Is it OK that PCSs tend to track TES?

[gkelly]

It's certainly possible to take notes on notable transitions, travel patterns, use of music, etc.

Some judges are better at it than others. Or care more than others.

E.g., some might pay more attention to the patterns, others to the use of music. Some might pay more attention to transitions using edge skills, others to use of the upper body. Etc.

But it's not impossible to pay attention to at least some of those things while also marking the elements.

And then, yes, a judge will also have an overall impression of the program aside from the details they took notes on.

 

[lariko]

I think there is upper body vs low body divide, like in everything with FS. I find that a lot of high PCSs skaters actually hold their hands down a lot, but they get high PCSs because of how they move their feet. And some fans would actually diss skaters who move upper body and arms a lot saying it's bad, but would go bananas for someone who made a small flick with a wrist 2 times in the whole program but right on music beat. While I didn't even see it. And then, there is TAT, who will die on the hill that the upper body should move up and down from the center line to grab the audience.

 

[Mathman]

The way judges put in their scores may lead to Tes focus. Each element is called in. Judges input their GOE. Then, only at the end, they give their PCs score. They are not looking in real time if there were nicely executed difficult transitions and if edges and turns were clean. They are, at the end of a program, giving their overall impression of the skate. It would be pretty difficult in my opinion to expect them from forgetting about a masterful technical display... Even if that performance was relatively empty from much else than elements. So there we have it, that's the whole issue. This is why having two panels would be better.

I agree with everything in this post except the last sentence. I do not see any advantage to the two panels solution.

In general I think that figure skating needs to understand what it is and what is not. What it is, is a competitive sport with an Olympic slant. What it is not is the Moscow Tchaikovsky prize (though both feature music) or the Boishoi ballet (though both feature graceful movement.) These entities are not requesting any figure skating judges to help them out with jurying and criticism.

One time in 2010 when it was Yuna Kim against Mao Asada for the Olympic gold medal, there was a very famous dancer serving as artist in residence at the University of Michigan. She delivered a long scholarly opinion piece on the theme that Mayo Asada should have been given the prize over Kimmie Yuna because Mayo was the better dancer.

Actually, it was an interesting piece.. But when someone asked, yes Asada was a better dancer, but who did you think was the better skater, the famous expert did not seem to understand the question.

 

[Arigato]

Perhaps the title of the thread should have a particular male skater's name in the title. Otherwise, I don't see how anyone could think a skater with high TES would also have high PCS. No one gave Boyang Jin PCS to match his TES. Hell, even Boyang would admit that. Shaidorov's PCS doesn't "track" his TES either. He is not a master of the blade by any stretch of the imagination any more than Boyang was. As a matter of fact, in my earlier post I posted a TES and PCS of a male skater:

TES 110.66
PCS 80.72

Who is it? Boyang Jin. 2016 4CC FS

Fast forward to 2025 WC FS

TES 110.17
PCS 82.53

Who is it? Shaidorov.

Both men were/are good jumpers with mediocre skating skills. And all those years in between won't change that. Just because a particular male skater is getting inflated PCS on a too-frequent basis doesn't mean the entire sport's scores are suspect.

I just looked above me and will quote @Mathman

"One time in 2010 when it was Yuna Kim against Mao Asada for the Olympic gold medal, there was a very famous dancer serving as artist in residence at the University of Michigan. She delivered a long scholarly opinion piece on the theme that Mayo Asada should have been given the prize over Kimmie Yuna because Mayo was the better dancer."


I have a major problem with that. This is a SPORT - not the ice version of the American TV show, So You Think You Can Dance.

 

[4everchan]

I agree with everything in this post except the last sentence. I do not see any advantage to the two panels solution.

In general I think that figure skating needs to understand what it is and what is not. What it is, is a competitive sport with an Olympic slant. What it is not is the Moscow Tchaikovsky prize (though both feature music) or the Boishoi ballet (though both feature graceful movement.) These entities are not requesting any figure skating judges to help them out with jurying and criticism.

One time in 2010 when it was Yuna Kim against Mao Asada for the Olympic gold medal, there was a very famous dancer serving as artist in residence at the University of Michigan. She delivered a long scholarly opinion piece on the theme that Mayo Asada should have been given the prize over Kimmie Yuna because Mayo was the better dancer.

Actually, it was an interesting piece.. But when someone asked, yes Asada was a better dancer, but who did you think was the better skater, the famous expert did not seem to understand the question.

