General Yuzuru Chat

[Henni147]

This is a bit OT, but since it's linked to the topic of Yuzu's thesis, I'll post it here:

 

 

The pre-rotation of jumps is often linked to that infamous "full blade" toe assist and I wonder, if Yuzu addressed that issue in his thesis as well.

 

While pre-rotation of edge jumps happens on the take-off foot, toe jumps used to be pre-rotated on the free foot (blade "rolls off" after the toe assist).

With that full blade assist Flips and Lutzes are technically no clean toe jumps anymore, but become edge jumps. During the pre-rotation the weight of the skater is transferred from the take-off foot to the free foot, so the Flip/Lutz gets morphed into a Loop jump.

 

This is crucial, if you consider all the strict repetition rules of the current system. Skaters are only allowed to repeat one quad in the freeskate. A layout with three quad Loops would end up with the following base value:

 

4Lo 4Lo + REP 4Lo* Total BV

10.50   7.35   0.00 17.85

 

Now take a skater like Daniel Grassl, who uses a full blade assist on his 4Lz and 4F and excessive pre-rotation on all quads. At Italian Nationals 2021 his protocol looked like this:

 

4Lzq 4F 4Lo< Total BV

11.50  11.00   8.40 30.90

 

Even with one UR call his base value was almost twice as high as the base value of three clean and fully rotated quad Loops with correct technique.

So these skaters not only profit from the uncalled pre-rotation, but they can also sell the same jumping technique as three different jumps and avoid repetition invalidations, which has an even bigger impact on the final score than the pre-rotation itself.

 

That puts skaters like Yuzu at big disadvantage, because he has to master 2-3 different jumping techniques to be on par with these new morph quads. If the ISU seriously intended to introduce a bonus for skaters, who land all 6 jump types in the free, these skaters would even triple profit (they could camouflage 2-3 techniques as 6 and grab the bonus with minimum effort).

[Umebachi]

Sharing my comment on a tweet about Yuzu's thesis.  The original tweeter was referring to that tabloid article and noted that his thesis was not written in anger, but an expression of his love and dedication for the benefit of future skaters.   I said I agree with her, and that it's because he truly loves Figure Skating that we can feel the passion in his research.  (The google translation does a miserable job in translation here). 

 

 

This post has been tagged by yuzuangel as [NEWS].

[yuzuangel]

6 hours ago, Umebachi said:

 

WARNING:  Please be careful and don't give clicks to this article!  This is published in that notorious women's gossip magazine that stalked him.  The article misreads the thesis and ascribes incorrect meaning to certain terms: there is no "anger" in his thesis.   

you're right....i'll delete it

[SitTwizzle]

Side-by-side GPF and WTT 4A attempts, decomposition :

 

 

Simultaneous video of 3A and 4A (figure skating boards are 106.7cm high, of course there's the camera angle) :

 

 

The 6 attempts with slow motion? :

 

 

[rockstaryuzu]

10 hours ago, CiONTUw4A said:

 

I wonder how the western media will react to this. How more dedicated can a figure skater be to this sport?

I just don't think the significance of what he's saying here will be understood well. 

 

Think about it: Yuzu has just proven the most fundamental assumption in figure skating - namely that it's not possible to judge it objectively - to be flat-out wrong. If you can use motion-capture to see if a jump is according to the rules, then you can use it for spins, for steps, for anything. You can use it to automate tech scoring so that bias (in TES at least) can never be a thing in this sport ever again. 

 

That's earth-shaking stuff to think about. But I doubt most sportswriters will figure that out, western media or not western. And the main, non-sports media won't care.

 

 

 

 

[Whoopiewoop]

48 minutes ago, SitTwizzle said:

Simultaneous video of 3A and 4A (figure skating boards are 106.7cm high, of course there's the camera angle) :

 

 

 

 

If this is slowed down to the same speed (frames per second) then you can see how much of a beast the 4A would be by just looking at the speed of his rotation.

 

It's insane. 

[Yuzurella]

4 minutes ago, rockstaryuzu said:

I just don't think the significance of what he's saying here will be understood well. 

 

Think about it: Yuzu has just proven the most fundamental assumption in figure skating - namely that it's not possible to judge it objectively - to be flat-out wrong. If you can use motion-capture to see if a jump is according to the rules, then you can use it for spins, for steps, for anything. You can use it to automate tech scoring so that bias can never be a thing in this sport ever again. 

 

That's earth-shaking stuff to think about. But I doubt most sportswriters will figure that out, western media or not western. And the main, non-sports media won't care.

 

 

 

 

 

I'm afraid you're right. The sad truth is that those who are in power are the ones who profit the most from biased, unfair scoring. So, why would they even want anything to change. Fair, unbiased scoring based on AI and technology would take away the power of those who profit the most and thus also take away their profit. They wouldn't want that to happen.

