What are good 'skating skills'? Who should get high SS scores? | Page 3 | Golden Skate

What are good 'skating skills'? Who should get high SS scores?

gkelly

Record Breaker
Joined
Jul 26, 2003
Firstly thanks for entertaining the hypothesis, I don't know all the answers but at least it is interesting to contemplate and refine ideas.

Yeah, let's just brainstorm how things might work.

In the ideal world at elite competitions (world, olympics, GPF) I had always hoped for separate panel of PCS specialist to improving judging standards (judges feel less pressured, and focus on their area of expertise),

So what is their area of expertise.

I'm sure some judges who established their judging careers under 6.0 already feel that some of their expertise has been taken away from them and that power given to the tech panel and to whoever makes up the scale of values.

like BV can maybe decided beforehand depends on if the judges is familiar with the program. At GPF, Olympics, Worlds afterall, most judges would have been familiar with program construct, ambition, difficulty, quality, history of the choreography already of the top 20 (may be during practice sessions to check? Not sure...), then perhaps all 9 judges then can just focus on ability to execute these 4 areas, while BV takes care of itself?

That's one reason why judges watched practices under 6.0 and may still with IJS. But skaters don't always do full runthroughs in practices. And judges won't be equally familiar with all the skaters from previous events.

And of course skaters don't always perform the same way in competition that they do in practice, or at one event the same way they did at the previous event.

So although judges might use whatever prior knowledge of this program they have to come up with a rough idea of the general range of PCS a skater might deserve for that program, I don't think it's possible to set a firm base value in advance. A base value based on difficulty of the choreography needs to be based on the difficulty of the choreography actually performed during the competition.

Even with the elements, we (skater, tech panel, TV commentators, etc.) might know what the base value for the technical elements would be if the skater gets full credit for all of them. But in practice a skater might execute one of the features in a spin or step sequence not quite well enough to get credit and only earn level 3 instead of the planned level 4, or the skater might underrotate a jump, or double a planned triple, and thus earn a lower base value than when she rotates it as planned. That's only taking into account the base value assigned by the tech panel. GOE from the judges is a whole different issue. It's entirely possible that in any two performances the skater will get higher base value for the element on one day and higher GOE for the same element on the other day.

Yes thanks for remind me, the factoring thing has always felt like funny mathematics to me. I know it is suppose to make PCS more or less equal to the tech score, but what it is really doing is just reinforce the PCS differentials and nothing more beyond that. Men get it twice, Women get it 1.8 times in the FS, it doesn't really say much about the quality of the skating components themselves. In any case it seems by awarding PCS cateogories out of 20, mathematically it works out factoring of 2.0 for the men already, so it need not change. So for women mark out 18 instead of 20 resolve any issues? Still seems funny and strange.

If it were up to me, I'd just let women's PCS have the same factors as men's, and let the PCS represent a larger percentage of the total score in the women's competition. So jump content would have less effect on the final results than with the 0.8 short/1.6 freeskate factors (sorry if I misstated the value of the factors earlier).

So best SS/PE/TR/CH/IN vs worst in the ladies short = 1x 3T only, but in Men's FS = 2 quads. :laugh2:! :shrug:

The best vs. worst total PCS in the ladies' SP at Worlds was 33.53 (averages in the mid 8s) vs. 19.24 (averages in the high 4s) -- total difference more than 14 points, or approximately equivalent to the base value for three triples including at least one harder one.

The skater who earned 19.24 for PCS placed 29th in the short program and did not qualify for the free skate. The skater who was the last to qualify in 24th place earned 20.20 for PCS, still more than 13 points behind the winner; the last-place skater in the SP earned 20.60 for PCS, higher than either of the above, so it was the elements more than the PCS that relegated her to last place.

I'm sort of throwing around the word "accurately" when I moreso accurately in the sense of each individual component actually evaluated separately, rather than just throwing a bunch of similar numbers for each category. Similar numbers in each category might indeed work for some skaters/performances, but certainly not all the way we see. Looking over the LP protocols from the top ten ladies at Worlds, the highest and lowest final component scores all had the range of less than one point away from another. For every skater, without exception (I believe). Scores from individual judges also tended to follow this pattern, through there were a few exceptions among the judges individually (like one judge gave Ashley a 7.75 in transitions but a 9 in P/E and interpretation. But individual judges giving scores outside of the one point range are the anomaly rather than the norm. Usually about one judge per skater, and usually the "anomaly" is caused by the transitions score.

