Ideas for how to rescore figure skating? | Page 3 | Golden Skate

Ideas for how to rescore figure skating?

cake

On the Ice
Joined
Jan 30, 2007
Someone took a video on how one judge scores Marin Honda's program.

I think judges already had the bullet points pre-calculated from watching the practice, and already had a GOE in mind for each element.

Thanks for finding that video! I`ve never seen before how the judging process actually works.
After I watched the video I realised how much I concentrated on the elements (and thinking how I would have judged them), I didn´t pay any attention on what Honda was doing inbetween. I can´t imagine judges even with their routine can really concentrate on both elements and PCS. I think they also have the component scores in mind already from watching the practice.
 
Joined
Jun 21, 2003
But that is not 0 GOE according to current requirements.

That was my point. This thread is not about what the current rules are but to make suggestions for improving the rules. I think the rules would be improved if they modified the requirements for positive GOE in such a way that an average jump would get 0 GOE. (Otherwise, what does "average" mean?)
 
Joined
Jun 21, 2003
Someone took a video on how one judge scores Marin Honda's program.

I think judges already had the bullet points pre-calculated from watching the practice, and already had a GOE in mind for each element.

Thanks for the post. Very interesting.

In reality, I don't see that there is any alternative to some degree of pre-judging. You go into the competition already with a notion of what to expect. I expect Yuzuru Hanyu to get exceptional height on his triple Axel (and also to display a bunch of other positive features). So I already have my finger poised on the +5 button. If the skater messes up, then I have an instant to change my mind.
 

[email protected]

Medalist
Record Breaker
Joined
Mar 26, 2014
i would love to to see spins and steps worth more points. both of them take quite a bit of energy to do nicely. there would definitely be more incentive to work on them instead of only jumps (worth way more), creating more well rounded skaters. i'd love to see that.

Yes, but only if the procedure of setting levels becomes much more tranparent and tech people become accountable for the levels they give. I already talked about it but I just keep digging. I looked at Trusova's StSq during her short programs at Nepela and Skate Canada. They both looked identical. In Nepela she recieved 1.6 ponits more for it. If we double the value of StSq the variance would be 3.2 points for exactly the same effort. For me it is unacceptable as "small step sequence" may become a real "deal maker" or "deal breaker" when establishing the placements.

Now tech people just bring the levels onto you. I don't know if there is any backstage work after that if we have such curious cases as Trusova's StSq levels. And I am eager to look at the CoR protocols. If Sasha gets level 3 that would be the funniest thing I saw in figure skating judging for quite a while. Like "guys, level 4 was an overkill - you were too excited but, well, level 2 was too harsh, we agree - let's settle at level 3 then". I shall post this quote if she indeed gets level 3 in Moscow.
 

andromache

Record Breaker
Joined
Mar 23, 2014
I'm so pressed. My faves can't win. Therefore ISU should change the rules

This is EXHAUSTING.

It's not about people's "faves" not winning.

It's about people valuing different aspects of figure skating.

If you're at all familiar with the history of the sport, what the sport of figure skating values has changed significantly over time. Figure skating used to value school figures. Then "free skating" was introduced, but figures mattered more. Over time, figures started to matter less and less...until they were eliminated entirely! Instead of the balance between figures and free skating, then, it became about the "technical" versus the "artistic."

Under IJS, these categories are broken down even further. Different elements are worth different points, balanced with PCS.

Every year and every Olympic quad, the ISU decides to reevaluate how different stuff is valued. Most recently, there was concern that technical difficulty was over-valued over quality, so the +5/-5 was introduced.

Now that the quad revolution is happening, people are questioning if jump rotations are over-valued other aspects of the sport.

Comments like yours are why we can't just have a good faith discussion/debate about the different things we value in skating.

If you are happy with how jump rotations are valued now, you should say that and explain why instead of accusing those who disagree with you of bad faith arguments.
 
Joined
Jun 21, 2003
Like "guys, level 4 was an overkill - you were too excited but, well, level 2 was too harsh, we agree - let's settle at level 3 then". I shall post this quote if she indeed gets level 3 in Moscow.

Actually, I think that thats is exactly how it goes. Like, he tech caller gives out a lot of <s, then the next time they say, you know, we have been giving out too many <s lately, we need to tone it down a little.

I also think they say things like this: Boy, we are seeing a lot of sloppy step sequences this year. Let's tighten up on the level calls and see if we can make the skaters work harder on them.
 
