Women's single skating at the ISU Grand Prix stages: statistics from 2021 to 2023

[Miller]

I actually like the 100%. It is compatible with the whole philosophy of the IJS, which says: you get 3.30 base valuie points for doing a double Axel. Period. If you do a solo double Axel you get 3.30 points. If you do 2A+3T you get 3.30 points for the 2A. If you do 3Lz+2A+2A you get 3.30 points for the first 2A and 3.30 points for the second.

True, some double Axels are more difficult than others. A solo 2A with intricate preceding steps is more difficult than a solo 2A without intricate preceding steps. This is rewarded elsewhere (GOE, Composition, Skating Skills), but the base value is still that good old 3.30.

As for the Russian Ladies effect, it's just common sense. If you take away the best jumpers then who is left are the second-best jumpers. Maybe one day we will decide to beat our swords into plowshares and our spears inbto pruning hooks so that sporting events can return to bornal.

I think the problem with this is the difficulty of doing combinations, or at least some of them.

For example you'll often get people mentioning the difficulty of 3Lz-3Lo combinations and why people don't do them.

For example if you've got 2 skaters, one doing a 3Lz-3T and solo 3Lo, and the other doing a 3Lz-3Lo and solo 3T, then the base value is exactly same, but from the number of skaters doing the former, but not the latter it must be much harder to do the 3Lz-3Lo combination than the 3Lz-3T one, and the ease of doing the 3T doesn't make up for the difficulty of doing the 3Lo - I would also add at this point that's there's a GOE boost for the first set of jumps in that the GOE for a combination is based on the highest value jump in the combination, hence the max for the 3Lz-3T is identical to that for the 3Lz-3Lo, but then the first skater can get more for the 3Lo (up to 0.35 points) than the second one can for the 3T. So there's even further incentive to do the 3Lz-3T rather than the 3Lz-3Lo.

When it comes down to the +2A sequences I think the reason why they've become so common is that the skater gets a chance to steady themselves a bit after the first jump in the sequence so the relative ease goes down compared with doing the 2nd/3rd jumps of a combination straight away, but the reward in terms of base value is exactly the same, hence why skaters are doing it.

Solution, or at least my proposed one - increase the base value for the 2nd and 3rd jumps performed in combination (but not the first one, it's no different as far as the skater is concerned compared with a solo jump of the same type, though of course there is the increased stress of knowing there's a big combination coming up rather than a solo jump). While for sequences reduce the base value of the jump done in sequence, but yet again not the first one, why should this be reduced when you think about it.

Such a thing would reward harder combinations more, punish easier sequences, but not too much, plus also go some way to reducing the GOE advantage of doing lower value jumps, e.g. 2A's and -3T's compared with -3Lo's, in sequence/combination. For example in the Ava Marie Ziegler layout quoted by Rebecca Moose above, Ava does 7 jumping passes each of which has its GOE based on a triple because her 2As are hidden away in her 3Lz+2A+2A sequence. Hence her Base Value for GOE purposes goes up by 2 or 3 points (I'd have to check the exact figure) compared with some one who say does 5 jumping passes led off with a triple and 2 solo 2As.

 

[Mathman]

I think the problem with this is the difficulty of doing combinations, or at least some of them....

For myself, I would not be eager to see the IJS go in the direction of listing every possible thing that a skater can do on the ice and then try to assess it's relative difficulty. The scale of values already lists base values for 174 different kinds of jumps, followed by base values for 179 different kinds of spins. Another few hundred pages listing every possible jump combination separately, what would that be, 174x174 = 30276 more lines?

What difficulty would we assign to a 2Lz + other direction 2Lz combo, for instance -- a cascade that is so hard that I don't think anyone has ever done it. Kristi Yamaguchi found the Salchow to be a great challenge (and in fact she flubbed her 3S at the 1992 Olympics which might have opened the door for Midori Ito, had Ito not already been out of contention for the gold before the LP even began.) I once saw Shizuka Arakawa in exhibition do a sequence of 23 double and single loops. What score should we assign to such a prodigy of endurance?

