Scoring bias at the national level

[Jumping_Bean]

If you'll look at my previous post to me it still sounds that he was found guilty of "not judging similarly with the other judges", and I would rather have a commission take element by element and pointing out exactly which element/component was judged incorrectly and why, instead of referring his total score being higher than the average.

Except that doesn't necessarily help in many cases. It definitely doesn't help in deliberate scoring manipulation - These judges know what they're doing and are going to give GOEs & PCS that are a little bit higher (or lower, depending on the goal) but not unjustifiably so, so that they don't get flagged automatically. They also hold other skaters to different standards and different GOEs for comparable execution, but of course, all within the judging corridor. (And before you suggest that they should then just review every single element and compare across skaters for the same judges, that's not feasible. The OAC reviews 20 competitions each season, with 3 or 4 disciplines each with 9 judges, and competitor numbers at times breaching 40).

For example, let's say two skaters have flat edges (that get called) on both Lutz and Flip, the jumps are otherwise similar in quality, and the skaters have the same layout. The mean GOE given on all four jumps is 0, and a judge gives one skater +1 and the other -1 on all of them - They won't be flagged, as there's no significant deviation. But at the same time, on those four elements alone, this judge has managed to "achieve" a TES difference of 4.48 points. Do that on every element, and you'll get a lot more of a difference. And at the same time, you did nothing wrong. At least on paper.

And if a judge is just scoring everyone differently, it is a very big coincidence for three skaters of the same country to be scored differently in only one direction (and significantly so). Very big coincidence, almost makes you think, no?

Let's take Isabeau's FS scored by Doug Williams as an example - There are no callable errors (apart from the 9.5 in composition, because she fell), but his score is almost 10 points higher than the average score. For Amber, it's the same thing - No actual errors, but 10 points higher in total.

 

[Mathman]

What the PCS were intended to measure were various aspects of the program as a whole.

The elements, such as jumps, are considered elements and are reflected in the Total Elements Score

I guess what I am trying to get at is that when it comes to the program as a whole it is not so easy (for me anyway) to separate out the "aspects," one from another. If you have good skating skills then you can put those skills to use in mastering choreography that is more intricate and better constructed. Those same skills are what gives you confidence and success in projecting to the audience. For that matter, if you have strong blade to ice skills you can put those skills to the service of your jump technique.

I am never surprised to see all program components, whether 3 or 5, go up.or down together. I can even understand why an observer might say, this guy just did 5 quads -- his skating skills must be pretty good.

 

[4everchan]

My working thesis at the moment is that giving higher marks to most and having a generally bigger difference in scores can be used as a technique to disguise the pushing of your favoured team a bit. Would have to look at that much closer, though.

how so? either way does the same... giving lower scores but a bit lower to other teams would do exactly the same.

my hypothesis would be the following : judges who actually want to play the system would know exactly how to do it to avoid "mistakes" and looking suspicious... until they are caught... like the OP case.

 

[icewhite]

how so? either way does the same... giving lower scores but a bit lower to other teams would do exactly the same.

my hypothesis would be the following : judges who actually want to play the system would know exactly how to do it to avoid "mistakes" and looking suspicious... until they are caught... like the OP case.

well, if there is nothing to be suspicious about it's probably because they are if not good at least fair judges
I think the problem lies with those who are suspicious, but it's difficult to draw the line or pinpoint.

I would think though that if someone gives out generous scores all the way people are less likely to get excited, find fault/ look closer - while more negative scores might immediately draw the attention not of a computer but fans or related ones of the lowballed skaters, so personally I'd probably be more inclined to be generous if I was trying to manipulate.

All of this requires much more detailed analysis than I without any mathematical knowledge can probably provide. But I find it interesting...

 

[Baron Vladimir]

I guess what I am trying to get at is that when it comes to the program as a whole iit is not so easy (for me anyway) to separate out the "aspects," one from another. If you have good skating skills then you can put those skills to use in mastering choreography that is more intricate and better constructed. Those same skills are what gives you confidence and success in projecting to the audience. For that matter, if you have strong blade to ice skills you can put those skills to the service of your jump technique.