2010 was well judged.. but ask many Canadians once in 4 year fans, and they may have thought Joannie had the most compelling performance.... Who cares about triple triple combos and 3axels . What's obvious to me is that the system is still not making fans confident enough

 

[Mathman]

This is exactly the point which I was to bring up. Correlation between different component marks is exactly the proof of faulty judging from the methodological point of view. If these factors are not factually related - and they are not - and the scores go together on all of them, then clearly the judging is faulty - incompetent or corrupt.

I vigorously disagree with the position that unexpected and unexplained statistical correlations are proof of incompetence or corruption or, really, much of anything. In the 1990s there was a study that found a striking correlation between teachers' salaries and crime rate ib Detroit. Every year they gave teachers a raise and every year there was a higher incidence of violent crime!

Anyway, about the issue that the three component marks are often similar to each other, this is hardly surprising. If you have superior skating skills you can enlist those skills in the service of more intricate, thoughtful and pleasing choreography. If your choreography is stronger you can more freely and confidently project your presentation to the audience. And if you are on a roll with the skate of your life and the audience is cheering wildly, somehow your feet might start to twinkle on the ice.

 

[Joekaz]

I think the best answer is that if it requires slow motion to see, it doesn't belong in PCS. If it affects the performance enough to be noticeable in full speed, then yes it can be part of the PCS calculation. Actually skating skills are much better judged at full speed, and better still in person where you can fully see the speed and ice coverage. The skating skills that are delineated in the rules for PCS actually have a great effect on artistry. Effortless speed and glide, flow, posture, use of edges, variance of speed all have a huge impact on artistry. Some of the small details on the GOE bullet points that require slow motion to see, really have no effect on the criteria that are used to judge skating skills.

 

[Mathman]

Perhaps the title of the thread should have a particular male skater's name in the title. Otherwise, I don't see how anyone could think a skater with high TES would also have high PCS...

TES 110.66
PCS 80.72

Who is it? Boyang Jin. 2016 4CC FS

I agree that there are two separate questions here. We can ask if there is a general trend for big techies also to get inflated PCSs, and separately we can ask about each individual skater, does he get better PCSs when his jumps are on target compared to when they are not.

As for Boyang Jin, I never really understood what was supposedly weak about his skating skills, choreography, etc. He tried hard to present entertaining programs, like Spiderman, but it didn't seem to matter. Esthetics? His Lutz edge was a thing of beauty all by itself.

 

[Mathman]

Would it help if we had some actual data to work with?

So I just looked at the men’s LP from Worlds and did a quick Spearman rank correlation test comparing the rankings, 1st, 2nd, 3rd,…,20th in TES against the rankings in PCSs. This is a non-parametric significance test that approximates the usual least-squares correlation approach. It’s advantages are that no assumptions are needed about the underlying distribution of data (symmetric, etc.) and that you don’t need a computer — or even a calculator, hardly — to carry out the arithmetic.

The disadvantage is that it does not discriminate between beating your opponent by a little or by a lot, so it will not mollify fans who feel that, yes, I am OK with skater A beating skater B, but the margin of victory should be smaller, those crooks.

Anyway, cutting to the chase, the final result was r = 69% correlation for a student’s t statistic of 6.60 (18 degrees of freedom). This is way, way, way goo-gobs statistically significant. Meaning that yes, there really is a tendency for high scores in tech to match up more or less with high scores in PCS, compared to two independent orderings produced at random by monkeys throwing darts at a dartboard. We are 99.9999… percent confident that no monkeys were involved.

 

[Arigato]

The disadvantage is that it does not discriminate between beating your opponent by a little or by a lot, so it will not mollify fans who feel that, yes, I am OK with skater A beating skater B, but the margin of victory should be smaller, those crooks.

HA!

Meaning that yes, there really is a tendency for high scores in tech to match up more or less with high scores in PCS, compared to two independent orderings produced at random by monkeys throwing darts at a dartboard.

What may be considererd high PCS to one may not be considered high PCS to another.

 

[4everchan]

I agree that there are two separate questions here. We can ask if there is a general trend for big techies also to get inflated PCSs, and separately we can ask about each individual skater, does he get better PCSs when his jumps are on target compared to when they are not.