Last edited:

[email protected]

Medalist
Record Breaker
Joined
Mar 26, 2014
Actually, I think that tats is exactly how it goes. Like, he tech caller gives out a lot of <s, then the next time they say, you know, we have been giving out too many <s lately, we need to tone it down a little.

I also think they say things like this: Boy, we are seeing a lot of sloppy step sequences this year. Let's tighen up on the level calls and see if we can make the skaters work harder on them.

You may be right if after SC Eteri will go deep into Sasha's StSq to see what has to be improved and will make improvements. If her reaction is: "what the heck, we all know that it was level 4 StSq and it was given level 4 before" then she may make some calls and we shall see zero changes in Sasha's steps in Moscow. Well, for me it's much more suspense in it than in Trusova's quads. We know that she can land them.
 

gkelly

Record Breaker
Joined
Jul 26, 2003
I think the rules would be improved if they modified the requirements for positive GOE in such a way that an average jump would get 0 GOE. (Otherwise, what does "average" mean?)

How do you define "average"?

Does that include jumps with errors, or only "clean" jumps? What is the pool of skaters you're looking at?

In reality, I don't see that there is any alternative to some degree of pre-judging. You go into the competition already with a notion of what to expect. I expect Yuzuru Hanyu to get exceptional height on his triple Axel (and also to display a bunch of other positive features). So I already have my finger poised on the +5 button.

Isn't it easier to have your eyes and brain and pencil poised to take note of how many bullet points the jump achieved as executed today and then press the appropriate button than to try to remember -- for every jump by every skater -- what scores you would have given the best or most typical attempts yesterday and adjust from there?

Jump execution is so variable from one attempt to the next.

Watching practices might be good for taking note that Skater X often underrotates or changes edge, or Skater Y has difficult turns planned before and/or after the jump, so be on the lookout for those in competition.
Or skater Z is planning a very unusual jump combination (e.g., 3Lz+3F or 2S+2S with the first jump landing on the "wrong" foot and no Euler in between) or jump approach (a salchow that looks like it's going to be a lutz), so figuring out what she did in practice will prevent confusion if you saw it for the first time in competition.
Or Skater Q goes directly from a jump landing into a flying spin or directly from a spin exit into a jump, so don't look down or you'll miss the transition and entry.

I'd think the main thing watching practices would be good for would be getting a sense of program layout. Assuming skaters actually run through their full programs, with or without full jump content.
 

cake

On the Ice
Joined
Jan 30, 2007
Another proposal to make judging more transparent: For a certain number of elements or components the judges or the tech panel must disclose their reasons after competition. For example, each skater could choose one element/one component they want to be openly discussed. Okay - each skater is too much, so maybe each national team could choose one. Or a certain number could be drawn from skaters´s proposed elements.

Maybe even goldenskate could propose elements for which the judging process must be disclosed ;) (No, I´m not serious)
 

andromache

Record Breaker
Joined
Mar 23, 2014
Another proposal to make judging more transparent: For a certain number of elements or components the judges or the tech panel must disclose their reasons after competition. For example, each skater could choose one element/one component they want to be openly discussed. Okay - each skater is too much, so maybe each national team could choose one. Or a certain number could be drawn from skaters´s proposed elements.

Maybe even goldenskate could propose elements for which the judging process must be disclosed ;) (No, I´m not serious)

I would love this - maybe a few elements per event (by which I mean program per discipline) could be randomly selected? Such as a few StSqs, a few different spins of different types, a few jumps marked either <, !, or e, and a few clean jumps.

Even better would be to make judges explain their PCS!!!

It will never happen, but I can dream!
 

gkelly

Record Breaker
Joined
Jul 26, 2003
If there is a post-event roundtable discussion among the judges and referee, maybe the ref could give a press conference afterward highlighting the most interesting points of disagreement and discussion among the panel.
 

chaser

Rinkside
Joined
May 15, 2018
what if all the jumps were the same value except the axel? Would that make a difference?
Tech panels need attention of some sort.
Choreo Sequence should include x, y & z and cover all the ice surface.
Spins now look like pretzels in all forms. Not elegant and very few with pleasing-to-look-at body lines.

On the whole, the judges do a fair job, given the complications of the rules they have to apply to each element in real time. (adding up and subtracting bullet points from a scale that is labelled as good, very good etc. What exactly is "very good" height as an example?). Remember the bicycle analogy? I still think that was very ha-ha funny.

The entries to jumps don't seem to carry enough importance - pre rotation/edge changes - especially for lutz and flip. Edge jumps look to be less important.
 