I am never surprised to see all program components, whether 3 or 5, go up.or down together. I can even understand why an observer might say, this guy just did 5 quads -- his skating skills must be pretty good.

Exactly.. Thats why for majority of skaters there is not big difference between different components... But lets say half a point difference or more between them can tell skaters what they need to work on in their programme for the next performance. For example, in Trusovas case, her TR score was bellow 8, while her SS were 8.65, that being a feedback from the judges what she needs to work on if she wants better components score...

 

[4everchan]

well, if there is nothing to be suspicious about it's probably because they are if not good at least fair judges
I think the problem lies with those who are suspicious, but it's difficult to draw the line or pinpoint.

I would think though that if someone gives out generous scores all the way people are less likely to get excited, find fault/ look closer - while more negative scores might immediately draw the attention not of a computer but fans or related ones of the lowballed skaters, so personally I'd probably be more inclined to be generous if I was trying to manipulate.

All of this requires much more detailed analysis than I without any mathematical knowledge can probably provide. But I find it interesting...

Here is what I will share with you : i am the opposite. I don't notice usually lower scores as much as higher scores... because, like in the example you brought up from 4cc... quickly, I could see the twist as a problem... it should be negative goe and it wasn't.

Same when there are falls that do not get -5
Same when there are falls and PCS is not affected (remember Chock and Bates fall at worlds yet their PCS didn't get the cap requested from some judges).

So personally, I notice a lot more the outliers that are high.

I am sure some notice them both. Some look at scores more than I do.

So your hypothesis doesn't work for me in that sense because my wiring tends to look at bigger scores as suspicious... probably because in my field, scoring is low. There is no perfection when results are judged subjectively in part. Though, when I would take a maths class, getting 100% on a test was possible and happened. So, when I see 10.0 or +5 I am always suspicious. Was this really worth a perfect score ? Same.. if I see a + goe on a fall or if a judge marks much higher than others... I get suspicious...

Just my personal reflexes but they are out there... and I really stand by what I said, the best way for a judge to really favour their skaters would not be by making "errors" like Fortin did... being the best judge and hiding their biases in there like Williams.

 

[gkelly]

Same when there are falls that do not get -5

If the GOE is -3 or -4 for an element with a fall, this is probably not a judging error but an intentional reflection that there were multiple positive aspects to the element before the fall occurred (or after, in the case of a step sequence).

The rules/guidelines are that judges should start by awarding any positive bullet points. If there is a fall (or specified other serious errors), the positives can't start at higher than +2, so the maximum final GOE after a fall deduction would be -3 or -4.

If you see -3 or -4 for an element with a fall, rather than being suspicious of the judge's motivations, it would make more sense to look for what else was good enough about the element that the judge(s) could still reward despite the error.

Only if you find a pattern of a judge consistently giving those generous points to skaters from their own country, including when no other judges on the panel do the same, and never to rivals, would it really be cause for suspicion.

So if you like to search for bias, you could flag such scores as potential data points in such a pattern. But on their own they could just as easily be evidence of good judging according to the rules.

Do you also like to search for excellence in the skating? Or would you be one of those judges who's always searching for reasons to mark down and less inclined to reward the positives, despite the explicit ISU guidelines to find the positives first?

So your hypothesis doesn't work for me in that sense because my wiring tends to look at bigger scores as suspicious... probably because in my field, scoring is low. There is no perfection when results are judged subjectively in part. Though, when I would take a maths class, getting 100% on a test was possible and happened. So, when I see 10.0 or +5 I am always suspicious. Was this really worth a perfect score ? Same.. if I see a + goe on a fall or if a judge marks much higher than others... I get suspicious...

Some judges are more pessimistic than others.

But I think all of them want to see great skating and great elements. They would like to see skating strong enough to reward with +5 or 10.0. So when they do see something really special, they may get excited that they finally get a chance to give those top marks, that they are privileged to have witnessed such an excellent element or excellent overall performance and want to mark the occasion with a top score.