 

If you ask me, it is indeed possible to make even a judged sport fair. Gymnastics used to have a lot of controversial judging and scoring, but after multiple (real) reforms, the system seems to be much fairer. But for this kind of reforms to happen, you need to have enough people in the feds and unions to be willing to do that. This doesn't seem to be the case in figure skating. Quite the opposite. Like I said, those who profit the most from biased scoring in figure skating are usually the ones in power and vice versa.

 

In figure skating, they even refuse to offer the judges slow motion reviews and such. Introducing new technologies and even AI? Laughable! Sadly, judging in figure skating is getting worse and is now as bad or even worse as during the 6.0 system. If it continues like this, the "sport" will lose more and more fans. Especially after Yuzu r******, who will still be willing to shell out thousands of dollars or Euros to watch pre-fixed "competitions" in this joke of a "sport"?

 

I think it's a bit of comfort that this ultimate level of corruption (aka the new GOE system) started after Yuzu had already achieved pretty much all of his own career goals in 2018. Of course he still got and gets disadvantaged and treated unfairly and was severely robbed on multiple occasions, but at least nobody can take away his earlier accomplishments.

 

But I do feel sorry for younger skaters like Rika or Kaori who started their careers years after Yuzu and might never get the medals, placements and recognition they deserve because of this corrupt system.

[SitTwizzle]

@rockstaryuzu & @Yuzurella

That's where we must be vocal. Those who refuse any AI scoring assistance (and transparency) are very likely trying to give unfair advantage to some skaters against others who skate better, this is very easy to understand. I don't believe IOC are a model of integrity but they can't really afford to be proven dishonest. That's the Anglo-Saxon notion of "respectability" : when in a function of power, you are "respectable" even if you are the most wicked and corrupted ever in this function, until you are publicly proven to be so, and then you become the last of the last.

 

Something quite different : I don't know if it has ever happened that a skater jumps all of 1A, 2A and 3A in a single open practice, but this is certainly the first one with 1A, 2A, 3A and 4A! The whole Axel family was introducing us 4A sama!

[EisElle]

9 hours ago, SitTwizzle said:

 I don't know how Rafael Arutyunyan views this thesis. After all, when it was known, it was still time to go back to a leeengthy ugly preparation into an awkward 3A, instead of the planned spread eagle into a << 3A. I suppose they hope this prerotation will be less huge by Olympics?

under spoiler for

Spoiler

idk if you know, but Lysacek is like kinda famous for his "3A" with backwards takeoff...and he won just fine. Maybe they're trying to channel his energy to replicate the feat

Though Nate, contrary to Lysacek, has the highest BV, which these days is all that matters (together with flag).

In 2010, instead, TraNstiOnS were all the rage (that, and, of course... flag). Ironical how the propaganda changes according to who USFSA has as their top man...

 

and who knows if now there are still Joe Inmans sending mails or whatsapping the judges to "suggests" how to score skaters....

 

 

What I love the most of Yuzuru's thesis --aside that as a scientist it's a development that I pretty much look forward to -- is that it's kind of a love letter to FigureSkating. He wants the sport to be the best it can be, and fair, consistent calls (starting from the highest BV elements) is the very first, fundamental step.

Any change will take probably decades (at the speed ISU implements new tech), but one day (if FS will not be dead yet) this kind of technology will be the norm in scoring, and people will marvel at how it was possible to score without that (though I bet certain fans will still be complaining about correct scoring taking away the "good ol' beauty" of the sport they used to know, just as they're nostalgic of 6.0).

 

Yuzuru would have had all the reasons to say f**k you all, I'm out, but he loves FS so deeply, not only for the expression value but as a competitive sport.

It probably pains him a lot to see ISU doind literally nothing to improve things with technology they already have and that it isn't exactly outrageously costly... (I mean, when a smartphone can give you a neat 4k superslowmo but tech panel has to rely on just one lower quality cam, there's something fundamentally wrong with the sport, and with its governing body that prefers to invest millions in their user-unfriendly website, podcasts, uninformed commentators, skating awards, songs and PR team that can't even disguise their bias instead of... decent scoring).

what Yuzuru had worked on (and that I'm sure he'll keep working on) will be part of his legacy for a better sport as much as his skating

 

Also, celebrating once again that he seem to have made peace with himself re: the scores he gets. Which is of course sad, he deserves a lot more (and at the very least basic respect, form the ISU), but let's remember that before 4CC20 he sounded very sad and kind of defeated, as he tried to make sense of everything, and it was heartbreaking to hear him doubt of himself to that extent.

The kuyashii vibes I got from him this whole season were very different. And I'm glad he can now look at himself and not be crushed by the ISU nonsense, and just keep following his path and smile and say "I still did well and I grew as a skater", because that's absolutely true.

His being more kuyashii for the "bad" 4A attempts than for any mistakes in his comp programs kinda says it all.

 

 

[Yuzurella]

 

I just discovered this today! Seeing the whole story behind the creation of Hana Wa Saku makes it even more amazing and touching.