Michelle Kwan used to train two versions of her LP. If she hit her 3T-3T combination at the beginning, then her last element was a split jump (base value = 0 under the IJS). If she missed her triple/triple, then she threw in an axtra 3T instead as the last element to bring up her triple count. The split jump version was rewarded in the second mark. In her iconic program to East of Eden at the 1998 World Pro, her most spectacular elements were (1) a frontward/backward change edge spiral sequence that lasted 14 seconds on one foot, and (2) a falling leaf (right on the pregnant musical pause preceeding the climatic cymbal crash) into a inside/outside spread eagle. Total base value under IJS, 0. (And in fact the only flaw in the whole program was a slightly unsteady edge on her triple Lutz landing.)

I basically am not a fan of turning the sport of figure skating over to the bean-counters (much as I lke counting beans nyself ).

 

[Jontor]

For myself, I would not be eager to see the IJS go in the direction of listing every possible thing that a skater can do on the ice and then try to assess it's relative difficulty. The scale of values already lists base values for 174 different kinds of jumps, followed by base values for 179 different kinds of spins. Another few hundred pages listing every possible jump combination separately, what would that be, 174x174 = 30276 more lines?

What difficulty would we assign to a 2Lz + other direction 2Lz combo, for instance -- a cascade that is so hard that I don't think anyone has ever done it. Kristi Yamaguchi found the Salchow to be a great challenge (and in fact she flubbed her 3S at the 1992 Olympics which might have opened the door to Midori Ito, had Ito not already been out of contention for the gold before the LP even began.) I once saw Shizuka Arikawa in exhibition do a sequence of 23 double and single loops. What score should we assign to such a prodigy of endurance?

Michelle Kwan used to train two versions of her LP. If she hit her 3T-3T combination at the beginning, then her last element was a split jump (base value = 0 under the IJS). If she missed her triple/triple, then she threw in an axtra 3T instead as the last element to bring up her trpile count. The split jump version was rewarded in the second mark. In her iconic program to East of Eden at the 1998 World Pro, her most spectacular elements were (1) a frontward/backward change edge spiral sequence that lasted 14 seconds on one foot, and (2) a falling leaf (right on the pregnant musical pause preceeding the climatic cymbal crash) into a inside/outside spread eagle. Total base value under IJS, 0. (And in fact the only flaw in the whole program was a slightly unsteady edge on her triple Lutz landing.)

I basically am not a fan of turning the sport of figure skating over to the bean-counters (much as I lke counting beans nyself ).

I get what you are saying, but I think the problem (if there is a problem?) is fundamental. It is easier to push the technical mark as a skater because that is something you can score yourself. If you end a free program with a 3T, that would score somewhere about 5.3 points. But if you end the program with a 3Lz then it would score about +2.50 points more. THAT is something you train for, and you know the results.

On the other hand, if I want to better my second PCS score, how do I do that? You can think all you want that you are getting more artistic but the judges might not. The subject view of the judges (and sometimes politics) is the biggest factor here. And I as a skater I can do nothing about that.

If we really want changes, we need to rework the IJS judging system at the very basic principles. As it is now, 75% of the points you get are de facto technical. Do we want that?

I love the examples that has been mentioned before as unique artistic elements, like the delayed 1A that is worth nothing today. I was also thinking of Lucinda Ruh, the mother of all spins, who sometimes did spins that changed shape all the time, she didn't hold her positions, she transformed them into something else and she would go on forever. That would also be worth 0 points today.

 

[Dawn825]

I seriously crave more original combos. I don't care about the points, I want see more triple loops, flips and salchows as the second part of combos. They're so exciting. I cannot make myself care about a double toe double loop, whatever it's worth.

 

[Mathman]

I seriously crave more original combos. I don't care about the points, I want see more triple loops, flips and salchows as the second part of combos.

