bit higher if there are many more respondents.

In short: The competition was a success overall, but there are some parts where there is room for improvement.