Skaters who can dominate both in IJS and 6.0.

jaylee said:

With all due respect, Mathman, you're changing the argument. Whether 6.0 or IJS punishes falls/mistakes more or less is not the issue that we were discussing.

Click to expand...

I was responding to this post by Ven (#159).

To be blunt, I don't consider 6.0 FS a sport, but I do consider CoP figure skating a sport. With 6.0, it seems like it's just people making qualitative judgements on a skater's performance, and then assigning random made-up numbers to back up their bias who was better than who. Yuna could go out there and skate her 2013 World Championship Les Mis performance and there could be someone ... on the panel who could and would rate her below Kostner.

Click to expand...

My point was that under IJS there is still a lot of qualitative judging going on.

For an example, I tried to point out that Kim's Lez Mis in fact won on the strength of the quality that the judges saw in her program, not because of quantitative measures.

But the judges are not making up their own scores at the end of the program. Each element has a base value, and this value can either increase or go down based on how the skater performs the element. This is as close to objective you will ever get, or need to get for that matter.

Ven said:

This is as close to objective you will ever get, or need to get for that matter.

Click to expand...

This I agree with 100%. This is as close to objective as we will even get in a judged sport.

Mathman said:

I was responding to this post by Ven (#159).

Click to expand...

Yet you quoted my post #175 in your response, #176? http://www.goldenskate.com/forum/sh...in-IJS-and-6-0&p=745123&viewfull=1#post745123
:think:

My point was that judging under IJS there is still a lot of qualitative judging going on.

For an example, I tried to point out that Kim's Lez Mis in fact won on the strength of the quality that the judges saw in her program, not because of quantitative measures.

Click to expand...

Actually, it won because of both the quality and because of quantitative measures. A skater can't win on GOE alone.

jaylee said:

Yet you quoted my post #175 in your response, #176? http://www.goldenskate.com/forum/sh...in-IJS-and-6-0&p=745123&viewfull=1#post745123
:think:

Click to expand...

Sorry, I thought your post was continuing the exchange between Ven and me. I didn't catch the change of topic.

Actually, it won because of both the quality and because of quantitative measures. A skater can't win on GOE alone.

Click to expand...

True, but Kim gave it a heck of a shot. She clobbered everyone in GOEs and that was the ball game. :yes:

Don't get me wrong. This is a good thing. Kim won because she was better than all the rest, not because she was more than all the rest.

gkelly said:

The judges are not keeping track of what is required for higher levels. That's the tech panels' job. Most judges who aren't also controllers have not memorized what's required for each feature.

Click to expand...

Then they shouldn't be judges. It's very easy to look at spins and know what level it should be. If elements aren't being judged based up the actual quality, then the outcome of the competition may suffer. Which is exactly what we saw with Patrick Chan at this past Worlds.

Blades of Passion said:

Then they shouldn't be judges. It's very easy to look at spins and know what level it should be.

Click to expand...

The system is designed to divide the labor. The tech panels determine the levels. The judges determine the grades of execution, and also evaluate the program as a whole according to the five components. They're two separate tasks. If the judges were busy trying to do the technical panel''s job, they would have less attention available for the job they're actually tasked with doing.

gkelly said:

The system is designed to divide the labor. The tech panels determine the levels. The judges determine the grades of execution, and also evaluate the program as a whole according to the five components. They're two separate tasks. If the judges were busy trying to do the technical panel''s job, they would have less attention available for the job they're actually tasked with doing.

Click to expand...

If you don't know how to tell the level of a spin when you're sitting there staring at it for 10 seconds, then you shouldn't be a judge. Plain and simple. The tech panel is there to set a base value because there has to be a single value for all of the judges to go off of, since it would take too long otherwise and would be too chaotic to look at on the protocols.

Most judges (even those who are "only" Bronze or Silver level judges in the US where they are just getting familiar with IJS) understand the requirements for levels.

Blades of Passion said:

If you don't know how to tell the level of a spin when you're sitting there staring at it for 10 seconds, then you shouldn't be a judge. Plain and simple. The tech panel is there to set a base value because there has to be a single value for all of the judges to go off of, since it would take too long otherwise and would be too chaotic to look at on the protocols.

Click to expand...