Unlike the 6.0 system, +5 and 10.0 are not defined as "perfect and faultless."
The +5 GOE is supposed to include positive bullet points with no reductions. 10.0 is defined as "Outstanding" and must have no errors.

So it's perfectly reasonable to look at an element or a performance that received those top scores and ask why/how it stood out so much from other strong elements/performances.

If pretty much all the judges gave at least +4 or 9.5, that shows general agreement that that really was something special and it would be curmudgeonly score lower.

If only one judge gives an outlier score that high, that could be reason to investigate further. Especially if they only give such scores to compatriots.

 

[4everchan]

If the GOE is -3 or -4 for an element with a fall, this is probably not a judging error but an intentional reflection that there were multiple positive aspects to the element before the fall occurred (or after, in the case of a step sequence).

The rules/guidelines are that judges should start by awarding any positive bullet points. If there is a fall (or specified other serious errors), the positives can't start at higher than +2, so the maximum final GOE after a fall deduction would be -3 or -4.

If you see -3 or -4 for an element with a fall, rather than being suspicious of the judge's motivations, it would make more sense to look for what else was good enough about the element that the judge(s) could still reward despite the error.

Yes. I have witnessed this and I am okay with a minus 3 or 4... But higher than that, is highly suspicious

I mean, we often see minus 5 for a step out...

My wish would be an automatic minus 5 for the fall. No need for a judge to find positives in a failed attempt.

Only if you find a pattern of a judge consistently giving those generous points to skaters from their own country, including when no other judges on the panel do the same, and never to rivals, would it really be cause for suspicion.

So if you like to search for bias, you could flag such scores as potential data points in such a pattern. But on their own

Do you also like to search for excellence in the skating? Or would you be one of those judges who's always searching for reasons to mark down and less inclined to reward the positives, despite the explicit ISU guidelines to find the positives first?

I look for excellence but do not believe in perfection. Two different things.

Some judges are more pessimistic than others.

But I think all of them want to see great skating and great elements. They would like to see skating strong enough to reward with +5 or 10.0. So when they do see something really special, they may get excited that they finally get a chance to give those top marks, that they are privileged to have witnessed such an excellent element or excellent overall performance and want to mark the occasion with a top score.

I understand but I don't necessarily like it. Let's say that a short program doesn't go well... Many falls and pops and bad spins.. then someone does a decent clean job.. that skater shouldn't get all the points compared to the others who were doing so badly. There should be a comparison with the level. If the combo was landed well but without much panache, it shouldn't get +4/5 just because everyone else popped theirs. The judges should compare the combo with a very good example. Otherwise, it becomes 6.0/ ordinals judging.

Unlike the 6.0 system, +5 and 10.0 are not defined as "perfect and faultless."
The +5 GOE is supposed to include positive bullet points with no reductions. 10.0 is defined as "Outstanding" and must have no errors.

So it's perfectly reasonable to look at an element or a performance that received those top scores and ask why/how it stood out so much from other strong elements/performances.

If pretty much all the judges gave at least +4 or 9.5, that shows general agreement that that really was something special and it would be curmudgeonly score lower.

Sometimes, in dance, we see such scoring of the top 4/5 teams... Yet, there have been mistakes... So I have a hard time with this. We don't even have to look far for this last year's worlds had interesting and creative scoring in the free dance.

If only one judge gives an outlier score that high, that could be reason to investigate further. Especially if they only give such scores to compatriots.

Yes.

 

[icewhite]

One major problem I have not found a solution for is the background of the skaters and the judges and the style of skaters.
I have looked at two judges closer, one I find very suspicious (A), one I don't find suspicious - the second one (B) seems to have a preference for a certain style of skaters - she gets away from the corridor more often than A. But with what I know about these skaters and looking at their nationality, coaches etc. I cannot see a preference of a certain background, but a preference for "beautiful"/ "complete" skaters. Looking at the skaters that A rated significantly higher than other judges I think that they do not have much in common as skaters, but tend to have certain coaches. Also, when A scores athletes lower than the average judge does (in absolute numbers or placements) it benefits skaters with a certain nationality.
But how do I put this in an algorithm? - There are ways for the nationalities and coaches, but for the style of the skater? I would have to have much more data at hand which proves in numbers what I can see as a viewer.

