Relative difficulty of types of jumps: triples versus quads

[Anna K.]

That is the scary part. If the infant/fetal Internet ever achieves self-consciousness, it will be insane, unable to distinguish truth from lies or even to understand that there is a distinction.

Even setting aside science-fictiony projections and Chicken-Little handwringing about Armageddon, as AI assumes an ever larger role it is a problem if the robots that we increasingly rely on have no means of deciding whether a statement is true or false.

Uhm... is there a thread in Le Cafe where we can discuss it?

All single jumps will be somewhat less than 360 degrees rotation in the air, except Lutz with counterrotated entry and of course axel--replace 360 with 540 in that case. It's part of the mechanics of the jump. How much less depends largely on the type of jump.

The entry edge curves, and the curve needs to deepen while the blade is still in contact with the ice to generate the rotation. That is normal technique, not cheating. It will happen somewhat more with multirevolution jumps because they need to generate more rotation.

At a certain point there could be excessive prerotation that should be penalized. To date the ISU has not defined that point except for "toe axels." It's considered a quality issue that can be reflected in GOE.

"Illegal" is not an appropriate term to use for quality issues, even those that are severely punished in base value as well as GOe like downgrades.
The rulebook uses the term "illegal" to refer to elements that are completely forbidden in competition and receive deductions if present. Like backflips until recently.

Even when skaters include elements in short programs that are not allowed in that disciplines short program at that level and therfore receive no base value, the official term is "not according to requirements."

Incidentally, the ballet jump called 540 (which skaters sometimes emulate in step sequences or choreo sequences) is not in fact 540 degrees in the air measured from the takeoff to the landing foot -- even the body rotation in the air is less than 540. Looks like it's usually a little more than 360 in the air, plus there is additional rotation of the body before and after.

Even in skateboarding, where the board has a front and back that can be used as reference points, the board does not make a full 540 degrees of rotation in the air:

- 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

And similarly with higher orders of rotation:

- 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

I know nothing about the technique, but I imagine the principle is similar: the very beginning and often the very end of the rotation has to occur while the board is in the process of leaving/returning to the ground.

I was waiting for somebody to say it

In ballet, the beginning and the end of the jump (of every move, actually) is defined by the body and face position towards the audience. Beginning facing the audience and the end facing the side? It counts as 540 regardless that the arabesque body position will be only partly turned to the side. But it doesn't matter because the purpose of the move is an aesthetic body position, not turning the core a certain degree.

In figure skating though, the beginning and the end of a jump is defined by the course of skating, which is a curve. Under-rotated means that the skater can't get back on his/her course smoothly. But skaters do not skate a straight line, so there is no way that a perfect jump can be exact 360/720/1080/1440 degrees measuring the take-off blade compared to the landing blade. But what would be a perfect jump rotation ratio in relation to the curve and optimal speed after landing? I don't mind if mechanical means of measurement are applied to find this out. It's interesting.

 

[Mathman]

Again your data is very valuable for long-term trends but I dont see how it was applied to effectively challenge my goal.

I don't think that Eppen is interested in challenging anyone's goal. Just in making a useful contribution to the topic of the thread, which invites discussion of the differences in apparent difficulty among different jumps, triples compared to quads..

 

[Blades of Passion]

With the flip the camera is perfectly place facing front on, on the landing the skate is completely flat on the ice running perpendicular to the camera.

No it isn't. You need to check your eyes or get a protractor. I'm now mad that you've made me waste my time needing to make better screencaps.

Kaori 3Flip takeoff, directly facing the back board:

3Flip landing:

That is more than 1/4 past the takeoff. Look at the front of the skate. It is not pointing exactly 90 degrees to the left. It is beyond that point. If you look at the jump from the side view, it can be seen there too:

The back of her skate is not pointing directly at the side board behind her, which is the 1/4 turn mark. It is past the point of being perpendicular to where she left the ice (facing the back board).