Judges have far greater things to concern themselves with during a program than the level of spins. They constantly make notes and if they look down for a second they might miss the fact that a spin rotation wasn't held for required rotations, or they might be looking at a position change in a forward camel spin and not notice if the edge change was for 2 rotations. For footwork, they would certainly have their work cut out for them if they checked off which turns were executed and what level the footwork should be. The technical panel is a god-send, because you'd invariably get judges propping up their favourites - even in the technical mark - under 6.0, over skaters who were cleaner. At least under IJS, the skater will only get technical credit for what they execute, and then it's up to the judges to give bonus for quality elements -- but that bonus isn't so severe that it can bestow significant benefit to top/favourite skaters. Of course, it's questionable when -2's are given for falls but the nature of the TES scores is that the judges have lesser impact on the actual elements score, which is how it should be. Under 6.0, one could argue that the technical mark should theoretically be exactly the same from all the judges, and the subjective presentation mark is where the scores would differ.

CanadianSkaterGuy said:

JThe technical panel is a god-send, because you'd invariably get judges propping up their favourites - even in the technical mark -

Click to expand...

Now all we have to worry about is the technical panel propping up its favorites in URs, edge calls, levels, etc. -- and then the judges propping up their favorites in GOEs and PCSs.

Hopefully they would have different favorites and it would balance out.

And hopefully the 9 or however many judges on the panel would each have different favorites.

Not so, CSG, with 6.0 having the "theroretically the same technical mark for all judges for a given skater". The idea of 6.0 is that it is a relative score all based around the median mark (first skater) and that median mark is different for each judge. Basically, 6.0 is was the next skater better or worse technically and presentation-wise than each of the last skaters and by how much. If one judge starts their median mark at 3.0, 3.0 and another judge starts at 2.0, 2.0, then the tech mark is different for the best skater in the event based upon that median mark.

^ Thank you for explaining this "median" mark mskater. This is an unusual way to use the word "median," since for each judge it is the median of a set containing only one number. I remember in 6.0 competitions there was always a delay after the first skater performer to calculate the "median mark," which I assumed was the median of all the marks given by the judging panel for that skater, tech and presentation separately.

So my question is, what use were the judges expected to make of this information? Suppose the announced median of all the judges marks is 3.0. I, an individual judge, gave 2.0. Now the next skater goes and is a little better. Presumably I give 2.3 and the other, more typical, judges give 3.3. Is that right?

How does it help me, as an individual judge, to know that I am scoring more conservatively than my fellows?

gkelly said:

Hopefully they would have different favorites and it would balance out.

And hopefully the 9 or however many judges on the panel would each have different favorites.

Click to expand...

Indeed. The more the merrier.

Setting aside deliberate bias and outright cheating, under CoP scoring the statistical reliability of the results goes up only like the square root of the size of the judging panel. If we increase the panel from 9 to 16, say, that is an increase of 77% in the expense of fielding the panel, but the decrease in statistical error is only 25%.

Mathman said:

^ Thank you for explaining this "median" mark mskater. This is an unusual way to use the word "median," since for each judge it is the median of a set containing only one number. I remember in 6.0 competitions there was always a delay after the first skater performer to calculate the "median mark," which I assumed was the median of all the marks given by the judging panel for that skater, tech and presentation separately.

Click to expand...

Yes, that's what "median" referred to.

Judges then had the opportunity to change their marks for the first skater so they would all be calibrated closer to the median.

But then for all subsequent skaters they were on their own.

So my question is, what use were the judges expected to make of this information? Suppose the announced median of all the judges marks is 3.0. I, an individual judge, gave 2.0. Now the next skater goes and is a little better. Presumably I give 2.3 and the other, more typical, judges give 3.3. Is that right?

Click to expand...

Yes, if you choose not to change your mark for the first skater. If you did change it, then you would score the subsequent skaters in relation to the mark you changed it to.

If you gave 2.0 and the median was 3.0 (which meant half the rest of the panel was above 3.0), you might not want to change your score to 3.0 if you had a clear idea in your head of what a performance would need to deserve 3.0 and this performance was nowhere near that. But you might change it to 2.5 or something like that, so your scores wouldn't be so out of line when read out loud and you'd be less likely to get booed.

For competitions that didn't read the marks out loud, they didn't bother taking the median.

How does it help me, as an individual judge, to know that I am scoring more conservatively than my fellows?

Click to expand...