There are a couple of different things going on here.

Even if every judge made a conscious effort to use larger gaps between highest and lowest component score for each skater, unless they all agreed on which component was best and worst for each skater, the averages would still end up in a narrower range than any individual judge would give.

Because the history of the sport in all eras has included so much generic choreography, judges probably come to expect "above average by skating standards" to qualify as "good." It might take more training in aesthetic principles in general for judges to recognize greater differences in choreography, or music interpretation -- especially those who have not studied the arts outside of skating and consider themselves more experts on skating technique than on artistic qualities.

There are surely some Golden Skate posters who are more knowledgeable about performing arts than some judges. Is the average level of arts knowledge among skating fans higher than among skating judges? Possibly.

On the other hand, the rules and traditions of the sport have taught judges to value certain qualities highly that may be unimportant and hard to discern on video for fans following along at home, and that permeate the whole performance. A skater who commands the ice with strong skating skills is also likely to make a stronger impression on judges in the arena in terms of performance/execution and maybe some areas of choreography and interpretation, and therefore earn higher PCS all around from the judges. Meanwhile, a slower, more tentative skater might have prettier body line and more subtle movements expressing the music that come across better on video than live, so most fans might agree that skater deserves higher PE, CH, IN scores even if the SS are lower and expect a bigger difference than the judges see fit to give.

Fans often watch just the top 6 or 10 or 20 skaters in the field, whereas judges see the whole field at that competition and have spent a lot more time watching non-elite skating. So what fans see as "worst" might actually look pretty good to most judges.

Does that mean that the judges are right and the fans are wrong? Or the judges are wrong and the fans are right? Or is it just a matter of different priorities and different experiences of the skating itself?

Of course there is room for subjectivity, but I do think that the majority of Golden Skate posters would agree that, for many skaters, some PCS categories should be higher/lower than others with more of a difference than the scores currently reflect. Most skaters have strengths and weaknesses to their presentation/program/performance and the scores do not reflect that.

Can we choose a few performances, preferably skaters who have not competed head to head so that we won't be influenced by how judges actually scored them against each other, and discuss what kinds of scores we think they deserve for each component and why? I think it would be interesting to see how much we agree or disagree with each other, and how much difference we see between different components.
 

OS

Sedated by Modonium
Record Breaker
Joined
Mar 23, 2010
That's very subjective though. You think that program deserves 9 out of 10 for choreography, I think it deserves 7 and I'm being generous. The only good thing about that program was the step sequence. The rest was telegraphing and shaking her belly. Now, her SP was the complete opposite. PCS can't work as GoE's for technical elements. They are much more complex to judge and PE, IN and to a certain degree the choreo, will always have the subjective elements in it. [/FONT]

But I think it is possible to be objective about something like Choreography? Just like there are musical pieces you can do that is graded according to level of difficulty and I don't mean just technical difficulty but also emotional sophistication, intricacy etc? For example, a Carmen for me is most straightforward to perform vs something abstract, emotional complex and never been used in figure skating. That is why it is good to divide give it a BV based on difficulty, appropriate musicality, balanced upper body movement, intricacy of transitions, things that are a bit more objective prior to the competition. Then have the other 50% of PCS categories mark based on fulfilling these objectives on a scale of 1 to 10. I am not sure the average judge are able to judge these correctly, that was why I think there should be specialists. May be all programs can be given levels prior to the competition by panelists. Where there are feedback letting the competitor know this is level 7 out of possible 10 because XYZ rather than just some random number that nobody can make sense of other than goes: !okay so she is 2nd in PCS between this person and that, i guess that about sound right, because she is 2nd overall."

Between you and I, I too agree with you 7 for Carolina's choreography would be generous, but I didn't want to annoy her fans by giving it a 5 or 6 for Bolero 2 seeing for some reason many people love it so much and wanted it be the winning program at Sochi. I find that there so much poseography, breathing room in that choreography, it deserves its own category. Just curious, who did you think have the best choreography at Sochi?