The 3T by any measure is a clear underrotation almost a downgrade.

Absolute lunacy.

This is her 3Toe takeoff:

3Toe landing:

That is very obviously past the 1/4 mark, unquestionably facing well beyond being perpendicular with the back board behind her, which is the direction she left the ice.

PLEASE stop writing misinformation about things you do not understand.

I don't understand this hostility to Petrosian when her 4T is clearly better than the 3F, the 3T, the 3Lz I posted.

"Hostility". All that matters is the factual rotation. Petrosian's 4T is 1/4 short. Kaori's jumps here are not.

How is that "3/4 turn pre-rotation"? Like genuinely wondering,

Her skate is facing directly away from the side board. The line you are drawing on the picture is wrong.

Most of your other screencaps are similarly crazy. Her skating foot has not launched off the ice in the picture you tried to make for that - her toepick wasn't even in the ice yet! Not that it matters, since the toepick foot is what needs to be looked at to see where the jump actually leaves the ice, but here are accurate screencaps:

First foot leaving the ice 1/4 turn into the takeoff

Toepick foot leaving the ice 3/4 turn into the takeoff:

Exactly as I said before, you can clearly see that Petrosian's toepick is directly facing the opposing side board of the rink when it leaves the ice (perpendicular to the short board behind her). From this point she does 3.25 rotations in the air, which is a 1/4 turn short jump, but her landing is completely "backwards" in relation to the short board behind her, so if you were to look only at where she is skating into the jump and then where she lands, it would superficially appear to be fully rotated. And sadly that is what most judges do.

 

[Arigato]

This is not happening.

 

[AxelLover]

Going back somewhat closer to the original topic: What's harder combo? +3T or +1Eu+3S? And what about +4T versus +1Eu+4S?

 

[Mathman]

I was informed by AY overwiewAnd what about +4T versus +1Eu+4S?

Has anyone ever done either of these? Maybe Hanyu could have done a 3A+4T,(?)

Edit: Just to make sure I Googled: Has Ilya Malinin ever done a triple Axel quad toe combination? I learned from AI overview, "yes, in multiple competitions" I am breathing a little easier about AI taking over the world.

 

[ramurphy2005]

Has anyone ever done either of these? Maybe Hanyu could have done a 3A+4T,(?)

Edit: Just to make sure I Googled: Has Ilya Malinin ever done a triple Axel quad toe combination? I learned from AI overview, "yes, in multiple competitions" I am breathing a little easier about AI taking over the world.

Ummm.....nope, the only people who have attempted 3A+4T in competition are Shoma Uno and Mikail Shaidorov, and only Shaidorov doing it successfully.

 

[Arigato]

Mikhail Shaidorov made history at the 2024 GP de France doing the first 3A+4T.

Skating Scores: Latest Figure Skating Results, Scores, Rankings & Statistics

The best resource for Figure Skating scores and stats on the web.

skatingscores.com

He repeated the 3A+4T at the 2024 Cup of China.

Skating Scores: Latest Figure Skating Results, Scores, Rankings & Statistics

The best resource for Figure Skating scores and stats on the web.

skatingscores.com

Upon Googling : Has Ilya Malinin ever done a triple Axel quad toe combination? The response was:

"No, Ilya Malinin has not been recorded performing a triple Axel-quad toe combination."

Upon Googling the same question using the correct spelling of Ilia's first name, the response was:

Yes, Ilia Malinin has performed a triple axel and quad toe combination in multiple figure skating competitions.



Competitions where Malinin performed a triple axel and quad toe combination:

• U.S. Championships
  Malinin performed a triple Axel and quad toe combination in a program that also included three other quads.
• 
  
• Grand Prix Final 2023
  Malinin performed a quad toe–triple toe combination in addition to a triple Axel.
  
