Dear ISU members, do you realize the idiocy you just voted on? In this way you have achieved the result that you had set for yourselves, simplifying the work of judges… destroying the discipline that judges should judge and that you should promote.
Is judging a figure skating competition difficult? Yes, I too know that. This means that the work of judges must be simplified, but without destroying the discipline. I’m not kidding. Do you remember that until the 2017-2018 season the jump not in combination in the short program had to be preceded by a step? The request has been removed. We now have jumps preceded by runs of 10 or more seconds that can take the same score as jumps preceded by complicated sequences of steps. This does not make figure skating more beautiful, it makes it poorer, because in order not to risk making a mistake in the jump, it is better to take a long run. Only someone crazy, or absolutely in love with the discipline he practice and willing to lose medals that he could win, in order not to accept compromises that would undermine the artistic integrity of the program, could decide to continue to perform the steps before the jumps. I don’t know many skaters like that. Only one comes to my mind.
From the evaluation of the PCS you have removed all the requests that made the program rich. Speed is now valued, not the ability to change speed. Did you know that speed skating already exists? In Beijing, medals were awarded in 14 disciplines in speed skating, so those who love speed already have their own sport. The athletes have it, and the spectators have it. There is no need to make a bad copy of an existing discipline. Nor is there a need to impoverish an existing discipline. When in gymnastics they decided to give more space to acrobatics, they did not destroy the sport they had, alongside artistic and rhythmic gymnastics they created trampoline competitions (which is also an Olympic discipline) and acrobatic gymnastics. Do you want speed competitions or jumping competitions? You have already speed competitions, and you can organize jump competitions, without destroying figure skating.
Were there no other solutions? If you really had no idea what to do, you could always go back to the +3/-3 system, which was better than the +5/-5 system. I know, it’s a difficult thing. Recognizing that you are wrong is never easy. You need to have a greatness of spirit that not everyone has, in addition to the intelligence to understand that you have made a mistake and the desire to do the right thing and not the simplest one. But at least you don’t need to keep making mistakes after mistakes.
Let’s think about some possible changes. Changes that can be implemented now, and that only for the dimension of the jumps, for which more data needs to be collected, should be made at least for the ISU Championships, Olympic Games and Grand Prix competitions by the 2024-2025 season, while the Challenger series could be given more time to adjust.
Technologies exist, let’s use them. You can measure all the jumps in all the major competitions for the next two years, in order to identify for each jump (toe loop/salchow/loop/flip/lutz/axel, single/double/triple/quadruple, Men’s/Women’s competition senior/junior competition) height and average length. From that number it is possible to establish precise numbers beyond which the jump deserves bullet 1, and under which it deserves the deduction because it is too small. Bullet and deduction assigned not by the judges but by the computer, which can measure the jumps in real time. In this way the judges have to think to one bullet less.
Since we are at the technologies… the technology evaluates the rotation of the jumps. Both underrotated and prerotated jumps. It can be done with video systems that allow you to see what the skaters did at the highest quality. Let’s look at the Beijing videos, for example. In the replays, some jumps slowed down a lot, to the point of leaving no doubt about what the skaters did. For his triple axel in the short program, for example, Yuma Kagiyama did the take off when he is already backward.
At page 20 of the TP Handbook 2021-2022 there is this passage:
Of course, the indication that the take off can only be watched in regular speed removes the possibility of see the cheated take off and to apply the rule. Here, I would replace the word regular with slow motion. To write a rule that can’t be applied is a non-sense. What interests us is the fairness of the competitions, right? That will be the best to win, and not the smartest. And after all, if the tech panel can watch the flip and lutz edges in slow motion, why can’t it watch the take off in slow motion? Let’s imagine the technical panel looking at one of these two jumps to determine if the edge is correct. Watch in slow motion, decide whether or not to make the edge call, then forget what he saw and evaluate the correctness of the rotation at take off. If someone made a rule like this, this guy would be ridiculous, but you’re not ridiculous, are you?
There is also another problem, linked to understanding when a prerotation is such as to be able to speak of a cheated take off. We need a more precise definition of how much a jump must be prerotated, evaluated by technology, instead of leaving everything to the approximation. And if there can be doubts on evaluating Kagiyama’s axel, on the 2T of the 3A+2T combination in the FS the prerotation is visible even at regular speed. We can discuss the take off of that axel, the toe loop take off is certainly cheated.
A very accentuated slow motion can also allow us to see when a jump is underrotated on landing, like the 3T of the combination performed by Nathan Chen in the SP in Beijing. Watch the official video if you don’t trust my screenshots. Look at the direction of the blade as it lands on the ice, and also at Chen’s effort to control a jump that lacks more than a quarter of a revolution and is definitely not effortless.
This high-definition images is an option, not the only one.
Multiple cameras can also be used to evaluate the rotations and also the correctness of the flip and lutz edges. It doesn’t take long, it doesn’t slow down the competitions. It takes less time to look at the same jump from two different angles, one of which is clear, than to look at it multiple times from the same unclear angle.
There is another system. The blade leaves a mark on the ice, which can be seen and used to evaluate rotation.
