The Level Calls

[gkelly]

I know the differences between the school figure turns. As to difficulty, I can understand this at the juvenile and maybe novice levels, but juniors, and above all SENIORS should not have a problem otherwise how did they get to SENIORS?

It's been 18 years since school figures were removed from international competition.

Some of today's senior skaters hadn't even been born yet.

None of them has had to compete school figures. Depending what country they learned to skate in, they may have been required to pass tests demonstrating mastery of the advanced turns or they might not; different federations have different structures for determining when and how a skater qualifies for various competition levels up to senior.

Specific turns have never been required in freeskating programs (including short programs). They have always been considered in evaluating the difficulty of the step sequences and connecting moves.

Now there are specific requirements for the number of different kinds of turns needed to qualify for level 2, 3, or 4 step sequence.

We can debate the specific choices and definitions of the step sequence features. But the fact that they are spelled out certainly makes establishing the relative difficulty of various sequences more objective than whatever each judge happens to notice and prefer to reward.

 

[mskater93]

Since you are clearly displeased with the current state of the sport and none of these explanations seem to satisfy you, perhaps you'd feel better if you actively did something about it? If "Difficulty only occurs with bad schooling together with bad training," why not start coaching?

 

[Joesitz (RIP)]

vlaurend indeed perceives me that way. and apparently you do also. You both have that right to your perceptions. But I can not change my view of Senior skaters. They worked hard to get to Seniors and mastered the basic connective steps of the basic terminology in every figure skating book. There are natural skaters who can flow with rockers and counters, and others must work with a choreographer to get the flow. That should not be difficult at the Senior level.

Also specific turns were never required in figure skating until the CoP came along, but the skaters during those pre-CoP years knew, learned and executed all those turns. Check out some of those pre-Cop skaters doing every turn, small jumps, and body movements on youtube. Some are better than others, but that has a lot to do with the choreographer, imo.

Difficulties must be overcome in every Sport. Why do we want to make figure skating an easy sport?

I am not against the CoP, but I am not overwhelmed with it either and I am definitely not in favor of Levels. Anything the skater does on the ice should be obvious for scoring. No need for a separate tech panel to show their preferences. Let the judges show theirs. And most of all let the Skaters do what they want to do.

 

[ChrisH]

When computer system is used, the judges do not see the level calls, only the type of element called ( Ccosp, Slst, etc).

I'd rather that the judges see the levels so that if they are undecided between the lower or higher of two GOEs, they can adjust to a stingy or generous level call. Remember that for jumps, judges must see levels, downgrades, and edge calls because they are required to adjust their GOEs. A wise judge, IMO, would moderate the technical calls.

 

[Joesitz (RIP)]

I'd rather that the judges see the levels so that if they are undecided between the lower or higher of two GOEs, they can adjust to a stingy or generous level call. Remember that for jumps, judges must see levels, downgrades, and edge calls because they are required to adjust their GOEs. A wise judge, IMO, would moderate the technical calls.

That certainly would help. All these judges and tech specialists sit in the same row with each other. They all see the same moves that the skaters do. They have all passed their judgeship tests, and some are former skaters who know exactly what is difficult and what is not. The Tech Specialists were ingrained into the way the ISU deems whether a move is difficult or not as compared with the judges who already know that without the ISU input.

I just do not see the need for Levels.

 

[Mathman]

I just do not see the need for Levels.

I see what you are saying. But at the same time, if a person buys into the whole idea of giving out points for "base values," I can see why they want to have a "base value" for footwork, spirals, etc., just like for jumps.

One of the criticisms of the 6.0 system was that jumps were everything. The "balanced program" idea seems to be that if you get base points for jumps then you should get points for other things, too. A level 4 step sequence earns 3.9 base points -- more than a double Axel (3.5) and just shy of a triple toe (4.0).

If you say, "well, senior skaters should be able to do this stuff in their sleep, why give them base points for it?" -- that is putting the whole tech score back on the jumps again. If you just skate down the ice doing little or nothing, that's level 1 and you get 1.8 points for just showing up (example: Yagudin in Winter, LOL.) If you do a bushel of different steps and turns, you get more.

On top of that, the judges can still add or subtract points for doing it well or poorly. Whether this is a good system or bad is a separate question, but at least it seems like the way the CoP handles step sequences is consistent with how it deals with other elements like jumps and spins.

Now if your argument is that we should eliminate the tech specialist's job altogether, letting the judges judge and the skaters skate -- basically returning to ordinal judging -- that position has merit, IMHO, although it takes some of the "sport" out of the sport.

 

[gkelly]

Another possible way to do it would be to give one base mark for all step sequences and let the judges add a possibility of more than 3, let's say 5, bonus points both for difficulty (now covered by the levels) and quality (now covered by GOEs). I think that's more or less what Joesitz is proposing.

The potential problem I can see with that approach is that it would be less accurate/more subjective.