The +flip and +Salchow sequence was a problem for years because of being on the wrong foot/edge after the first jump. When they first came out with the +euler+S or F sequence (for less value back then than a true combo), the skaters weren't used to it and the awkwardness of the euler just made the whole thing look like a mistake/step-out.

For me, Patrick Chan was the first skater who actually got the technique down (the key is the deliberateness of the step and landing of the euler, IMHO) and made it look cool. Anyway, that's why I like the 100% rule that makes these sequences worthwhile pointwise.

 

[Mathman]

If we really want changes, we need to rework the IJS judging system at the very basic principles. As it is now, 75% of the points you get are de facto technical. Do we want that?

In my opinion, it is what it is. It desn't really matter what we want or don't want. I don't see any possibility of turning the clock back to an earlier version of what used to be. I for one am OK with it. We can still look at tapes of memorable performances of yore if we like, while also appreicating the new run-and-gun kids.

Here's Lucinda Rue in competiton. Whatever kind of spinnerade it is that they put in the water in Switzerland, Denise Biellmann had enough left over to share with Lucinda and then with Stephane Lambiel.

Dick Button's commentary here says it all.

Lucinda Ruh - 1999 Worlds LP - Gliere Harp Concerto

Lucinda Ruh of Switzerland skates in the long program at the 1999 World Figure Skating Championships in Helsinki, Finland. Ruh ends up in 13th after placing...

www.youtube.com

 

[gkelly]

Whatever kind of spinnerade it is that they put in the water in Switzerland, Denise Biellmann had enough left over to share with Lucinda and then with Stephane Lambiel.

Not to mention Nathalie Krieg

 

[4everchan]

Not to mention Nathalie Krieg

I adored her spins !!!! Still think about them.

 

[Mathman]

I think this might be Nathalie Krieg's best competitive performance. She got through all her jumps without any major errors. Could have made better use of the music in the first part, but presentationwise she made up for it by spunk and verve. (Well of course she did. She was 15 and the world was her oyster.) And those spins.

NATHALIE KRIEG 1993 EUROPEANS FP

For the spin lovers

www.youtube.com

 

[Miller]

For myself, I would not be eager to see the IJS go in the direction of listing every possible thing that a skater can do on the ice and then try to assess it's relative difficulty. The scale of values already lists base values for 174 different kinds of jumps, followed by base values for 179 different kinds of spins. Another few hundred pages listing every possible jump combination separately, what would that be, 174x174 = 30276 more lines?

What difficulty would we assign to a 2Lz + other direction 2Lz combo, for instance -- a cascade that is so hard that I don't think anyone has ever done it. Kristi Yamaguchi found the Salchow to be a great challenge (and in fact she flubbed her 3S at the 1992 Olympics which might have opened the door for Midori Ito, had Ito not already been out of contention for the gold before the LP even began.) I once saw Shizuka Arakawa in exhibition do a sequence of 23 double and single loops. What score should we assign to such a prodigy of endurance?

I wasn't really proposing all combinations of combinations to be in the scale of values, that would be crazy. What I was proposing was something like a 3Lo done as the 2nd part of a combination would always be worth something like 6.0 points, and the overall combination would take care of itself e.g. a 3Lz-3Lo would be 11.9 points, a 3T-3Lo would be 10.2. That way you wouldn't need masses and masses of new lines.

Of course there might be laws of unintended consequences taking place and you'd have to really think what skaters might do to exploit the new rules (as has occurred with +2A sequences IMO). But get it right and perhaps you could have opposite way -2Lz's, plus didn't Artur Dmietriev Jr do a 3Lz-3F combination a few years ago which no one else has attempted since then, but if the -3F was worth say 6.5 points compared with 5.3 might'nt more people try it?

 

[Magill]

In my opinion, it is what it is. It desn't really matter what we want or don't want. I don't see any possibility of turning the clock back to an earlier version of what used to be. I for one am OK with it. We can still look at tapes of memorable performances of yore if we like, while also appreicating the new run-and-gun kids.