 

 

Yuzu is the best thing that could've happened to figure skating. Period.

 

 

"Can the me of tomorrow be proud of the me today?"

He's such an inspiration!

[yuzuangel]

5 hours ago, rockstaryuzu said:

16 hours ago, CiONTUw4A said:

 

I just don't think the significance of what he's saying here will be understood well. 

 

Think about it: Yuzu has just proven the most fundamental assumption in figure skating - namely that it's not possible to judge it objectively - to be flat-out wrong. If you can use motion-capture to see if a jump is according to the rules, then you can use it for spins, for steps, for anything. You can use it to automate tech scoring so that bias (in TES at least) can never be a thing in this sport ever again. 

 

That's earth-shaking stuff to think about. But I doubt most sportswriters will figure that out, western media or not western. And the main, non-sports media won't care.

 

TBH it's not that hard or earth-shattering, other sports do SO MUCH more.

But those from western feds + the media will just play dumb and pretend it can't exist, or that it hurts the "beauty" of figure skating which can't be measured using computers or something, as if calling underrotations or wrong edges fairly will make skating uglier. Of course, the people who have "beautiful" skating that will be hurt the most from this all happen to be from one or two feds...

 

The only change that will happen is if the US and Russia both have a leading lady and those two feds can try to out-politick each other and propose technologies that highlight each other's weakness. Otherwise things are all fine and dandy and there's no need to change.

[rockstaryuzu]

5 hours ago, Yuzurella said:

Fair, unbiased scoring based on AI and technology would take away the power of those who profit the most

*cough*RusFed*cough*

 

Granted, it would still only be TES scores that would become more fair. PCS would still be abused wantonly for the sake of national bias. 

[Melodie]

18 minutes ago, rockstaryuzu said:

*cough*RusFed*cough*

 

Granted, it would still only be TES scores that would become more fair. PCS would still be abused wantonly for the sake of national bias. 

It can help a bit with PCS IMO, SS is most likely quantifiable. But I agree with @yuzuangel, unless the ISU itself is changed, technologies, no matter how advanced and people, no matter how dedicated, won’t change anything 

 

Edit: actually, the rest of PCS can be quantified too, albeit some less than others (so will still need the human factor, but bias can be reduced). 

[rockstaryuzu]

5 hours ago, SitTwizzle said:

@rockstaryuzu & @Yuzurella

That's where we must be vocal. Those who refuse any AI scoring assistance (and transparency) are very likely trying to give unfair advantage to some skaters against others who skate better, this is very easy to understand.

 

2 hours ago, LadyLou said:

u

  Reveal hidden contents

idk if you know, but Lysacek is like kinda famous for his "3A" with backwards takeoff...and he won just fine. Maybe they're trying to channel his energy to replicate the feat

Though Nate, contrary to Lysacek, has the highest BV, which these days is all that matters (together with flag).

In 2010, instead, TraNstiOnS were all the rage (that, and, of course... flag). Ironical how the propaganda changes according to who USFSA has as their top man...

 

and who knows if now there are still Joe Inmans sending mails or whatsapping the judges to "suggests" how to score skaters....

 

 

What I love the most of Yuzuru's thesis --aside that as a scientist it's a development that I pretty much look forward to -- is that it's kind of a love letter to FigureSkating. He wants the sport to be the best it can be, and fair, consistent calls (starting from the highest BV elements) is the very first, fundamental step.

Any change will take probably decades (at the speed ISU implements new tech), but one day (if FS will not be dead yet) this kind of technology will be the norm in scoring, and people will marvel at how it was possible to score without that (though I bet certain fans will still be complaining about correct scoring taking away the "good ol' beauty" of the sport they used to know, just as they're nostalgic of 6.0).

 

 

 

 

 

Well, the true caveat here is that while Yuzu has proven that it's feasible, taking motion capture and AI-type technology from a research environment and making it into a robust and easy-to-use system that can be deployed at competitions around the globe by the average end user (that is to say, any Luddite fossil) for a reasonable cost and upload results in real time, is a whole other ball game. It will probably happen but I expect it to take at least ten years, maybe more. 

 

 

47 minutes ago, yuzuangel said:

TBH it's not that hard or earth-shattering, other sports do SO MUCH more.

 

Precisely my point - in figure skating, it is a 'mind blown' moment. Consider how many actual figure skaters competing today obediently parrot the party line of 'you can't be objective in figure skating". Consider how hide-bound and resistant to change the sport is. 

 

To play devil's advocate for a moment though, using motion-capture technology to judge technical elements might just encourage the ISU to go even further down the road of having a 'technical' program and an 'artistic' one instead of SP and FS. If they decided to adapt technology-based judging at all, the so-called 'technical' program is where I'd expect to see it come up. 

 

[Melodie]

 

 

The evolution of THAT part in LMEY. Yuzu really killed it with the WTT encore