 

[4everchan]

One major problem I have not found a solution for is the background of the skaters and the judges and the style of skaters.
I have looked at two judges closer, one I find very suspicious (A), one I don't find suspicious - the second one (B) seems to have a preference for a certain style of skaters - she gets away from the corridor more often than A. But with what I know about these skaters and looking at their nationality, coaches etc. I cannot see a preference of a certain background, but a preference for "beautiful"/ "complete" skaters. Looking at the skaters that A rated significantly higher than other judges I think that they do not have much in common as skaters, but tend to have certain coaches. Also, when A scores athletes lower than the average judge does (in absolute numbers or placements) it benefits skaters with a certain nationality.
But how do I put this in an algorithm? - There are ways for the nationalities and coaches, but for the style of the skater? I would have to have much more data at hand which proves in numbers what I can see as a viewer.

and the style can be shared by a few countries.

As a skating fan, I tend to like certain skaters... and it's not limited to one nation... but perhaps 4-5 nations produce more of these skaters... not always, but more often than some other nations...

So, if I were a judge, you would probably notice that I score favourably a style of skating. It happens that some Canadian athletes share this style (not all of them obviously)...

I call this taste, not national bias LOL

 

[icewhite]

and the style can be shared by a few countries.

As a skating fan, I tend to like certain skaters... and it's not limited to one nation... but perhaps 4-5 nations produce more of these skaters... not always, but more often than some other nations...

So, if I were a judge, you would probably notice that I score favourably a style of skating. It happens that some Canadian athletes share this style (not all of them obviously)...

I call this taste, not national bias LOL

I want to have your trust in people.

Well, you stated you don't want to talk about this person, but isn't it a bit odd how I'm casually scrolling through recent event scores and when I google the first judge I find suspicious they turn out to be politically very problematic and accused of basically corruption? Either that's a big coincidence, or the rate of such judges is pretty high or there's something real behind my intuitive alarm going on.

 

[gkelly]

Yes. I have witnessed this and I am okay with a minus 3 or 4... But higher than that, is highly suspicious

I mean, we often see minus 5 for a step out...

The reduction for a step out is -3 to -4. So when there's a -5, there was probably something else wrong with the jump.

My wish would be an automatic minus 5 for the fall. No need for a judge to find positives in a failed attempt.

Even for something like this? https://www.youtube.com/watch?v=iIdqjryYbQI&t=1m08s

I look for excellence but do not believe in perfection. Two different things.

Fortunately for the skaters, the scoring system does not require perfection in order to earn top scores.

Under the former judging system, the top score of 6.0 was defined as "perfect and faultless" -- meaning, originally, drawing a perfect circle on the ice, with no expectation that humans would approach that perfection. Even scores in the low 5s were rare for figures.

In free skating, 6.0s were sometimes used, and 5.9 often. Of course, in that system, the actual scores were only placeholders.
And it was not unknown for skaters to earn 6.0 in one mark or the other even with one visible error.

Under the current IJS rules, 10.0 is defined as "Outstanding" with "No errors."

For +5 elements, all that's needed is 5 positive bullet points and no negatives. On that single element.

I understand but I don't necessarily like it. Let's say that a short program doesn't go well... Many falls and pops and bad spins.. then someone does a decent clean job.. that skater shouldn't get all the points compared to the others who were doing so badly. There should be a comparison with the level. If the combo was landed well but without much panache, it shouldn't get +4/5 just because everyone else popped theirs. The judges should compare the combo with a very good example. Otherwise, it becomes 6.0/ ordinals judging.

I'm not talking about waiting all afternoon to see a clean program.
With 10.0 on one or more components, I'm talking about waiting across a 20- or 50-year judging career to see a performance for the ages.