But reforming IJS does not mean turning the clock back. Quite to the contrary, it means responding to challenges of today while facing forward and moving on with the times. Otherwise, no change would have ever occurred in the first place and we would still be back in the figures and 6.0 system today.
Whatever system has been introduced once, it is bound to be changed with time. That's life. We bid farewell to 2023 and welcome 2024. We bade farewell to ordinals once and welcomed IJS. Now, one day we will say goodbye to today's IJS and we will see a new version of it or a completely new system. That's how it works, whether we like it or not.
I am pretty sure FS fans from the future will look back on today's judging system, with its subjective assessment of perfectly measurable values, with its technological backwardness and giving priority to the power of human eye over technology, with a mixture of amusement and disbelief. The only question is when these changes will occur and how many more generations of skaters will be judged the way it is done now. IMHO, it is not "will it change?" but "when?" (and obviously "to what?").

 

[Mathman]

In my opinion, it is what it is. It desn't really matter what we want or don't want. I don't see any possibility of turning the clock back to an earlier version of what used to be. I for one am OK with it.

But reforming IJS does not mean turning the clock back. Quite to the contrary, it means responding to challenges of today while facing forward and moving on with the times.... The only question is when these changes will occur and how many more generations of skaters will be judged the way it is done now. IMHO, it is not "will it change?" but "when?" (and obviously "to what?").

Aren't we saying the same thing here? We can't go back. I am Ok with that.

As to the direction that changes may take, I think that we can make some guesses. One guess that I would hazzard is that scoring will continue to place the most emphasis on "tech" and especially on the subcategory "jumps." My other guess is that we will see improvement in measuring devises -- better clocks and yardsticks. Skater X will be clocked at so many meters per second across the ice, and his quad Lutz will come in at a height of so many centimeters with such-and-such distance covered, with so many degrees of rotation in the air and so many degrees of of angle to his take-off edge.

Can we go farther? Should we? Can we imagine an AI program (perhaps programmed by a master AI program without human intervention) that gives an evaluation of "variation of speed and energy reflecting the character of the music and the overall shape and intent of the choreography." Would this be a good thing?

 

[gkelly]

Can we go farther? Should we? Can we imagine an AI program (perhaps programmed by a master AI program without human intervention) that gives an evaluation of "variation of speed and energy reflecting the character of the music and the overall shape and intent of the choreography." Would this be a good thing?

I think it might be possible for technology to measure the "Pattern and ice coverage" aspect of the current Composition component, approximately as soon as it will be possible for technology to measure and evaluate the execution of elements.

But I think that AI is still very far away from being able to evaluate qualitative aspects of performance as well as trained humans can, let alone better. And that includes technical aspects of skating quality as well, such as Clarity of edges, steps, turns, movements and body control; Balance and glide; Flow. Anything that involves understanding cultural considerations about understanding movement to music would be even further away. Not in my lifetime.

Nor do I foresee the ISU deciding that quality shouldn't count and that only what is quantifiable should determine results. That would result in a completely different sport.

 

[skatingguy]

I am pretty sure FS fans from the future will look back on today's judging system, with its subjective assessment of perfectly measurable values, with its technological backwardness and giving priority to the power of human eye over technology, with a mixture of amusement and disbelief. The only question is when these changes will occur and how many more generations of skaters will be judged the way it is done now. IMHO, it is not "will it change?" but "when?" (and obviously "to what?").

Maybe, but it's not like when electronic timing was introduced to swimming, and they didn't trust it so they awarded Olympic medals based on the judgement of the people calling the lines, and so the winner was not the faster swimmer. The technology exists, but it's still pretty expensive. Electronic line calling in tennis is available, but most professional tournaments still include lines people because that's the cheaper option. If the video review technology gets lot cheaper, and easier to set up then we might see it in skating. It would be interesting to see the 3D recreation of jumps similar to what we see in tennis line calling, or VAR in football.