As for Boyang Jin, I never really understood what was supposedly weak about his skating skills, choreography, etc. He tried hard to present entertaining programs, like Spiderman, but it didn't seem to matter. Esthetics? His Lutz edge was a thing of beauty all by itself.

Really? Just watch his blades on the ice and compare to Chan... Judges got it right

 

[yelyoh]

I have been saying for years TWO SEPARATE PANELS, one for tech and the other for components!!!

 

[Mathman]

Really? Just watch his blades on the ice and compare to Chan... Judges got it right

Jin was the only skater I ever saw who rode that outside Lutz edge all the way in, ever deepening it throughout the whole approach and lift-off. I thought that was really cool. OK, he wasn't Captain Blade Skills ike Patrick (who is?) but I never thought that he was so bad that he deserves to be the poster boy for big jumps, no skill.

 

[Mathman]

What may be considererd high PCS to one may not be considered high PCS to another.

That's the beauty of this kind of test. It doesn't matter whether the scores are high or low, just higher of lower than the next guy's. This was a super-useful statistical test of judging consistency back in the 6.0 days, not so much now, though.

 

[Tutto]

It is my humble opinion that SS do not belong to PCS but should be judged as part of TES (apologies if this was already discussed I do not have time to read all the posts here)
I've always thought that judges look at the speed and ice coverage when considering PCS more than anything else but I am not sure of even that anymore. Example: Nikolaj Memola walks on the ice from one jump to another as slowly as it humanly possible and gets away with 7.71 in SS. I would give him 5 or 4 I think. Yes, he is exceedingly handsome young man (and he knows it!) which must have some effect on the ladies and as most of judges are female it is obviously working (+ Olympics are in Italy).
Andreas Nordeback in the same competition (Worlds in Boston) 'rewarded' by 7.46. ???
I feel like we could, of course, continue to pass the time and vent out our frustration over unfair PCS but the reality is the FS is a subjectively judged sport and nothing will ever change whether it is dressed under the 6.0 system or COP -the end result will always be the same. Judges will do what they want or what they are told to do.

 

[gkelly]

It is my humble opinion that SS do not belong to PCS but should be judged as part of TES (apologies if this was already discussed I do not have time to read all the posts here)

Skating Skills should be considered part of a technical score, but not part of the Technical ELEMENTS Score. It's not (primarily) about the elements.

When a skater starts the program and starts skating around, before they ever do the setup for the first element, that is showing skating skills and setting each judge's initial expectation of what range the SS might fall in (to be adjusted up and down depending on how the skating goes during the rest of the performance)

If we wanted to split the scoring into technical vs. artistic scores (instead of element scores vs. whole-program scores, which is what PCS are in the current system), then the technical score would need to be split between TES and SS.

E.g., assuming everything else stays the same as now, scores could be reported as:

Technical Score: Technical Elements Score (calculated by adding all the base values and GOEs, as shown at the top of the protocol) + Skating Skills score (a single score rating the skating skills throughout the program on a scale of 0 to 10)

Artistic Score: The sum of the remaining full-program-based component scores, currently Composition + Presentation

If the panel were in fact to be split into different groups of judges focusing on technical vs. artistic parts of the performance, that would be a reasonable way to split the workload

I've always thought that judges look at the speed and ice coverage when considering PCS more than anything else but I am not sure of even that anymore. Example: Nikolaj Memola walks on the ice from one jump to another as slowly as it humanly possible and gets away with 7.71 in SS. I would give him 5 or 4 I think.

Have you compared his speed and ice coverage to that of juniors (or weak seniors without quads or triple axels and maybe inconsistent triples) who are regularly earning scores of 5 or 4?

How about his Variety of edges, steps, turns, movements and directions; Clarity of edges, steps, turns, movements and body control; Balance and glide?

We accept that these may be significantly weaker than that of other quad-jumping medal contenders.

But how do they compare to the skill level of skaters who do regularly receive scores of 4 and 5?

There are a lot more skaters who deserves 4s and 5s than who deserve 7s. Most of them are also not capable of earning the minimum technical scores required to enter ISU championships, and/or of placing higher than their national rivals to earn assignments to those events. So if you mostly ever watch ISU championships, and Grand Prix, and maybe the top group at a Challenger or JGP Final/top group at Junior Worlds, you're mostly going to see skaters who deserve scores in the 7s and higher, occasionally 6s, rarely 5s. You need to look elsewhere to see what 4s and 5s tend to look like.