If the commentators are making comments like "We're so lucky to be able to see this!" (especially about skaters not from the same country as the commentator), wouldn't we expect judges to be similarly impressed and to score accordingly? Some years they might be lucky enough to be on the panel for several such performances. Other times they might go years between being so fortunate.

+5s on individual elements would be more common. Did the element meet all three of the mandatory bullet points, including effortlessness? Did it also have two more positives? If the judge can add those up, it is appropriate to award the +5. On a great day of skating, there might be multiple skaters earning multiple +5s.

If there are multiple skaters with popped combos and then one skater finally lands a clean combo, the judges could award anywhere from 0 to +5 depending on the quality of the element successfully completed. If it just meets the requirements but has no positive qualities other than being clean, then the judges would be happy to give the 0s it deserves, after all the -5s and *s for the other skaters.

Or suppose a skater starts the program with their best element, meets the necessary criteria on that element and earns +5s there, and then falls on the next element and it's all downhill from there. That wouldn't cancel out the excellence of the first element.

 

[Mathman]

Even for something like this? https://www.youtube.com/watch?v=iIdqjryYbQI&t=1m082

Proving that no rule book can ever be so capacious as to anticipate and provide guidance for every possible thing that can happen on the ice.

OK, OT but I can't help myself. https://www.youtube.com/watch?v=_qqHEd0Bq9o

 

[Diana Delafield]

Proving that no rule book can ever be so capacious as to anticipate and provide guidance for every possible thing that can happen on the ice.

OK, OT but I can't help myself. https://www.youtube.com/watch?v=_qqHEd0Bq9o

Off topic, but as an old (in both senses) pair skater, for posting that glorious performance. I'm swooning. That last big lift's entrance must have felt wonderful to do!

 

[4everchan]

The reduction for a step out is -3 to -4. So when there's a -5, there was probably something else wrong with the jump.

there often is

Even for something like this? https://www.youtube.com/watch?v=iIdqjryYbQI&t=1m08s

this is probably the exception to confirm the rule but then, you know, I am one of these people who would be OK with the following rule :
you fall, the element is invalid. I mean think about it... there is a zayak rule where a skater can triple a quad and then later on, lose his+3t combo because one too many 3t.... so that jump = 0... or a skater failing to tag a combo can get a REP on a very perfectly well landed quad... So... why should there be points given to falls ?

Fortunately for the skaters, the scoring system does not require perfection in order to earn top scores.

I get that... but then, in the end, giving a 10 does mean a perfect score right? because nobody can get a higher score. So if the element doesn't require perfection, the score awards it.

Under the former judging system, the top score of 6.0 was defined as "perfect and faultless" -- meaning, originally, drawing a perfect circle on the ice, with no expectation that humans would approach that perfection. Even scores in the low 5s were rare for figures.

In free skating, 6.0s were sometimes used, and 5.9 often. Of course, in that system, the actual scores were only placeholders.
And it was not unknown for skaters to earn 6.0 in one mark or the other even with one visible error.

Under the current IJS rules, 10.0 is defined as "Outstanding" with "No errors."

For +5 elements, all that's needed is 5 positive bullet points and no negatives. On that single element.

yes I am aware of that.

I'm not talking about waiting all afternoon to see a clean program.
With 10.0 on one or more components, I'm talking about waiting across a 20- or 50-year judging career to see a performance for the ages.

If the commentators are making comments like "We're so lucky to be able to see this!" (especially about skaters not from the same country as the commentator), wouldn't we expect judges to be similarly impressed and to score accordingly? Some years they might be lucky enough to be on the panel for several such performances. Other times they might go years between being so fortunate.

fair.

+5s on individual elements would be more common. Did the element meet all three of the mandatory bullet points, including effortlessness? Did it also have two more positives? If the judge can add those up, it is appropriate to award the +5. On a great day of skating, there might be multiple skaters earning multiple +5s.

In ice dance, we can see +5 quite often. Perhaps in some lifts in pairs but hot often to we see +5 for jumps or spins or steps ? I am curious about this.