Re:gkelly

Alot of food for thought as usual, I need a bit more time to think about this. Perhaps I don't pay enough attention to those outside top 10 enough to be aware of the gap. My view is 7.5 - 9.5 for those among the 10 doesn't seem much of a difference. The Mathematics kind of makes me laugh. Top 10 Women's best worst SS gap in the SP = 3t, less worthy than best worst Men's FS SS = 2 quads?! I don't even know how to begin to answer that! I agree maybe not having them factored may remove the confusion, but still I don't know. Are we saying PCS is more important for women then Men, PCS should out weighs tech? Why don't we give women a different set of TES values then? It will seem make sense consider 3A is more difficult for the ladies, and 3lz is pretty hard too and don't get enough rewards, and they don't have any quads. One thing I do know is that it is usually those at the top 6 where the stake are the highest, is where PCS often takes a life force of its own that is not necessarily about what is performed infront of them. In many ways, if you are to take unrelated performance and give them subjective marks, is quite different than if these unrelated performance all happen to eyeing the Olympics podium while each federation judges battling it out to support the home team. Then the question becomes how to

- improves objectivity
- minimise subjectivity, or allow subjectivity to be categorized and easier to spot
- minimise inflation
- give reward and credit where it is due there and there
- give penulties where it is due there and there

instead of a few competitions later due to cognitive latency effect for judges who only has 10-15 seconds to come up with a number, and may not have the confidence to step outside the corridor and outside the opinions/influence of their peers. At the same time, those who deliberately tries to slant the results, by giving boost to those they support vs and deflation to rivals, it is easier to spot and red flagged.
 
Last edited:

Alba

Record Breaker
Joined
Feb 26, 2014
But I think it is possible to be objective about something like Choreography? Just like there are musical pieces you can do that is graded according to level of difficulty and I don't mean just technical difficulty but also emotional sophistication, intricacy etc? That is why it is good to divide the categories into BV based on difficulty, appropriate musicality, balanced upper body movement, intricacy of transitions, things that are a bit more objective prior to the competition.

Absolutely. I'm just saying that still there is an element of subjectivity especially in PE and IN so it's difficult to put a BV for that.

Just curious, who did you think have the best choreography at Sochi?

From the top or in general? Some that immediately comes to my mind are: Mao SP and LP, Yuna's programs (I don't think they were masterpieces but I think the LP in particular was a very good choreo), Yulia's LP, Gracie's SP, Adelina's LP, Osmond SP (the best choreo for me), Carolina's SP (loved it), Suzuki SP.

p.s speaking about TR Osmond deserved 9s for her TR in the SP during 2013-2014. The best among the ladies.
 
Last edited:

OS

Sedated by Modonium
Record Breaker
Joined
Mar 23, 2010
Absolutely. I'm just saying that still there is an element of subjectivity especially in PE and IN so it's difficult to put a BV for that.

May be the BV of the PE and IN should be a factored version of what they received at their last outing, and be graded up and down according to how they did this time around? It seems more accountable than the current system we have?

e.g
Scenario 1
Julia received got 8 last competition.
This competition 8 BV + 7 PE
7 is +2 apart from median of 5 (mid point of 10). 2/20 = + 0.1
Her next base value for her competition is 8.1 to start

Scenario 2
This competition 8 BV => 10 PE (very rare once in a life time type of performance, we are talking about Sochi arena filled up with only fans of Julia)
10 is + 5 apart from median of 5. 5/20 = +0.25
Her next base value for her competition is 8.25 to start.

Scenario 3
This competition 8 BV, total disaster in PE => 2 PE
2 is - 3 out of medium of 5. -3/20 = - 0.15

Her next BV of PE for the competition will drop to = 7.85

--------
This gives it a threshold of no more than 0.25 points max per competition per category which is unlike to happen. or 0.25 PE + 0.25 IN = 0.5 for these 2 categories. Will this stop inflation?

The only disadvantage of this seems if you are under marked as a newbie at the start, you are pretty much stuck, on the other hand, the more you compete, the more you can improve. Also maybe undervalued will give the judges some incentive to score you higher?
 