 

[Mathman]

I think it might be possible for technology to measure the "Pattern and ice coverage" aspect of the current Composition component, approximately as soon as it will be possible for technology to measure and evaluate the execution of elements.

It shouldn't be impossible to measure things like spacing of elements, steps and turns, and evaluating the extent to which the skaters covers the whole surface. It would be a challenge of a different sort, I would think, to decide whether all this adds up to a coherent and aesthetically/musically purposeful program.

It would be interesting to see the 3D recreation of jumps similar to what we see in tennis line calling, or VAR in football.

Vary interesting, yes. But it might turn out to be more like a CAT scan in medicine. Yes you can get a nice 3D reconstruction of the patient's liver, but the subject has to lay still for half an hour while the machine grinds away at it.

 

[skatingguy]

Vary interesting, yes. But it might turn out to be more like a CAT scan in medicine. Yes you can get a nice 3D reconstruction of the patient's 's liver, but the subject has to lay still for half an hour while the machine grinds away at it.

That's very true. At the moment the use of AI in sports is limited to tennis, where it tracks the ball, and soccer where it tracks the ball, and off-side calls. That's not nearly as complex as the movements of skater moving through the air with various body parts moving in different directions at the same time.

 

[Alex Fedorov]

wrt to the current womens field, I'd like to see more creative jump layouts. too many skaters are attempting 3-3s that are not consistent or rotated. I want to see rotated jumps with good technique. obligatory shoutout to my girl Ava Marie Ziegler, who I think has the smartest LP jump layout I've seen all season:

3Lz+2A+2A
3F+2T
3Lo+2T
3S
3Lz
3F
3T

better base value here than many skaters trying 3-3 in the LP!

Ava Marie Ziegler competed in only one Grand Prix stage this season (NHK Trophy). She performed the sequence 3F+2A+2A, not 3Lz+2A+2A. You were probably a little mistaken (or maybe you were talking about some other tournament). Her jump set could be improved significantly. The first thing that comes to mind is to move the sequence to the second half of the program and replace the last solo jump (3T) with 3Lo. As a result, the program's base value will increase from 63.44 to 65.33.

Also, I can dream a little and replace both 3+2 combinations (or at least one of them) with 3+3. Provided that the solo 3T is replaced with 3Lo, this is quite acceptable. In this case, the BV will increase by another 2.9 + 2.5 points and reach 70.73.

Well, in my wildest dreams, instead of 3S, 4S will be implemented. And then the BV will reach 76.13. If we return from fantasy to reality, probably 3S in this program can be used as the basis for a “plan B” in case of failure at one of the cascades. This is exactly how Loena Hendrickx uses this jump.

 

[Alex Fedorov]

As for a performance like Curry's what I admire about it is that the jumps might be doubles, but they are placed in the choreography as "juist right" musical exclamation points. They enhance the program without being the program.

This very important detail gives me another reason to say that the change in the rules for performing sequences was not thought out enough. The fact is that the sequence 3+2A+2A is not perceived as a single element. It's more like three separate jumps. And it’s hardly possible to increase impact of some musical accent with these three time-stretched jumps. At the same time, 3Lz+3Lo is like a lightning strike, everything is very fast and there is no doubt that this is exactly one element, and not two separate jumps. It is no coincidence that this cascade looks so spectacular. But now skaters have no motivation to study this element.

 

[Alex Fedorov]

The number of +2A sequences has increased dramatically now that it's 100% of the base value rather than 80%. This may explain the increased number of solo 3S and 3T jumps as skaters may not be doing two -3T combinations, or a -3T and a -1Eu-3S one, anymore. Instead they're doing a -3T and a +2A sequence which allows them to do their maximum of 7 non 3A triples and two 2As in an easier manner, so one of the -3S or -3T combination jumps becomes a solo one. I agree with Rebecca Moose just above who says that Ava Marie Ziegler has the best jump layout going to exploit what seems the relative ease of doing +2A sequences. 2 goes at a 3Lz and if you land one do 2As after it, then your other combinations are both -2Ts which must be relatively easy for skaters at this level. I would be interested to know if the number of -1Eu-3S combinations has dropped away dramatically as skaters find easier ways of doing the same technical content in terms of Base Value - it certainly feels that way.