If there are multiple skaters with popped combos and then one skater finally lands a clean combo, the judges could award anywhere from 0 to +5 depending on the quality of the element successfully completed. If it just meets the requirements but has no positive qualities other than being clean, then the judges would be happy to give the 0s it deserves, after all the -5s and *s for the other skaters.

Or suppose a skater starts the program with their best element, meets the necessary criteria on that element and earns +5s there, and then falls on the next element and it's all downhill from there. That wouldn't cancel out the excellence of the first element.

i totally agree that each element should be scored separately. That's not an issue with me. I actually would advocate for that. Too often, we see the opposite, especially in dance again... but at times in pairs too... Missed opening twizzles or missed opening twist ... then lower GOE throughout.

 

[4everchan]

I want to have your trust in people.

Well, you stated you don't want to talk about this person, but isn't it a bit odd how I'm casually scrolling through recent event scores and when I google the first judge I find suspicious they turn out to be politically very problematic and accused of basically corruption? Either that's a big coincidence, or the rate of such judges is pretty high or there's something real behind my intuitive alarm going on.

I just don't think we can talk about this person without getting into politics and that's not allowed and at this point, there is enough going on in the skating world... but yeah... it's super problematic.

Trust is a very interesting concept, isn't it ? I find it better for my mental health to trust than not. It doesn't mean I accept everything blindly

 

[gkelly]

there often is

this is probably the exception to confirm the rule but then, you know, I am one of these people who would be OK with the following rule :
you fall, the element is invalid.

Would that include jump combinations with a fall on the last landing? Spin combinations with a fall at the end after enough revolutions had been successfully completed to meet the requirements of the spin (and maybe even earn level 4)? Falls at the end of a step sequence or choreo sequence when all the requirements had been met and maybe the step sequence had already earned several level features? Or a fall in the middle of the sequence in which the skater is able to recover and continue the sequence to the end -- including "falls" from low moves like knee turns and hydroblading that may just constitute a slight loss of balance with additional body parts briefly touching the ice?

Back in 6.0 days, when a single score covered all technical content, there was wording in the rulebook that jumps with falls or landed on two feet would not be counted. (Which, of course, led to some discussion of what constituted landing on two feet vs. just touching down the free foot.)

Which can make some sense in a free skate and was probably written to discourage skaters from attempting jumps they hadn't mastered yet in hopes of getting some reward for the attempted difficulty.

In a short program, though, judges set a base value for the program a whole and then took deductions, with the maximum deduction for an attempted element being 0.4 at least as of the mid 1990s. Would it really be fair for the base value for a SP with a fall on a triple axel to be exactly the same as the same program with a fall on a double axel or on a "waxel" wipeout, and then the deduction also to be the same 0.4?

IJS is a lot more flexible in harshly penalizing elements with multiple errors including underrotation and falls and in balancing out errors and strengths in the same element. (And separating the technical scores for the individual elements.)

I get that... but then, in the end, giving a 10 does mean a perfect score right? because nobody can get a higher score. So if the element doesn't require perfection, the score awards it.

10.0 is for program components, not elements.

Yes, a judge who gives 10.0 can't give a higher score on that component to another skater. They can give the same score if it is equally outstanding, perhaps in different ways.

Or more 10.0s if the first performance was only that wonderful in one component and another performance on more than one.

In ice dance, we can see +5 quite often. Perhaps in some lifts in pairs but hot often to we see +5 for jumps or spins or steps ? I am curious about this.

Out of curiosity, I checked the Men's and Women's protocols from the recent Europeans and 4Continents.
Aside from isolated 5s on elements that otherwise got 4s and 3s from most of the panel, I found:

Loena Hendrickx got six 5s on her layback in the short program.
Nina Pinzarone got five 5s on her layback in the free skate.

Deniss Vasiljevs got four for his flying sitspin in the short program.

Yuma Kagiyama got five 5s on his quad salchow in the free skate (and three in the short program).

I'd expect a few more than that at Worlds.

 

[Mathman]

this is probably the exception to confirm the rule...