  
• 
  
• World Figure Skating Championships 2024
  Malinin performed a quad toe loop–triple toe combination in a program that also included a quadruple Axel, quad Lutz, and quad Salchow.
• 
  The answer to the question is FALSE. At no time has Ilia Malinin done the combination.
  Jump history found here:
  
  

  Skating Scores: Latest Figure Skating Results, Scores, Rankings & Statistics

  The best resource for Figure Skating scores and stats on the web.

  skatingscores.com
• 
  
  Google AI is not infallible. I have found numerous mistakes using it for various reasons.
  

 

[4everchan]

Going back somewhat closer to the original topic: What's harder combo? +3T or +1Eu+3S? And what about +4T versus +1Eu+4S?

Ask Shaidorov I think he did try both this season (quad versions)

I think it depends but we cannot have a real answer again due to code of point.
The +3t combo is necessary in the SP so skaters work on this better... however, we tend to see more often skaters who do not have that combo manage to do a +3s euler combo in their free... so I tend to believe the euler combo is easier

 

[Blades of Passion]

I can't stand the google AI thing, it's wrong a lot of the time and wastes space at the top of search results, pushing actual articles down the page.

Going back somewhat closer to the original topic: What's harder combo? +3T or +1Eu+3S? And what about +4T versus +1Eu+4S?

I'd say it depends on the individual for the Triple combo. Most people have always practiced +3T more though, since it's needed for the SP. Would be interested to see the stats after like 5 years if +3Sal became allowed in the SP.

For the Quad, I have to think +4S is more difficult, with the loss of speed that happens from the half-loop beforehand. Swinging a triple from low momentum is very doable, comfortable even, but a quad is a different galaxy. Shaidorov's +4T was cleaner than his +4S.

 

[Mathman]

Yeah, I am really disappointed with this Google AI thingie. Where's the "I" (intelligence) part. All the program seems to do is spot a couple of key words and then search around to find some other words that people have often put with them.

If this is the best a big blowhard company like Google can do, they should go back to the task of recoloring their maps to show the Gulf of America and Red, White and Blueland.

 

[AxelLover]

The +3t combo is necessary in the SP so skaters work on this better... however, we tend to see more often skaters who do not have that combo manage to do a +3s euler combo in their free... so I tend to believe the euler combo is easier

This is my hypothesis as well. I suspect the +1Eu gives you more time to adjust and compensate (to an extent) for mistakes in the landing of the first jump. You can't afford to lose much speed doing the euler though.

 

[AxelLover]

For the Quad, I have to think +4S is more difficult, with the loss of speed that happens from the half-loop beforehand. Swinging a triple from low momentum is very doable, comfortable even, but a quad is a different galaxy. Shaidorov's +4T was cleaner than his +4S.

I hope we'll see more people doing these combos in the future so that we can collect some data.

 

[CanadianSkaterGuy]

	2005-06	%	2009-10	%	2015-16	%	2018-19	%	2022-23	%
Quads	144	3,1	152	2,9	538	7,8	743	9,8	773	10,7
3T	913	19,7	1054	20,2	1269	18,5	1312	17,3	1208	16,8
3S	589	12,7	673	12,9	760	11,1	789	10,4	737	10,2
3Lo	735	15,8	515	9,9	776	11,3	838	11,0	911	12,6
3F	799	17,2	973	18,7	1056	15,4	1187	15,6	1029	14,3
3Lz	917	19,8	1112	21,3	1412	20,5	1476	19,4	1468	20,4
3A	546	11,8	735	14,1	1061	15,4	1251	16,5	1083	15,0
	4643		5214		6872		7596		7209

Got curious about triples and their significance in men's comps and used my trusted friend https://fses.sakuraweb.com/ to collect data. That database includes mostly senior internationals, major juniors and Japanese nationals. Obviously only IJS data and I decided to use 2005-06 as a starting point because before that the number of competitions using IJS was smaller. And then chose the season before the big change in jump scoring, one from the middle of the 2010s (before the first real quad explosion) and the last season the database contains. Shows the actual count and then also percentage which are easier to compare.