If in doubt, you can watch the video:
Or again… I don’t know if you know, but a while ago a student from Waseda University graduated with a thesis entitled “A Feasibility Study on Utilization in Figure Skating by A Wireless Inertia Sensor Motion Capture System“. If you are not aware of it, you can find it at this link:
I know, it is in Japanese, but there are also Japanese among you, who can read it and, in any case, the scholar who made it knows English, so you can ask him directly to illustrate the content, and perhaps to collaborate with him to further develop the system. He has the technical knowledge, and I’m sure he would be happy to help you improve your discipline.
With what emerges from the thesis it is possible to evaluate the rotation of the jumps, both at the take off and at the landing, the edge of flip and lutz, but also the problem of the full blade, which, for example, makes a flip with a full blade take off assist no different from a loop.
We must simplify the deductions. Why from -1 to -2 for “!” and from -2 to -4 for “e”? Pick a value, and it’s the same for everyone. And it’s not the judges who apply the deduction, they don’t even have to think about it. If the technical panel (or rather the technology) makes the call, the deduction is triggered automatically. The only thing that the judge must remember, in the end, is to remove bullet 2, if they have assigned it (and if the judge does not do so, and assigns a score higher than +2, the maximum possible in the absence of bullet 2 and with a deduction of -1, computer automatically reduces it to +2. This if the deduction is -1, if the deduction is higher, the computer reduces the mark accordingly).
Abolition of the “q”. If there is no middle ground between < and <<, I don’t see why there should be between fully rotated and <. And the same deduction for all the skaters for the signs << and for all the <, assigned by the computer on the call made by the technology (or by the technical panel for the competitions in which a better technology has not yet been introduced).
If there is a step out, the call comes from the technical panel, with the automatic application of a -3. Obviously the judges can only assign bullets 4 and 6. At the end of the program the judges are shown the replay of that particular landing, and the judge decides if the step out is small, and therefore the judge does nothing, or if the step out is big, and then the judge presses a button to report it and increases the deduction to -4. Similar goes for the various touch downs, minimum deduction automatically assigned on the call of the technical panel, eventually increased at the end by the judge.
For the step before the jump or the long preparation, the task can be entrusted to the technical panel, with the help of the computer. A member of the technical panel presses a button when there is a step or a turn more difficult than a mohawk, another when there is a difficult turn, yet another when there is the technical element. Three buttons aren’t too hard to hit while watching a program, and if the tech panel can call the step sequence level, I suppose it can recognize the steps it sees. Then the computer enters the scene again: a step up to two seconds before the jump? Bullet 4 assigned. Eight seconds or more without steps? Deduction for long preparation. The judges must have nothing to do with these decisions, it is the computer that makes them. The judges have to evaluate only the other four bullets.
Reduction of the power of judges. At the moment a +1 (or a -1) is worth 10% of the base value of the element. Bring that value to 8%.
Blocking the GOE to -5 via computer for Jump element not according to requirements in the SP, without allowing the judges the possibility of assigning those -4, or even -3, that they sometimes happen to see.
Increase the importance of what is not a jump. In the Men’s competition in Beijing, the best in the short program and the last qualified for the free skate had these base values:
| SP | Jumps | Sp+StSq |
| Chen | 36,27 | 13,50 |
| Britschgi | 25,29 | 13,60 |
In the free skate the difference was even more pronounced. These are the values of the best and the worst skaters:
| FS | Jumps | Sp+StSq+ChSq |
| Chen | 78,86 | 15,48 |
| Shmuratko | 49,33 | 14,10 |
Overall, the first classified and the 24th had these base values:
| SP+FS | Jumps | Sp+StSq+ChSq |
| Chen | 115,13 | 28,98 |
| Shmuratko | 73,26 | 26,40 |
There is a huge disproportion. Increase the base value, and consequently also the GOE, of the spins, steps and choreographic sequence by 10%. This does not eliminate the disproportion, but slightly reduces it, and incentive the skaters to work for performing better elements.
Removing the step sequence in junior programs means leading young skaters not to improve their steps, because they don’t score any points. This automatically results in more empty programs. And it is not an assimilation with the senior programs, because the senior programs have both the step sequence and the choreographic sequence. If junior skaters can’t do everything, it’s best to take a jump off.
If you simplify the work of judges in the GOE, this will make it easier for them to evaluate the PCS, because they will have less things to deal with, without watering down the PCS. For the assessment in PCS you can ask the Waseda University graduate for help. Maybe it doesn’t solve everything, but probably interesting ideas can be found in his work.
Apart from that, the skaters can be asked to provide the layout of the steps and turns, as they already provide that of the technical elements. Among the tasks of the technical panel, there is to check that the layout of the steps is respected at least 85%. If they don’t, save for any steps they don’t take because they’re rising from a fall, their SS, TR and CO score is reduced by 50%. By having the layout of the steps and turns in advance, it is possible to provide to the judges graphic such as those made by Roseline Winter and Elisa, to help them assign correct marks:
It is also possible to create graphs and tables like those made by me based on the analysis of Roseline Winter and Elisa (and which would be easily made on the layout presented by the skaters):
For this table I looked at what precedes (top row) and what follows (bottom row) a difficult turn (difficult, column on the left) or an element (element, column on the left, box in light blue). The subdivision is between a push (power skating, PS, box in light green), a transition (tr) a difficult turn (D, box in yellow), and an element (E, box in red).
