- Joined
- Jun 21, 2003
I think the reaction of many fans might be, "Nuts to you, you stupid AI robot. This was a beautiful performance. If you didn't like it, so much the worse for you!"
Why do "full rotations" matter?
- Thread starter Anna K.
- Start date
- Joined
- Jun 21, 2003
I wanted to go back to this comment, comparing the role of the ! mark with the q.
A properly done Lutz presents a clear outside edge. If it is definitely inside, that's a flaw in the jump that gets a well-defined penalty. If it is not a clear outside edge but rather wobbles over toward the flat at the last moment, that, too, is a technical flaw, but not as serious as the e. So it gets a smaller penalty.
Is there anything to be gained by some advanced angle-measuring technology that can discern the difference between 1 degree to the outside of vertical versus 1 degree to the inside? Or that can determine that the degree of flatness is within +/- five degrees, so give it a !, but if it is 6 degrees to the wrong side, that's an "e"?
- Joined
- Sep 23, 2020
Seemingly I am of better opinion of people, in the end of the day ....I think the reaction of many fans might be, "Nuts to you, you stupid AI robot. This was a beautiful performance. If you didn't like it, so much the worse for you!"
I can only repeat I believe if there was a feeling that calls are justified, non-selective and consistent for the whole field - and use of technology could create it - I think scores would be much better accepted and complaints would be much fewer compared to now.
It is their selectiveness and inconsistency that make them perceived as unfair, which is difficult to negate really, as inconsistent and selective scoring actually is unfair, generally speaking. Which creates a heyday for complaints.
Last edited:
- Joined
- Sep 7, 2023
It's the story of her gymnastics career- all the highs and lows.
She won her first Asian all around title in 2015. Her second- 2025.
I tried to follow NCAA but the scoring confuses me.
Last edited:
IMO the measurements should focus on the depth of the curve before Lutz instead on focusing on skate tilting.
Just saying.
- Joined
- Feb 17, 2010
One way I think AI could be really helpful is to measure rotation and enable judges to make deductions on a more granular basis. A perceived 1% difference in rotation can mean a "perfect" jump goes to "q" along with the associated deduction in GOE, or a "q" jump going to "<". AI could allow these deductions to be more level, where those small differences can result in smaller differences in marks taken off. It would also allow a fully rotated jump to earn higher marks than one barely meeting rotation requirements.
- Joined
- Jun 21, 2003
I was alays impressed by Boyang Jin's technique on the quad Lutz. The opposite of flattening out, his outside edge continued to deepen all the way to launch. That was wonderful technique, and hardly anyone else could do it.
As for the curve, I don't mind a relatively straight and long lead-in. In fact. I prefer it to the short tight curve approach. I think it depends on whether we want the jump to appear to "come of nowhere," or whether -- the old-fashioned thing -- to advertise and showcase it.
Yes to graduated deductions. Also, to deductions being applied consistently. It really bugs me that < and << affect the base value, while q is reflected in the GOE. That makes no sense to me whatsoever. It seems especially convoluted because in general, the technical panel sets the base values and the judges give the GOE. Having a tech panel call reflected in GOE seems illogical and hard to implement.
- Joined
- Aug 24, 2016
Yes! Watching Boyang's 4Lz takeoff in slow-mo is an absolute joy.
- Joined
- Aug 24, 2016
Yes. I don't like the fact that < means lower BV and also lower GOE. Like why punish the skater twice? I think it would be better to just decrease the BV and give the freedom to the judges to decide whether the underrotation had a negative impact on the aesthetics of the jump. If the jump is well incorporated into the program, with good transitions and speed in and out AND the underrotation is not apparent in real time, why deduct GOE? Lowering BV is enough, imo.
Look, we talked about measurable parameters of a jump, didn't we? So, every jump has height, length, speed (entry speed, rotation speed, exit speed), but in the Lutz jump the skater also has to fight the torque - and the torque can be calculated based on the curvature. Humans can't do such calculations in real time but computers can; which opens a gate to how increase/decrease the difficulty and value of the Lutz jump.
Just saying.
