At this point, I think it would be more feasible to improve this new system that we've got than to start from scratch.
About ten years ago, when there were some complaints about the ordinal system (from Cinquanta, from some new fans who maybe were more familiar with gymnastics scoring at the time, etc.), I had some thoughts about ways to make things a little bit more objective than just letting judges give two marks to produce ordinals with all the decisions that figure into those marks being made in their heads and hearts but closed to the skaters and the public.
My plan A thought was to keep the 6.0 system and to define the official expected technical content for, let's say, 4.5 and 6.0 benchmarks, maybe 5.0 and 5.5 as well. Also define specific bonuses that could be added to those benchmarks and deductions to be subtracted.
Presentation marks would still have been based on each judge's overall subjective impression, adjusted as necessary to account for the technical level and comparison with other skaters in the event if we're still talking about ordinals.
Not really much of a change from the 6.0 system actually -- just more specific guidelines as to what it takes to earn a specific technical mark.
[Later, about the time that the current system was first proposed, the Australian federation submitted a proposal with guidelines for breaking the 6.0 technical mark into separate parts for jumps, spins, etc. But that never got anywhere.]
The other suggestion I was thinking about back then was similar to what we actually ended up getting, but not quite so complicated.
Set base marks for each kind of element.
Have an official (I was thinking referee, but it could be a separate "technical specialist" or "technical controller") identify which elements a skater actually completed. But they wouldn't micromanage the identification of errors or extra difficulty. At most, they would decide that a 2.5-revolution jump attempt counts as double rather than triple.
Judges could take 1-5 deductions from each element (I was thinking in terms of SP-type deductions of 0.1 to 0.5) for errors of varying numbers and severity. They could also give bonuses of 1-5 for added difficulty ("features" in the current system) or quality (+GOEs).
I did foresee some potential difficulties with that approach, some of which are solved by the more active technical panel function that we have now.
I don't remember what I had in mind for judging the presentation and in-between skating.
About ten years ago, when there were some complaints about the ordinal system (from Cinquanta, from some new fans who maybe were more familiar with gymnastics scoring at the time, etc.), I had some thoughts about ways to make things a little bit more objective than just letting judges give two marks to produce ordinals with all the decisions that figure into those marks being made in their heads and hearts but closed to the skaters and the public.
My plan A thought was to keep the 6.0 system and to define the official expected technical content for, let's say, 4.5 and 6.0 benchmarks, maybe 5.0 and 5.5 as well. Also define specific bonuses that could be added to those benchmarks and deductions to be subtracted.
Presentation marks would still have been based on each judge's overall subjective impression, adjusted as necessary to account for the technical level and comparison with other skaters in the event if we're still talking about ordinals.
Not really much of a change from the 6.0 system actually -- just more specific guidelines as to what it takes to earn a specific technical mark.
[Later, about the time that the current system was first proposed, the Australian federation submitted a proposal with guidelines for breaking the 6.0 technical mark into separate parts for jumps, spins, etc. But that never got anywhere.]
The other suggestion I was thinking about back then was similar to what we actually ended up getting, but not quite so complicated.
Set base marks for each kind of element.
Have an official (I was thinking referee, but it could be a separate "technical specialist" or "technical controller") identify which elements a skater actually completed. But they wouldn't micromanage the identification of errors or extra difficulty. At most, they would decide that a 2.5-revolution jump attempt counts as double rather than triple.
Judges could take 1-5 deductions from each element (I was thinking in terms of SP-type deductions of 0.1 to 0.5) for errors of varying numbers and severity. They could also give bonuses of 1-5 for added difficulty ("features" in the current system) or quality (+GOEs).
I did foresee some potential difficulties with that approach, some of which are solved by the more active technical panel function that we have now.
I don't remember what I had in mind for judging the presentation and in-between skating.