Increase in quads and 3As and decrease in all the other triples except 3Lz. Suprisingly little in most cases.

What this made me think was also the significance of the BV. Since most men's six different triples (at least the elite ones) are pretty consistent, it is possible for them to prefer the ones which bring most points, particularly 3A and 3Lz.

But with quads it is a little trickier since even most elite men usually only have 1 or 2 of them. You can't really pick and choose which ones to learn, maintain them in consistent form and use in competition, you take the ones you hopefully can do. (Except if you are Ilia Malinin.)

The BV for all jumps in the beginning of the IJS was quite low, for example 8,00 for 4T and 8,50 for 4S. By the end of the 2000s, they were raised to 9,80 and 10,30 but this did not encourage anyone to jump more quads or try the more difficult ones with the hopes of getting all those points. The major rearranging of jump scoring to encourage risk-taking was accompanied by even higher BVs and these put together did make quads more lucrative than before. However, increases in numbers happened very slowly and very few skaters tried the big jumps with even bigger BVs until mid-2010s. And even then the big favourite was 4Lz and not one of the easier difficult quads or the 4A which would have been the obvious choice if BV was considered.

The huge increase in attempting all kinds of quads and learning multiple types has happened only after the TV for all quad types were lowered to current levels for the 2018-19 season. This happened maybe partly as a response to the repetition rules set at the same time and partly because it is perceived that more quads were needed to beat the elite skaters.

I would still argue that the types developed to competition consistency are not chosen simply by BV but rather by what the skater can do in the end.

E

This and the quads data are super interesting thanks for posting these.

It seems the loop is just not a jump skaters like. The attempt and success rate of a triple loop is significantly less than a triple lutz and more recently even less than a triple axel …. but I don’t think people could argue that means a triple loop is harder than a triple lutz - and certainly not a triple axel.

 

[Blades of Passion]

The two of you have serious issues with judging angles and generally have no idea what you're talking about. Seek the help of a geometry professor or something. I'm just going to respond to the most important things in that crazy spam list:

The skate is facing between 10 and 11 o'clock maybe slightly closer to 11 o'clock. This prerotated.

She pre-rotates 1/2 turn, as is standard. Her skate leaves the ice facing the back board.

You don't seem to understand the camera angle in relation to the rink (the camera at the start of her takeoff is not directly facing the back board) and also that the camera is moving between the takeoff and the landing. What matters is where the skate takes off and lands in the actual rink. You use the camera to mark those points.

The toe of the blade hit the ice earlier than this.

No it didn't. STOP LYING. She lands directly on the edge of her blade on the 3Flip, the toe wasn't pointed. Totally obvious to see in the slow-mo replay: https://www.youtube.com/watch?v=QL68-ltEQoM&t=270s

Defending this as a fully rotated combo is untenable.

What's untenable is your constant misinformation and inability to understand basic angles. You keep drawing all kinds of nonsense lines on these pictures. The picture shows that she is not landing before the 1/4 mark. Her skate was facing the lettering on the board behind her when it left the ice. If you can't see and understand that her skate is pointing at least 90 degrees away from where it left the ice, then you have a serious problem and need to stop writing about this subject until you fix it.

Here is an actual useful measurement picture of her 3Toe from the side angle, while she is still fully in the air, with both boards of the rink visible. THIS is how you measure rotation if you're trying to use lines, not the crazy crap you've been posting:

She's clearly nowhere close to << territory as you keep wrongly trying to assert, VERY past that line while still rotating in the air, and as we can see from the closeup of the landing from the other angle, her skate is definitely making it to this 1/4 mark.

Even more proof its off the ice before 180:

This gif literally shows that her toepick was still on the ice 180 degrees before her landing. Here is the pic of where she leaves the ice, if you are having trouble seeing:

Way over quarter. Shes begging for a downgrade with just a few more degrees here or there.