Watch more JGPs, and full fields at non-Challenger senior B events, to see more skaters at that skill level.

Or national competitions, focusing on the whole field and not just the medalists. If you get the opportunity to attend a national competition in person, that could also give the opportunity to see seniors and juniors live in the same place in the same week.

 

[Mathman]

...
Technical Score: Technical Elements Score (calculated by adding all the base values and GOEs, as shown at the top of the protocol) + Skating Skills score (a single score rating the skating skills throughout the program on a scale of 0 to 10)

Artistic Score: The sum of the remaining full-program-based component scores, currently Composition + Presentation

First, thank you for keeping before us the point that the iSU judging system has to work for skaters at all levels, not just for the tiny handful of elite seniors that we see on TV.

About technique versus "artistry," I think that GOEs also have some artistic judgement going on, like timing of the element to the music and within the plan of the choreography. In general I would prefer to minimize the distinction. You can't have artistry without technique, while technique without artistry is just... technique without artistry.

Back when the three components were SS, Choreography and Musical Interpretation, I felt that the latter two were too intertwined to warrant separation into two distinct components, I would have been OK with just two, SS and Esthetics (possibly with weights if we don't want "artistry" to get swamped into irrelevance altogether). However, I am quite satisfied with the three that we have now. Presentation and Composition are sufficiently distinguished and (reasonably) well defined, although maybe "good presentation" falls somewhat into the category, "I can't say what it is but I know it when I see it."

 

[gkelly]

Back when the three components were SS, Choreography and Musical Interpretation, I felt that the latter two were too intertwined to warrant separation into two distinct components,

There were never just those components.

Originally, the five components were Skating Skills, Transitions, Performance/Execution, Choreography, and Interpretation of the music. There was some rewording of the names of some components over the years, until ca. 2021 when they were reduced to the three we have now.

I agree that there was a lot of overlap between PE and IN and between CH and IN. Which, I understand, is why they redefined them.

I would have been OK with just two, SS and Esthetics (possibly with weights if we don't want "artistry" to get swamped into irrelevance altogether). However, I am quite satisfied with the three that we have now. Presentation and Composition are sufficiently distinguished and (reasonably) well defined, although maybe "ggpd presentation" falls somewhat into the category, "I can't say what it is but I know it when I see it."

Judges should be able to say what "good presentation" is. That's what the written guidelines are for.

But no matter how much the ISU tries to put it into words and to use those words in judging training, it still comes down to each judge making a qualitative assessment in real time for each performance that they see. How and how much they apply the wording of the criteria when doing so will likely vary a lot from judge to judge.

 

[lariko]

I mean, the guys with a big jump in them, but no SS don't get high PCSs. For example, Fedir Kulish of Latvia landed big jumps, got low PCSs scores. When people want low PCSs for so-called low PCSs skaters, they drop to the PCSs values that are way too low (I've seen people saying that Malinin 'should' get 2's). On the top 12 level, the PCSs imo appropriately start in the 6's and 7's. I don' mind people disagreeing about PCSs, but going too far is going too far. I had seen Memola live. He is slower, but impressive, and certainly not on the 4 level of SS, because if he gets 4, what do you even give the skaters who are not in the leet tier of international competitions, who compete nationally, regionally, on a city level, etc. Negative values?

Also, GoE. So, watching both WTT and WC, I am feeling that Kagiyama's edges on jumps are actually too much because his propensity to exaggerate the edges make the landings look loop-sided and unstable even if he stays on his feet. Malinin's jumps (despite straighter knees and lesser edge) are way higher, effortless, sure and, therefore, more impressive. So, is it actually a perfect technique if it brings the jump this close to a failure to land to show more edge? Too much of a good thing...

 

[gkelly]

When people want low PCSs for so-called low PCSs skaters, they drop to the PCSs values that are way too low (I've seen people saying that Malinin 'should' get 2's).

"People" in this situation being fans who don't understand the full range of skills the PCS scales need to cover.

Judges certainly don't do that.

On the top 12 level, the PCSs imo appropriately start in the 6's and 7's. I don' mind people disagreeing about PCSs, but going too far is going too far. I had seen Memola live. He is slower, but impressive, and certainly not on the 4 level of SS, because if he gets 4, what do you even give the skaters who are not in the leet tier of international competitions, who compete nationally, regionally, on a city level, etc. Negative values?

Exactly.