Last edited:

gkelly

Record Breaker
Joined
Jul 26, 2003
But I think it is possible to be objective about something like Choreography? Just like there are musical pieces you can do that is graded according to level of difficulty and I don't mean just technical difficulty but also emotional sophistication, intricacy etc?

I don't think it's possible to be completely objective. Different skating experts, or choreography experts, will disagree about subtle points. Even if all agree that piece A falls into the general range of good (7) and piece B into the general range of very good (8), if they're in completely different styles/take completely different approaches (let's say syncopated and angular vs. seamless and smooth), then those experts who favor the A approach, who believe it's more challenging than their peers do, and believe that B is actually not as difficult as it seems, might think that both could be worth 7.5.

(And I don't think it's possible to be completely objective in music either, although I'm far from an expert there. The score at least stays the same, so everyone playing the same piece can start from the same baseline. But I'm sure experts disagree on whether one piece is more or less challenging than an other, and any set values are based on consensus that might not reflect any single experts exact rankings of difficulty.)

And then, of course, on any given occasion a skater might leave out some of the planned choreography, whether because of mistakes/getting behind the music, or a predecided plan with the coach to simplify something that had been giving her trouble in practice, etc. But the panel who decides on the base value wouldn't know something was left out if they never knew it existed in the first place. And they can't know every detail of all planned choreography in advance. It's not like a printed score. Every skater has her own choreography and is free to change it on the fly if she chooses.

I do think it's possible for a panel to reach a consensus despite their own subtle disagreements with each other. But not in advance. They have to evaluate the skating in front of them on that day, not what the skater did in practice or in some previous performance or what they said they were going to do on paper.

Therefore I'm absolutely opposed to the idea of scoring the base value of the choreography before the skater has even skated it in this competition, on theoretical grounds.

That's not even taking into account the practical grounds.

How would this work at the first competition of a season? And not everyone will have the same first competition. Some seniors will start at different senior B events, some will start at the first Grand Prix and some will start at the fifth Grand Prix, some will first compete internationally at Four Continents or Europeans, some might skate only at Worlds or Olympics that year, having qualified on the strength of the previous season's results. We can't expect international officials to take into account what happens at every single national championship, not all of which are televised even in their home countries.

Could base values be set equally fairly for programs that are being performed internationally for the eighth time that year vs. programs being performed for the first time at a senior international, by a skater the members of the panel never paid attention to before?

(This isn't taking into account that some skaters keep programs for more than one season, but often make subtle changes from one to the next.)

Re:gkelly

Alot of food for thought as usual, I need a bit more time to think about this. Perhaps I don't pay enough attention to those outside top 10 enough to be aware of the gap. [/quote]

I think it's important to keep the whole field in mind when setting rules.

My view is 7.5 - 9.5 for those among the 10 doesn't seem much of a difference. The Mathematics kind of makes me laugh. Top 10 Women's best worst SS gap in the SP = 3t, less worthy than best worst Men's FS SS = 2 quads?!

I think you're getting confused.

There are 5 components. SS (Skating Skills) is only one of them.

In short programs men's components are worth the face value and women's 80%; in free skates, they're worth twice that.

So the difference between 7.5 and 9.5 -- two points -- in ONE component, e.g., SS only, would be 2.0 (about half a triple toe) in a men's short program, 1.6 in a woman's.

In a free skate, the difference between 7.5 and 9.5 in SS only would be 4.0 for men and 3.2 for women.

If there are similar gaps in ALL FIVE components -- e.g., if the judges just give 7.5s across the board to one skater and 9.5s across the board to another (we'd expect a little more variation to reflect what each skater is best and worst at componentwise but let's say it averages out that way if one skater is one of the very best in the world having a good day and the other is just one of the good ones but not one of the best) -- then the average 2-point difference over five components, with factoring, will add up to 8 points in a ladies' short program, 10 points in a men's short, 16 points in ladies' free skate, and 20 points in a men's free skate.

Then the question becomes how to

- improves objectivity
- minimise subjectivity, or allow subjectivity to be categorized and easier to spot
- minimise inflation
- give reward and credit where it is due there and there
- give penulties where it is due there and there

instead of a few competitions later due to cognitive latency effect for judges who only has 10-15 seconds to come up with a number, and may not have the confidence to step outside the corridor and outside the opinions/influence of their peers. At the same time, those who deliberately tries to slant the results, by giving boost to those they support vs and deflation to rivals, it is easier to spot and red flagged.