There wouldn't be time for each judge to do the job that's currently done by the three-person tech panel to keep track of all the difficulty in the step sequence -- it would be especially difficult for judges who weren't skaters themselves and don't have the muscle memory of the different turns in their own bodies -- and also do the job that they currently do in evaluating quality. At events where video replay is available, if they each had to rewatch all the step sequences one or more times, that would slow down the wait for the marks even further than is currently the case.

And if they aren't given specific guidelines to reward difficulty such as the current level feature definitions, then you would often see huge differences of opinion between judges as to how difficult each sequence was.

If you just skate down the ice doing little or nothing, that's level 1 and you get 1.8 points for just showing up.

Well, if you just skate down the ice doing nothing, it won't be called as a step sequence, so you'll get no points.

How little is too little to be recognizable as a step sequence? E.g., the opening pass in Kwan's Lyra Angelica would probably be a borderline case.

If you do a bushel of different steps and turns, you get more.

However, you have to do the right combination of different kinds of steps and turns plus at least one other feature to earn even level 2. Some fairly difficult step sequences both currently and from the pre-COP era end up getting called level 1 because they're difficult in the wrong ways according to the rules.

A solution to that would be to make the level rules more flexible. E.g., make something like "at least half the sequence performed all on one foot" a feature, or make the variety of steps and turns feature once again no longer mandatory.

Or just accept that the ISU has determined that variety of steps and turns is important enough that they won't give extra difficulty points for a step sequence that doesn't have that variety.

In which case you could choreograph the sequence knowing it'll be level 1 and use it to earn higher GOEs and higher PCS.

 

[Mathman]

Another possible way to do it would be to give one base mark for all step sequences and let the judges add a possibility of more than 3, let's say 5, bonus points both for difficulty (now covered by the levels) and quality (now covered by GOEs).

This would be the same as eliminating base value altogether. Since it is a competition of one skater versus another, if you gave every skater 0 points base value or 1000 points base value, it wouldn't affect the outcome.

 

[gkelly]

This would be the same as eliminating base value altogether. Since it is a competition of one skater versus another, if you gave every skater 0 points base value or 1000 points base value, it wouldn't affect the outcome.

Well, it would make a difference between skaters who do a recognizable step sequence and those who don't. Plus you have to have some kind of base to add the bonus points onto.

 

[Joesitz (RIP)]

But what makes the Tech Specialists more special than the judges at this?

(Not sure how the base values would be affected by eliminating the Levels. There are clear definitions for each element. Doesn't the GoEs come in to play?)

 

[Mathman]

Not sure how the base values would be affected by eliminating the Levels.

If levels were eliminated, there would be only one base value for your step sequence no matter which steps and turns your sequence featured.

The way it is now, your step sequence starts out with a base value that depends on which level is called.

Level 1 base value = 1.8
Level 2 base value = 2.3
Level 3 base value = 3.3
Level 4 base value = 3.9

If we changed the system and eliminated levels, then there would be a single base value for a step sequence. Say eveyone gets a 2.0, no matter how many different turns, etc., they do. Then the judges add or subtract GOEs.

There are clear definitions for each element.

Although there are clear definitions for what a rocker is, what a bracket is, etc., these elements do not recieve any base value themsleves.

Same for spins. There is a clear definition of what it means to change edge, but changing an edge does not receive a base value. It does, however, count toward the level. If you do a whole bunch of changes of edges and other such things, you get a higher level, hence a higher base value, for your spin as a whole.

Doesn't the GoEs come in to play?

In principle, the GOEs are not supposed to come into play at all. The level, and it's corresponding base value, is supposed to refelct only what the skater did. The GOEs are supposed to be about how well they did it.

In practice, I am not sure how much the judges are influenced by level calls. Chris H argues above that it is a good thing for judges to be able to "correct" the tech panel's call by giving an extra high GOE if the judge thinks the tech panel erred in giving too low a level. It is not intended to be this way, but judges and tech callers alike are human.

 

[mskater93]

The judges don't see the level calls for spins/steps/spirals so only judge quality.

The idea of the technical panel - Controller, Specialist, Assistant Specialist - is that they are not beholden to a federation like judges *may* be and so are more likely to be unbiased and calling based on the rules they are given so that judge 1 doesn't give a step sequence a L4 because the skater is from their home country while judge 7 gives it a level 1 because it's their home country's skater's biggest rival.

 

[antmanb]

The judges don't see the level calls for spins/steps/spirals so only judge quality.

The idea of the technical panel - Controller, Specialist, Assistant Specialist - is that they are not beholden to a federation like judges *may* be and so are more likely to be unbiased and calling based on the rules they are given so that judge 1 doesn't give a step sequence a L4 because the skater is from their home country while judge 7 gives it a level 1 because it's their home country's skater's biggest rival.

While everything you have written is true, the reality is that a tech specialist who is a particular nationality may call things in a biased way towards skaters from the same country, if the two others on the panel don't challenge it, it simply goes down as fact. Whereas 12 judges making the level calls would more likely, as a majority, end up with the correct levels.

Ant

 

[Joesitz (RIP)]

I'm confused. The skater begins a step sequence for his footwork. Which scoring comes first? The Tech Callers for announcing the Level or the Judges for their scores? And how do the twains meet for a final score?