Joined
Jun 21, 2003
Isn't it easier to have your eyes and brain and pencil poised to take note of how many bullet points the jump achieved as executed today and then press the appropriate button than to try to remember -- for every jump by every skater -- what scores you would have given the best or most typical attempts yesterday and adjust from there?

Well, I am not a figure skating judge, but to me I think it would be very, very challenging to take note of how many bullet points each jump satisfied, in real time before the program swept on to the next jump. I was thinking of major senior championships where the judges do pretty much know in advance what to expect from each skater, barring mistakes. Or if not on an individual basis, at least what we expect to see from skaters at a particular competitive level. Surprises can always happen, of course.

For competitions at a lower level, or where skaters participate that you have never seen before, I think it would be hard to say anything more than just, well that's wasn't bad -- a little better than the average jump. I guess i'll give it a +1.
But hats off to the judges if they can do better.

How do you define "average"?

That is the hard part. What I had in mind was a jump that takes off from the correct edge, has satisfactory air position, completes the rotations (or almost does), lands on an edge with reasonable flow -- but makes me say Eh? rather than Wow! Jumps that have some superior features and some weak ones, that's harder.

i think "average" would have to mean compared with the standard for the competitive level. An average jump by a Grand Prix contestant would be different from an average jump (0 GOE) for a novice or local club member.

This is more subjective (one judge's Wow is another judge's Eh) but this is the "quality" part of the judging, so I don't think that this can be helped (or that it need be lamented).
 

Baron Vladimir

Record Breaker
Joined
Dec 18, 2014
That was my point. This thread is not about what the current rules are but to make suggestions for improving the rules. I think the rules would be improved if they modified the requirements for positive GOE in such a way that an average jump would get 0 GOE. (Otherwise, what does "average" mean?)

But what an average means in a jump? How can you get to +5 then, and what is the point of +5 to exist if only one skater is able to acheive it in one exact way? Like i've already told if your jump felt as integral part of the music and integral part of the program you will get positive GOE, just because of that. If your jump is just a separate element but with very good height and distance, take off and landing it will get +2 too. I don't see something wrong with that. GOE is not just a technical part of jump itself, like components are not only an 'artistic' part of the program. And different skaters can achieve the same +2 GOE or 8.00 in components in different ways. Otherwise, skating competition would be boring, with all the skaters skating and performing the elements the same way to achieve that same mark. Now, i think GOE is reworked in a way it is, to make a better difference between very bad jumps which worth -4 or -5 and 'average' jumps which can worth 0, 1 or 2. The thing i heard is that ISU is thinking to up GOE in positives even up to +7, to make a difference between the elements who are achieving all the GOE bullets and elements which are achieving all but one GOE bullet, because now they worth the same.
 

cohen-esque

Final Flight
Joined
Jan 27, 2014
How do you define "average"?

Does that include jumps with errors, or only "clean" jumps? What is the pool of skaters you're looking at?

Setting aside the grading of the elements as a whole considering all aspects both positive and negative—there is a logical issue with the current GOE guidelines on some elements where this is no possibility of being “average” (or “adequate” or “acceptable” or “basic” of whatever) with respect to just a singular aspect. This an area the scoring guidelines could improve on, since it currently leaves some elements without a clear, logical baseline around which to score positive and negative GOE.

For example, take step sequences. A basic expectation of step sequences, explicitly laid out both in the rulebook and the technical handbook, is that they are to be executed in accordance to the music. Since this is a basic, minimal expectation, then for a step sequence to not match the music is a “serious error” for which starting GOE evaluations are capped.

But then, just achieving “element matched to the music” earns bonus points in +GOE for step sequences.

There is a binary choice here, where either a step sequence matches the music or it does not. Under the current guidelines, in regards to musicality, a step sequence can only ever be exceptionally good, or exceptionally bad. It can never just be...unexceptional, either way.

This skews the GOE in favor of positive, since skaters are explicitly rewarded for achieving the bare minimum. And it’s just illogical. If executing a step sequence in accordance to the music is a basic, prerequisite expectation, then why should doing so get rewarded with bonus points? And if doing a musical step sequence is instead above and beyond the basic expectations and so therefore worth extra points, then why would failing to achieve this be an error worthy of reduced points, instead of being simply adequate, and worth no bonus or penalty either way?
 

gkelly

Record Breaker
Joined
Jul 26, 2003
Well, I am not a figure skating judge, but to me I think it would be very, very challenging to take note of how many bullet points each jump satisfied, in real time before the program swept on to the next jump.

I think it would get easier with practice.