Last edited:
- Joined
- Dec 9, 2019
The first video has normal speed in context, then motion picture, then slow motion:
- YouTube
Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
www.youtube.com
The second video has normal speed in context, then slow motion under different angles:
- YouTube
Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
www.youtube.com
- Joined
- Dec 9, 2019
I think that there would be two ways: either aiming at continuous scoring of rotation, which as you say would be granular; an advantage would be that the call would be exactly proportioned to the lack of rotation, and a potential advantage would be that it could be based on the total rotation in the air; the inconvenience would be that it would be hard to make it appear on score sheets and the public may lack trust because they wouldn't be able to calculate by themselves?
The other way would be to make more "slices". For instance, a jump lacking less than 30° wouldn't be called, lacking between 30 and 60° it would be called r (for rotation), lacking between 60 and 90° it would be called R, lacking between 90 and 120° it would be called u, lacking between 120 and 150° it would be called U, lacking between 150 and 180° it would be called <, lacking over 180° it would be << as today; with graduated penalties. I don't know if it would be good to have no base value reduction before the downgrade? Nowadays it seems that there's double sanction but most often it's deserved because the jump is far from perfect, which may be the main cause for the lack of rotation? To the opposite Shun Sato at the NHK Trophy got a wrong q call on his 4Lz which is far from being as beautiful as Boyang Jin's but still deserving a decent Grade of Execution, and he got a well negative Grade of Execution... Being on the high Base Value 4Lz, this wrong call and the too-automatic low Grades of Execution barred him from reaching the 100 symbolic bar; and I'm not speaking of his "restrained" Components... (less restrained than what they have been, I must say).
I believe that it would be a bit longer to determine clearly how to measure edges into jumps, after that of course "teaching" it to the software would be easy.
- Joined
- Aug 24, 2016
The first video has normal speed in context, then motion picture, then slow motion:
- YouTube
Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
The second video has normal speed in context, then slow motion under different angles:
- YouTube
Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
I know these videos very well!
- Joined
- Aug 24, 2016
That specefic 4Lz we are talking about is really one of the "hard to tell" cases. I'm sure @DizzyFrenchie didn't mean to say the q was unfair just because he "didn't like it". I'm not 100% sure the q was justified either.
- Joined
- Dec 9, 2019
I really think it's lacking really less than 90°, having seen it under two angles (one where it looked "hard to tell", one where the rotation angle was more visible).
I didn't think that Arigato was meaning me though, because we had further reasons than "just because (we) don't like" to contest the q claim. I supposed that she was speaking of other cases; of course, one can see more frequently omissions in calls than wrong calls.
Are we really in a position to override a judge? Yesterday someone claimed a skater did not deserve a q call. Now, I don't know what they were using to watch the event, but that skater's foot was so far away from my eyes on the ice, I could no more agree or disagree than the man on the moon. How the hell that person thought they were right, is nothing short of ludicrous. Furthermore, we do not know at what angle(s) the judges are viewing a skater compared to ourselves or what they can see on replay that we may not. I am mindful of that.
- Joined
- Dec 9, 2019
Oh, sorry, thanks to your latest post Arigato I have just realised that I had posted the slow motion videos under two angles on Shun Sato's Fan Fest, so you may not have seen them (although the first had been posted on the NHK Replay thread), here's the link to the page, it's the post n°238:
As you can see, there is no doubt that the jump was lacking less than a quarter of a turn.
Right, you are a Shun Sato fan and you post the slow-mo with "there is no doubt that the jump was lacking less than a quarter of a turn".
Was that jump deemed clean, the next day the slow-mo would be posted by somebody who is not a Shun Sato fan with "how can those lenient techs deem this clean when there clearly was a three-turn on the exit".
This is becoming ridiculous. RIDICULOUS. The endless fan complaints posed as "we want fair score" are becoming ridiculous especially because in cases like Sato's the score will be frickig the same q or not q.
The current rules have made it unclear if marred exits like Sato's should be punished by giving a q and then judges can give their remaining +GOE or there is no call and judges apply their -GOE but at the end of the day it's the same.
I need a break. I'm angry.