In this way it is possible to understand if a skater has skating skills good enough to be able to do several difficult things in a row or if he needs time between one difficulty and another. The score will be more accurate, and the judges’ work easier. You can also find a graph, always based on data from Roseline Winter and Elisa, at the end of this post:
Elimination of the roof on the PCS. In the Beijing free skate, Yuzuru Hanyu fell twice, and despite his mistakes his skating skills are better than everyone else’s, and his program is richer in transitions. Why can’t he get the score he deserves? Falls are errors on technical elements, they can have an impact on PE and IN, they don’t have it on SS and CO, not falls like his, and the former has no impact even in TR. The impact of the fall must be assessed from time to time, not all falls are the same and the roof has become something arbitrary, used at the discretion of the judges even when the rule would not be applicable, as in Hanyu’s own short program.
Elimination of the corridor in which the judges must stand. Mr. Mika Saarelainen made it clear: the judges are afraid to leave the corridor. If a judge thinks a skater’s SS deserves 7.50 points, but he knows that all of his colleagues will award him a minimum of 9.00, that judge will award a mark he knows is wrong to stay in the corridor. Instead the corridor can be wrong, all judges can be wrong. Don’t believe it? Check out the Men’s free skate at the 2019 Grand Prix Final. In the Choreo Sequence Chen stumbled. At a minimum, bullets 3 and 6 are missing and a deduction of -1 for the stumble had to be applied. The maximum possible GOE was +2. All marks were +4 or +5, so the judges stayed in the corridor because they all got it wrong. If a judge had assigned a +1, assuming there were only two bullets and applying the deduction, he would have been the one to walk out of the corridor and have to justify himself.
No, the judges must be able to leave the corridor by assigning the mark they deem right. If they can justify the mark (with the application of the rules and with the evidence of the images, not with simple feeling), they are right.
Turning on a light on the judge’s monitor if he is assigning a mark too far from the average of his colleagues. Not to force him to stay in a corridor that doesn’t exist, but to avoid mistakes. In 2020 Mr. Jerome Poulin awarded 0.75 in Skating Skills in the free skate of the Junior World Championship to Yuma Kagiyama. I am convinced that Mr. Poulin simply pressed the wrong button, and that he would have appreciated a report that would point out the error in order to give him the correct mark. And if you think it’s a little mistake, not worth to think at, I remind you that in 2014 Anna Cappellini/Luca Lanotte won gold at the World Championship with a lead of 0.02 points over Kaitlyn Weaver/Andrew Poje and 0.06 points over Nathalie Pechalat/Fabian Bourzat. Every fraction of point is important, and skaters deserve to be given correct scores. A mistake is only acceptable if you have done everything possible to avoid it, otherwise it is just indifference. And it’s not a mistake that took you by surprise. In the short program at Skate America 2003 a judge awarded to Sasha Cohen 0.75 in Performance.
Judges must be professional, get paid for their work, and be appointed by the ISU and not by individual national federations, who may want to put pressure on them. If you still don’t know it, I suggest you read an article written 12 years ago for Slate by Ray Fisman, Is Figure Skating Fixed? You can read it here:
Fisman’s conclusion is disheartening:
To be a skating judge, you need a sense of nationalism; for skiing, a sense of integrity.
If you are interested, on my blog I have put together links to numerous other articles that talk about judging problems, in some cases also providing the name and surname of the judges, in this post:
If a judge gives clearly biased marks, he must be disqualified. But the judges’ marks must be carefully watched every time, you can’t ignore it. Last summer Ms. Salome Chigogidze was suspended for national bias for the votes she awarded at the 2021 World Championship. I didn’t need to see that competition to know that Chigogidze helps her compatriots, I wrote it publicly as soon as I saw her name in the panel of judges:
How is it possible that I knew that she was aiding her compatriotes, and that you did not know? There is something wrong, and that something has allowed her to influence the result of another competition. No, judges’ votes must be controlled, not because they have to stand in a corridor, but because they have to explain them. And on the protocol everything must be indicated, the bullet assigned, the deductions, and the reasons for the marks in the PCS, even at the cost of having protocols hundreds of pages long. Skaters must are given until 12 hours after the end of the competition to contest the marks, their own or those of the other skaters. If the results can be corrected for doping, they must also be able to be corrected in the case of serious errors by the judges.
Probably if I think about it I can think of something else, for now I’ll stop here. There would also be considerations to be made on the score (parameterize the PCS to the maximum possible TES in each category, review the BV of some elements, such as 4A), but given that I focused on the judges, I do not go into these aspects. Do you want people to watch figure skating? Ask the skaters to create memorable programs, and not a series of jumps joined out of thin air, write rules that encourages the creation of richer programs and make sure that the marks assigned by the judges correspond to what the skaters did on the rink, giving them all the better technologies to do their jobs. And disqualify all the judges who cheats.