It doesn't, really, if you feel that you know what you're doing and you're satisfied that the marks you gave the first performance are exactly what you think that performance deserves, and that you'll have plenty of room above and below to fit in all the subsequent skaters.

Except that if you're scoring much lower you're likely to get booed often, but maybe you take pride in being the hardass and consider that a good thing.

It does help the panel as a whole if the range of scores for the first skater is relatively narrow and therefore the rest of the scores are also within the same general range and it looks as though the panel as a whole is more or less on the same page. And you're part of that panel. So if you want the panel to look cohesive, you'd adjust your initial score toward the median. If you think the rest of them are all wrong or want to look like a maverick, you'd leave it as is.

Also, knowing that you're much higher or lower than the rest of the panel might help you if you have less of an idea of what to expect from this field than some of the other judges. E.g., if you're judging in a new area or at a new level and haven't seen most of the skaters before, don't have much idea what the standard is likely to be at this event.

E.g., if this is an intermediate event and you're used to seeing intermediates with lots of clean doubles and several with double axels and maybe triples, when the first skater skates slowly with lots of cheated doubles you may think she's a weak intermediate and give her 2.0/2.0, expecting that this skater will finish near the bottom. If the median score lets you know most of the rest of the panel, who are more familiar with the standards in this region, all give scores in the high 2s and 3s, that would alert you that this might be one of the better performances you're going to see today after all, maybe no one else will land any clean double jumps either, and you'd better leave plenty of room below this skater fit in everyone else.

The same issue would come up at the top of the scale. You don't want to give 5.5 or higher to a clean junior or senior performance with a few easy triples, even if that's the best you ever see in your home area, and then find out that ten more skaters in this event are going to land 6 or 7 triples, triple-triple combinations, etc., making it hard to slot them each in the order you think they deserve.

(I'm just using jump content as an example -- the quality of the basic skating would be a consideration as well.)

gkelly said:

maybe you take pride in being the hardass

Click to expand...

Haha! I don't think I've ever seen gkelly use that kind of slang before.

Thank you, gkelly.

Except that if you're scoring much lower you're likely to get booed often, but maybe you take pride in being the hardass and consider that a good thing.

Click to expand...

Yeah, now yer talkin'. Boo this!

Actually. scoring at home, I have the opposite problem. The first skater out did great! Yay! 5.9. 5.9. Then the next one goes. Oops.

CanadianSkaterGuy said:

Judges have far greater things to concern themselves with during a program than the level of spins. They constantly make notes and if they look down for a second they might miss the fact that a spin rotation wasn't held for required rotations, or they might be looking at a position change in a forward camel spin and not notice if the edge change was for 2 rotations.

Click to expand...

Yes that's fine, let the tech panel determine exactly, but you can still tell approximately just by looking at it. EVERY lady holds one of the layback positions for at least 8 revolutions. That in itself (combined with the other required positions for the levels) shouldn't merit +GOE.

Mathman said:

So my question is, what use were the judges expected to make of this information? Suppose the announced median of all the judges marks is 3.0. I, an individual judge, gave 2.0. Now the next skater goes and is a little better. Presumably I give 2.3 and the other, more typical, judges give 3.3. Is that right?

How does it help me, as an individual judge, to know that I am scoring more conservatively than my fellows?

Click to expand...

It doesn't necessarily help you to know you are more conservative other than perhaps you might want to move your first skater mark up so you have more room at the bottom so to speak to separate your skaters. If you start low and this was one of the better performances you are going to see at that level and the other judges are more generous and knowledgable, it becomes difficult to find scores to differentiate below that median mark, especially if it's a big group!

You can see in the link here a group that the judges understand their median mark, even in a big group, and you can see that there are different things that these judges are each looking for and that not one skater hit on each of the judges' hot buttons, so to speak. http://www.usfigureskating.org/leaderboard/results/2013/68551/results.html

I had a judge explain 6.0 judging to me like this. This judge's favorite thing is corn. Judge #2 really likes carrots best. Judge #3 is only a meat and potatoes guy. Judge #4 only likes to nibble a little salad. Judge #5 ho hums her way through dinner to get to dessert. A person that can provide a well balanced plate that hits on each judges' favorite thing is going to get more high marks than someone who only serves up meat & potatoes, and corn and that person is going to get more high marks than someone who only drops off a dessert. In other words, it's best to be as well rounded a skater as possible and to have as few weaknesses as possible in order to place the best.