Yes, those would be the goals.

Whatever rule changes might be adopted to try to achieve them, they have to be applicable for all competitions at all levels, obscure senior Bs as well as Worlds and Olympics.

It might be possible to budget for extra officials at the big championships to split up the duties, but there would not be much opportunity for officials to get experience with the split duties at many competitions per year since most -- including probably Grand Prix events -- couldn't afford it.

And any differences in rules between different levels of senior competition that would require skaters to plan their programs differently depending where they're competing would confuse things unnecessarily.
 

usethis2

Medalist
Joined
Feb 11, 2014
My issue with PCS is that TR can either be included in either SS or CH. These components are so much overlapped. At the same time, IN can be included in PE. Personally I think there are too many categories in PCS. IMO, 3 categories are enough. Why not just SS, CH and PE?
I think the same and I participated in this discussion long time ago. I actually think 2 categories are enough - performance and interpretation. To be sure the current 5 subcategories can be distinguished in the abstract, but in reality they are used for judges to correct their mistakes (at best) or to manipulate the results (usual).
 

Meoima

Match Penalty
Joined
Feb 13, 2014
I think the same and I participated in this discussion long time ago. I actually think 2 categories are enough - performance and interpretation. To be sure the current 5 subcategories can be distinguished in the abstract, but in reality they are used for judges to correct their mistakes (at best) or to manipulate the results (usual).
Another solution is that adjusting the factoring. I wonder if the factoring is different, judges might give more variation across all the components.

My friend have calculated and suggested:
In sp: 0.4*TES bv + 0.6* total PCS (in scale of 10).
In lp, 0.4*TES bv + 1.2* total PCS (in scale of 10).
 
Last edited:

usethis2

Medalist
Joined
Feb 11, 2014
os168 said:
I think Judges should just make more effort to mark them correctly since PCS should be as important as TES. Just like each technical elements can be broken down into base values then with their GOEs, why can't PCS categories with some effort? Why not award a base value (e.g Choreography difficulty, interpretative difficulty etc) then add own GOES (positive and negative scoring depends on how the skater did on the day) too! It would seem to makes more sense.

This is an interesting concept. Any thoughts about how it might work, specifically?

Only way that can work, IMO, is by having compulsory programs. I share the concern over the PCS just like everyone else, but what os168 suggested presupposes judges' awareness of Choreos, before the skating takes place. Pre-judging and corridor-juding may be inevitable to a degree, but there is no reason to encourage them.
 

AsadaFanBoy

Final Flight
Joined
Feb 14, 2014
Only way that can work, IMO, is by having compulsory programs. I share the concern over the PCS just like everyone else, but what os168 suggested presupposes judges' awareness of Choreos, before the skating takes place. Pre-judging and corridor-juding may be inevitable to a degree, but there is no reason to encourage them.

There used to be a compulsory round. What do you have in mind when you say compulsory programs? I'm very interested/curious.

=) I also like the old format of the GPF - lots of skating all in one day. Exhausting for skaters, exciting for fans.
 
Last edited:

usethis2

Medalist
Joined
Feb 11, 2014
Similar to compulsory round in Gymnastics (floor), I guess. But that has its own issues..
 
Last edited:

andromache

Record Breaker
Joined
Mar 23, 2014
Only way that can work, IMO, is by having compulsory programs. I share the concern over the PCS just like everyone else, but what os168 suggested presupposes judges' awareness of Choreos, before the skating takes place. Pre-judging and corridor-juding may be inevitable to a degree, but there is no reason to encourage them.

Obviously it's completely unrealistic and would never happen, but I love the idea of compulsory programs IN THEORY. Compulsory programs with compulsory choreography (yet something fairly generic, not with a specific character or attitude that would definitely give some skaters advantages over others) in order to compare skaters execution of elements, moves in the field, and choreography side by side. It would be fascinating to consider and debate which skaters perform/execute identical choreography better than others.

It would be boring, bad for the popularity of the sport, too much money, never going to happen etc etc but I can dream.
 
Top