That is the hard part. What I had in mind was a jump that takes off from the correct edge, has satisfactory air position, completes the rotations (or almost does), lands on an edge with reasonable flow -- but makes me say Eh? rather than Wow!

So it's probably better to call that "adequate" rather than "average" and to award a 0 if that's all it is. Then if there is anything "good" rather than just adequate, that can be a plus.

Jumps that have some superior features and some weak ones, that's harder.

True. If you focus on one most obvious positive or obvious negative aspect of a jump, then you might miss the other negative or positive aspects. But again, you could improve with practice.

Even so, in that case you might get disagreements within the panel whether the positive outweighed the negatives or vice versa, with some judges scoring the element positive and some negative. A range of +1 to -1 was not uncommon in the old GOE range, so now with more and smaller increments, sometimes we might see +2 to -2 out of 5.

i think "average" would have to mean compared with the standard for the competitive level. An average jump by a Grand Prix contestant would be different from an average jump (0 GOE) for a novice or local club member.

There may be some mental adjustments taking that into account, especially adjusting for age and size of the skater. A "big" jump for a 70-lb 4'8" skater is going to be smaller in absolute terms than a big jump for a grown man.

But what do you do when you have a mix of skill levels in the exact same event?

How do you set an average for a JGP event that includes national champions with strong skating and triple axels and quads who may be too young for senior internationals competing against the best teen skaters from a tiny/new federation that has yet to produce anyone with any triples at all?

Not to mention pre-growth-spurt 13-year-olds (from either of the above skill categories or anywhere in between) competing against full-grown 18-year-olds who for whatever reason chose or were sent to compete as juniors?

Or a minor senior B event mostly catering to skaters who are just hoping to have a good enough day and generous enough tech panel to earn the Euros/4Cs minimum tech scores, along with a world medalist getting mileage on a new program before the Grand Prix.

Or actually at Euros/4Cs, where skaters who did just manage to squeak out the minimums might skate in the same short program warmup as several world medalists?

There is a binary choice here, where either a step sequence matches the music or it does not. Under the current guidelines, in regards to musicality, a step sequence can only ever be exceptionally good, or exceptionally bad. It can never just be...unexceptional, either way.

Good point, in this example.

Perhaps the basic minimum is that the movement more or less reflects the style and general phrasing of the music.

Negatives would be if it completely ignores or actively clashes with the music.

There could be separate positive bullet points for steps/movements skated in time to the musical rhythm throughout, or expressing multiple nuances of the music throughout (didn't the bullet point used to say "enhances the musical structure"?). So a skater who does one of the above can get one + and a skater who does both can get two. And then "creative or original" bullet point often applies when the choreography is specific to that particular music choice as opposed to just same old technical skating moves adapted to this phrasing, so there could be up to three pluses that relate to the music above and beyond just skating generally lyrically to lyrical music.
 

WeakAnkles

Record Breaker
Joined
Aug 1, 2011
Yeah right now there's just too much brass. I'd go with more strings. Strings and woodwinds. You can't go wrong with a lovely cello and oboe interlude. Maybe a cor angelis.

:hap57:
 

skatenewbie

Medalist
Joined
Mar 16, 2017
But that is not 0 GOE according to current requirements. If a jump is landed OK, that already can be +1 GOE. Actually if a jump doesn't have any jumping quality to itself, and its just matching the music it will get +1 GOE. With matching the music and being connected with the whole skating (by having transitions in and out) it will get +2. And for +5 you even don't need to have all the GOE bullets presented in your jump, except the core bullets. I mean, 0 GOE doesn't mean that jump is OK, it only means that there is nothing visually wrong with it :biggrin:
you need to get pass the bullet 1-3 first if you want to get to the bullet 4-5 which is music and creative entry/exit iirc. So if you dont tick all 3 first bullet, judges SUPPOSED to not give it bullet 4-5 even though the skater tick the bullet requirement. I believe the actual practice was not like this though 😂
 

Baron Vladimir

Record Breaker
Joined
Dec 18, 2014
you need to get pass the bullet 1-3 first if you want to get to the bullet 4-5 which is music and creative entry/exit iirc. So if you dont tick all 3 first bullet, judges SUPPOSED to not give it bullet 4-5 even though the skater tick the bullet requirement. I believe the actual practice was not like this though ��

Not really. You need to have all of the core buletts (1-3) to be able to get +4 or +5. But without meeting all of them, you can still get +2 or even +3, if you have any of two or 3 bullets. That's why Satoko for example can get +3 without very good height and distance, but she can't get more than that. Because she can't meet one of the core bullets (1-3) she is just capped at max of +3.
 
Top