This right here shows how completely lost you are. Your takeoff pic showed the skate leaving the ice perpendicular to the side board, therefore the 1/4 turn mark is where the skate is at a 0 degree angle with the side board (aka, directly facing it). The landing here is very clearly past that, pointing to the left of the 1/4 turn mark.

Now lets forget Adeliia

How convenient. After being proven wrong, "lets ignore that."

-----

This is my hypothesis as well. I suspect the +1Eu gives you more time to adjust and compensate (to an extent) for mistakes in the landing of the first jump.

Eh, it goes both ways. Needing to do a euler from an unsteady landing, instead of being able to steady yourself with the toepick into a toeloop, is not something everyone is most comfortable with. And while some people can still do a sufficiently rotated triple after the euler from an initially wonky landing, the GOE of the combo should usually be impacted.

Even more importantly, if the euler between the jumps is too underrotated (we've seen a TON of these, many that are just step-outs from the previous jump, masquerading as a euler) then that shouldn't be considered a jump combo at all. It's just a jump sequence at that point.

 

[TallyT]

All these screencaps mean very little, you all know, since anyone can take a whole heap and then cherrypick (even if unintentionally) the ones that prove what they are arguing. What is pretty clear just for watching the actual whole videos (remember them?) is that jumps for many skaters are increasingly sketchy and even when efficient increasingly unbeautiful.

That's on the judges, the feds and ISU - basically officialdom - since let's be honest most stars in any competitive sport will do what they can get away with, especially where wins and therefore money is involved.

 

[gmyers]

This thread needs a cool down!!!!

It’s like doping plus war combined

 

[Mathman]

[T]he camera is moving between the takeoff and the landing.

This is the consideration that leaves me unconvinced by all these screenshots. The camera is continually moving, distorting the apparent angles against the background of the boards. I don't know how many degrees of freedom the robot arm has that the typical figure skating camera is mounted on. 6?

The same thing happens in spins. A spin might be perfectly centered, but the moving camera makes it seem like it is traveling. If you take a screen shot at 1:00 and another at 1:01 the background has shifted due to the moving camera. The stationary judges have a more reliable view.

 

[Blades of Passion]

This is the consideration that leaves me unconvinced by all these screenshots. The camera is continually moving, distorting the apparent angles against the background of the boards.

IMO it's not that hard to spot, but either way this is how calls are made in competition - looking at camera replay. So the eye needs to be well trained to gauge the angles there. Unfortunately the entire system of judging rotation is still quite formless in the rules and there isn't that much training required of the judges on this. It's hurt the sport more than it's helped, at this point.

 

[Mathman]

IMO it's not that hard to spot...[refering to estimating angles from various views]

I think it is quite a thorny undertaking, although it might be possible to get pretty good at it with enough practice.

I once did some consulting work with a group of orthopedic surgeons whose practice included scoliosis patients (sideways curvature of the spine, especially in adolescent girls). The procedure had been to take Xrays of vertebrae from front to back and monitor the angles of displacement. Then it was decided that it would be better to go back-to-front (PA) so as somewhat to protect developing breast tissue from excessive radiation. So to monitor a continuing patient they had to be able to convert angles on the old X-rays with the same angles, shot from a different perspective, on the new.

It turned out to be a startlingly challenging problem in three-dimensional projective geometry. Plus the fact that the subjects were human beings, and individual ones at that, rather than geometric abstractions, threw everything off. And this was merely the equivalent of shooting with a still camera from one side of the rink instead of the other. Add in complicated relative motions of both the skater and the camera. I do not think that superimposing two-dimensiona red lines on still photographs is as helpful as we might think.

That's why I think that if we MIST use technological aids to assist judges' eyes, the idea of motion sensors is more promising. Even if we view a jump in slow motion, or even if we study a single frame, we still must depend on our eyes to interpret what we think we see (as this thread attests).