Battle scoring (placing points)
Comments about this discussion:
Started
While creating my proposal for battle scoring, I realised the difference in comparing overall total scores and total scores by individual judges.
Example:
Judge1: 50 points for Rider A, 25 points for Rider B
Judge2: 30 points for Rider A, 40 points for Rider B
Judge 3: 30 points for Rider A, 40 points for Rider B
Overall score for Rider A: 110 Points
Overall score for Rider B: 105 Points
So Rider B would place behind Rider A, despite the majority of judges believing that Rider B was the better rider. This is an extreme example but it shows that just adding up scores allows for individual judges potentially having a greater weight, even though the majority of judges agrees on a different outcome.
I would propose to change prelim scoring to translating total points into placing points. For example, if there are 3 riders, the rider with the highest total score by an individual judge would receive (1) placing point, the second highest score (2) placing points, the third (3) placing points from this individual judge. Placing points by all judges would then be added up and the rider with the lowest placing points would be ranked first. If there is a tie between riders based on placing points, it would be possible to look at overall scores as a next step.
I think this would give all judges equal weight, which in my mind, makes for fairer judging.
What do you think?
Comment
I think this is particularly relevant when riders scores are close.
Comment
First to clarify. You named this topic 'Battle Scoring ...' but this is only meant for prelims right? Because in battles it is exactly how you would like to see it. Judges create a score for each of the two riders and the higher scoring one gets one voting point from that judge and a single majority vote decides. So there is no adding scores across judges for battles.
For prelims though I have to disagree. While you might be right that in this specific extreme case the decision could arguably be 'unfair' when adding scores, this only highlights an important part of finding judges.
Having good ones.
Because scores like this should never be given. Or when they have, this judge would most probably not be judging again anytime soon.
Let me explain from my perspective as chief judge of a number of competitions by now:
It does happen (quite often even) that a judge tends to give higher or lower points in general than the other judges. This is not problematic as long as they are consistent in the level of given points. This is just caused by Flatland's subjective system of judging, which is dependent on the suspected level of riding instead of having fixed scores for every trick.
What should never happen is a judge giving a rider B a higher score compared to rider A when all other judges give rider B a lower score. So the changes of scores between different riders (raising it or lowering it) should be somewhat consistent across the whole panel. And after every competition I go through the judging sheets to find those anomalies which would then lead to either a more detailed workshop for that judge next time I ask them or me not asking them again.
But why not go with your proposed system just to be sure?
It would make the scoring system a lot less transparent. Having points given in relation to the scores of other riders (these points also only being able to be determined after every rider has been scored) could lead to a lot of confusion and misunderstandings.
It would also make it impossible to take the audience into the competition by displaying scores. So we would go back to having every rider do their prelim without anyone even having an idea of how they did and I think it was a really nice addition in terms of engagement with the riders and the audience to know the scores right after each run.
Comment
I think it is good think you brought this topic up Marie, this thought came to my mind a few times too; but after giving it some thought myself I have came to the conclusion that is well aligned with Ian's point.
I agree with him that this really just shows how important it is to have good judges on the panel. Instead of making the scoring system more "blunt", I agree that it is better to have well trained judges. As long as they as consistent with their scoring, these anomalies should balance themselves.
This is how I see it: in Flatland, judging is subjective. This is why we have people as judges, otherwise we could just have an excel sheet where we enter all tricks the rider did and it would give us a score instead of people giving scores after a run. This means that the judges have different perspectives, and they may have different ideas on how difficult tricks are, and what they find stylish and how much. This is why it can't be said enough that it is a great responsibility to put together a judging panel, because you have to keep all this in mind (plus more, to make sure not all judges are from the same country, club, team, etc., don't have the same style, ... ). For instance, the example with the scores you brought up could be very realistic for a rider who is doing many coasts and freestyle influenced tricks. Judge 1 probably understands these tricks better and therefore gives higher scores for them. Judge 2 and 3 are probably coming from a more flip tricks based background and they are less used to these styles, so they give lower scores to them. This is not a problem as long as there is balance on the panel, and both (all) perspectives are represented.
As an idea for the future, we could consider to remove the highest and lowest judging scores for every run. If I am correct they do this in many Olympic sports, for instance in figure skating or figure diving. So if there are 7 judges, and they score a run 5.9 , 6 , 6.3 , 6.45 , 6.46 , 6.55 , 6.89 ; then 5.9 and 6.89 will be removed from the athlete's final score. I think this could be a good system BUT this would require more judges otherwise you just remove the variety of the judging panel. This would also require a sort of standardization in scoring levels, because if there is a judge who tends to give higher scores, then basically their score will be removed all the time. There are great formulas in statistics btw to measure the consistency of judges and judging panels.