That is exactly why I am in general supportive of human judges and not so much concerned about variation in judges' application of the rules. The rule book cannot consider all the "exceptions." But human judgment can.

 

[4everchan]

That is exactly why I am in general supportive of human judges and not so much concerned about variation in judges' application of the rules. The rule book cannot consider all the "exceptions." But human judgment can.

Yes...i don't disagree... at the same time, it opens the door to weird stuff

 

[4everchan]

Would that include jump combinations with a fall on the last landing? Spin combinations with a fall at the end after enough revolutions had been successfully completed to meet the requirements of the spin (and maybe even earn level 4)? Falls at the end of a step sequence or choreo sequence when all the requirements had been met and maybe the step sequence had already earned several level features? Or a fall in the middle of the sequence in which the skater is able to recover and continue the sequence to the end -- including "falls" from low moves like knee turns and hydroblading that may just constitute a slight loss of balance with additional body parts briefly touching the ice?

In some sport any fall = 0. Maybe that's where I come from.. A dive that is not "landed" head first is 0. In snowboard, if you don't land your trick = your run is over.
Look there are perks to being more strict about falls= invalid element = 0.
Skaters would aim for clean programs. I think it may be more enjoyable in the long run.

Back in 6.0 days, when a single score covered all technical content, there was wording in the rulebook that jumps with falls or landed on two feet would not be counted. (Which, of course, led to some discussion of what constituted landing on two feet vs. just touching down the free foot.)

Which can make some sense in a free skate and was probably written to discourage skaters from attempting jumps they hadn't mastered yet in hopes of getting some reward for the attempted difficulty.

yup

In a short program, though, judges set a base value for the program a whole and then took deductions, with the maximum deduction for an attempted element being 0.4 at least as of the mid 1990s. Would it really be fair for the base value for a SP with a fall on a triple axel to be exactly the same as the same program with a fall on a double axel or on a "waxel" wipeout, and then the deduction also to be the same 0.4?

give it a *

IJS is a lot more flexible in harshly penalizing elements with multiple errors including underrotation and falls and in balancing out errors and strengths in the same element. (And separating the technical scores for the individual elements.)

more flexibility is a double-edged sword

10.0 is for program components, not elements.

I know that.

Yes, a judge who gives 10.0 can't give a higher score on that component to another skater. They can give the same score if it is equally outstanding, perhaps in different ways.

Or more 10.0s if the first performance was only that wonderful in one component and another performance on more than one.



Out of curiosity, I checked the Men's and Women's protocols from the recent Europeans and 4Continents.
Aside from isolated 5s on elements that otherwise got 4s and 3s from most of the panel, I found:

Loena Hendrickx got six 5s on her layback in the short program.
Nina Pinzarone got five 5s on her layback in the free skate.

Deniss Vasiljevs got four for his flying sitspin in the short program.

Yuma Kagiyama got five 5s on his quad salchow in the free skate (and three in the short program).

I'd expect a few more than that at Worlds.

I didn't watch Euros but layback spins is perhaps an element more conducive to +5. Yuma's quad salchow is stellar.

I don't have a problem with + 5 on some elements. I have more of a problem in ice dance or pairs lift... where we may see "too many + 5"

I tend to have more of an issue with 10.0 components as soon as there is a bobble, a small mistake, some moment of hesitation or lack of flow or synch, why should there be any + 10.0 ? I don't agree with that. I am still not over judges giving Chock and Bates the PCS they got for a program with a fall at worlds... If I had been a judge there, I would have been an outlier So this raises the question : do judges feel they can give a 8.75-9 for world champions in ice dance when they have a major mistake like a fall and some small imperfections ... because, if normally the program could gather 9.5.... with a fall, it could easily be a 9.0... but who did that? Who felt like they had the freedom to do so? I don't recall the protocols by heart but if I am not mistaken, there were a lot more mid 9s than low 9s let alone high 8s. At least if judges still want to reward the composition and the skating skills, they should have the freedom to give a low mark for a fall in an a free dance.