The increased number of -3T attempts for non-Russians does seem somewhat counter-intuitive bearing in mind the above. However you need to remember there's now 12 non-Russians at each GP rather than the 9 or 10 of before. This probably also explains the number of -2T combinations. This will be because the skaters replacing the Russians will not be as strong, and so more -2T jumps are to be expected.

Finally I'd say the increase to 100% for +2A sequences has gone too far. They've become for too ubiquitous IMO, plus the number of 'real' 3-3 combinations has reduced (taking account of my comments about more non-Russians competing on the GP). It is somewhat cool however to see +3As in the Men's though. Overall I'd say 90% is about right, more incentive to do them, more variety in programs, plus more of the old jump combinations should re-appear.

the number of cascades 3+еu+3 has really decreased sharply. In 2021, there were 20 such combinations (or 25, if you take into account 2A+еu+3S). In 2022, 11 attempts were made, of which 4 were 2A+eu+3S. In 2023 there will also be 11 attempts, of which 2 are 2A+eu+3S and one is 2A+eu+3F. But here it is impossible to say what exactly caused the sharp reduction in the number of such combinations, because in 2021, out of 25 attempts, 13 were made by Russian figure skaters who did not take part in these competitions in subsequent years.

Generally speaking, sequences were not supposed to displace 3+3 cascades. There are several good reasons for this:

1) The sequence cannot be performed in a short program.
2) The maximum base value of a sequence of two jumps (3Lz+2A) is 9.2. This is less than the value of the two most popular combinations 3F+3T (9.5) and 3Lz+3T (10.1).
3) Only one sequence can be performed in a free program.

Thus, it would be more logical if skaters replaced the usual 3+2 combinations with “short” sequences. Or if the combination 3+2+2 were replaced with a much more profitable “long” sequence 3+2A+2A. But here we have to admit that nothing like this happens:

1) The number of conventional 3+2 cascades is not decreasing, but increasing (and this is accompanied by a decrease in performance efficiency).
2) The number of “short” sequences in 2023 compared to 2022, on the contrary, decreased.
3) In 2023, only one sequence 3+2A+2A was performed (the same already mentioned sequence from the Ava Marie Ziegler program).
4) The number of “regular” 3+2+2 cascades has increased from 16 in 2021 to 31 in 2023. The efficiency of performing these elements has decreased significantly - but for some reason the skaters do not try to switch to “long” sequences.

 

[Jumping_Bean]

Ava Marie Ziegler competed in only one Grand Prix stage this season (NHK Trophy). She performed the sequence 3F+2A+2A, not 3Lz+2A+2A. You were probably a little mistaken (or maybe you were talking about some other tournament). Her jump set could be improved significantly. The first thing that comes to mind is to move the sequence to the second half of the program and replace the last solo jump (3T) with 3Lo. As a result, the program's base value will increase from 63.44 to 65.33.

Also, I can dream a little and replace both 3+2 combinations (or at least one of them) with 3+3. Provided that the solo 3T is replaced with 3Lo, this is quite acceptable. In this case, the BV will increase by another 2.9 + 2.5 points and reach 70.73.

She can't replace the solo 3T with a 3Lo, nor can she do a +3T unless she gives up one of the +2As or does a double or quadruple to replace the solo 3T. Skaters are only allowed to repeat two triple jumps and Ava already does that with two 3Lz and two 3F.
GOE-wise, this layout is the best thing she can do, as she "hides" all doubles in combinations with higher-value triple jumps.

The only thing she can do is move all of the combinations to the second half, but with Ava's history of inconsistency, I would not recommend it to her at the moment.