 

[mskater93]

JOE:
The level call happens as the step sequence (or spin or spirals) ends by the tech panel (or a review is called by someone on the panel who doesn't agree with what has been called and it will be reviewed at the end to make sure all are in agreement after the program is over). The judges grade GOE for each element during the performance unless the tech panel decides to downgrade a jump or give an "e" call upon review in which case the judge can change his/her GOE. Everything gets entered into the system during the program (or it is changed after reviews) and is accepted at the end of the program and reviews. Once everything has been entered and accepted by the tech panel (and the judges have entered all GOEs and PCS scores), the program crunches it all together and spits out the Base Value, the Scores of Panel for TES and the Scores of Panel for PCS multipled by the factor. Judges never see the level called for spins/spirals/steps.

ANT:
I am only giving the ISU party line as to why a tech panel was created versus having the judges do their own calls as was one of the questions. I did not state my opinion as to what I think of it.

 

[gkelly]

I'm confused. The skater begins a step sequence for his footwork. Which scoring comes first? The Tech Callers for announcing the Level or the Judges for their scores? And how do the twains meet for a final score?

Here's how I understand it.

The tech callers call it a step sequence of whatever general shape (it may take half the sequence before they're sure what its shape is). They also assign a level at the end of the sequence, after they know how many and which features it fulfilled. They may also call "review" and then go back after the end of the whole program and rewatch the sequence to verify exactly what the skater did to confirm that they assigned the correct level according to the rules.

I believe the person who's actually doing the calling out loud will say the parts of the name of the element in the order that they know them. E.g., "step sequence, circular, level 2."

So by the end of the sequence the data input operator has input the whole code for the sequence: CiSt2. After review at the end of the program the tech panel might determine that the skater actually did meet all the requirements for level 3 and change the call to CiSt3 -- or, more likely, the other way around.

What shows up on the judges' computer screens is only the code for the general type of sequence without the level, e.g., CiSt.

While they're watching the sequence, the judges recognize that it is a step sequence and are evaluating its quality, thinking about what GOE they think it deserves. They can't finalize their decision about the GOE until the sequence has ended, because what if the skater falls on the next-to-last step? That would change the GOE.

The judges are not going to know if the tech panel initially considered the sequence to qualify for level 3 and then changed it to level 2 because replay revealed that the intended counters were really three turns preceded by edge changes. They'll only know whether the turns looked to them to be well performed or sloppy/incorrect and will take that into account in evaluating the quality.

The input into the computer from the tech panel will be CiSt2, and that will determine the base mark. Base mark for CiSt2 is 2.3.

For easy math, let's say there are 7 judges. The input from the judges might be GOEs of 0 0 1 0 0 -1 1. The computer will throw out the high and low 1 and -1, leaving 0 0 0 0 1. According to the scale of values, GOEs for CiSt2 are worth 0.5 for each +1. So you get 0+0+0+0+0.5 = 0.5, average that across the 5 scores after trimming, and the value of the panel's average GOE for that element will be 0.1.

So the final value for this step sequence would be 2.3 + 0.1 = 2.4.

 

[vlaurend]

The judges don't see the level calls for spins/steps/spirals so only judge quality.

The idea of the technical panel - Controller, Specialist, Assistant Specialist - is that they are not beholden to a federation like judges *may* be and so are more likely to be unbiased and calling based on the rules they are given so that judge 1 doesn't give a step sequence a L4 because the skater is from their home country while judge 7 gives it a level 1 because it's their home country's skater's biggest rival.

Exactly. In addition, the technical specialists have received extensive specialized training and had to pass tests to demonstrate their abilities to call difficulty levels. While some judges have also been certified as technical specialists, many have not, and would therefore not be qualified to determine level of difficulty. Also, while the judges do not have to agree on the marks and/or GOE's they give a skater, the technical panel must present a single set of protocols for each skater that has been agreed upon between them via a prescribed process. Keep in mind that the members of the technical team may be from different countries, which would make it more difficult for a single member of the technical team to make calls based on national or personal bias.

 

[Mathman]

Do technical specialists have to be nominated by their national federations, as is the case for judges? Or does the ISU tap people independent of input from the federations?

Tne reason that judges cannot stray too far from the wishes of their federation chiefs is, as I understand it, that if they did, the federation could deny them further assignments as judges by withholding their names. That is, for each event there is a random draw to determine which country gets to send a judge, but the actual individual is selected by the federation from the list of ISU qualified judges from that country. Is this correct?

Is it the same for tech specialists, or could the tech specialist blow off his or her national federation by being tight with Speedy?

 

[mskater93]

No, not nominated, tapped by ISU to be World level tech team member.

 

[Wrlmy]

I have one question. I don't know this has been discussed or not, but why is it that we see so many spiral sequences called lv1 or lv4 but not lv2 or lv3?

I've seen so many descent spirals, which would be called lv4 under most circumstances, called lv1 just because of one tiny error such as holding a position for only 2.99 seconds.