South Korean federation's complaint to the ISU about judging

[Mathman]

If seven judges gave Yuna higher GOE, she would have won GOE, isn't it?

That is the whole point. We don't know.

Here are the total GOEs of the nine judges, in two different orders.

KIM 27 26 22 21 20 20 19 17 16
SOT 26 23 22 19 16 15 31 29 26

If this matching of judges' scores is the correct one, then five judges favored Kim with one tie. Three judges greatly over-scored Sotnikova while simultaneously lowballing Kim. With this pairing the three Sotnikova judges, by giving exaggerated scores, dominated the majority, who favored Kim.

Or…

SOT 31 29 26 26 25 22 19 16 15
Kim 27 26 22 21 20 20 19 17 16

Now six judges favored Sotnikova, with one tie, and only two thought Kim's elements were better.

Which scenario is the factual one? We will never know.

 

[cooper]

That is the whole point. We don't know.

Here are the total GOEs of the nine judges, in two different orders.

KIM 27 26 22 21 20 20 19 17 16
SOT 26 23 22 19 16 15 31 29 26

If this matching of judges' scores is the correct one, then five judges favored Kim with one tie. Three judges greatly over-scored Sotnikova while simultaneously lowballing Kim. With this pairing the three Sotnikova judges, by giving exaggerated scores, dominated the majority, who favored Kim.

Or…

SOT 31 29 26 26 25 22 19 16 15
Kim 27 26 22 21 20 20 19 17 16

Now six judges favored Sotnikova, with one tie, and only two thought Kim's elements were better.

Which scenario is the factual one? We will never know.

that's why they should reveal the scores!! based on ISU's decision that they found no bias from the judges.. so why not reveal those scores?! after all russia and the US advocates to get rid of the anonymous judging.. why not make an example out of this if they have guts and to promote transparency and accountability for the next season/s!

 

[mich2]

That is the whole point. We don't know.

Here are the total GOEs of the nine judges, in two different orders.

KIM 27 26 22 21 20 20 19 17 16
SOT 26 23 22 19 16 15 31 29 26

If this matching of judges' scores is the correct one, then five judges favored Kim with one tie. Three judges greatly over-scored Sotnikova while simultaneously lowballing Kim. With this pairing the three Sotnikova judges, by giving exaggerated scores, dominated the majority, who favored Kim.

Or…

SOT 31 29 26 26 25 22 19 16 15
Kim 27 26 22 21 20 20 19 17 16

Now six judges favored Sotnikova, with one tie, and only two thought Kim's elements were better.

Which scenario is the factual one? We will never know.

Lol. You are desperately trying to use 6:0 system to make something up. It is cheat. If the majority of judges indeed favoured Kim, she would have won by simple math. She did not.
The only real point is that nobody will remember the lawyer's name, sore fans' posts or even the bad KSU. The legacy will go to Kim personaly. Hardly a favour.

 

[Mathman]

Lol. You are desperately trying to use 6:0 system to make something up. It is cheat. If the majority of judges indeed favoured Kim, she would have won by simple math.

Utterly false. You have it exactly backward. Under 6.0 it is true that if the majority of judges favor one skater then that skater will win. Under add-up-the points systems it is possible that a majority favors one skater but the other skater wins. Scenario #1 above shows how this is possible for GOEs. Post 85 shows how it is possible for PCSs. It is embarrassing to have to give a simplified example showing how this can occur:

6.0: Judge 1 gives first place ordinal to skater A. Judge 2 gives first place ordinal to skater A, judge 3 gives first place ordinal to skater B.

Skater A wins, two first place ordinals to one.

CoP. Judge 1 gives 9 points to skater A and 8 points to skater B

Judge 2 gives 9 points to skater A and 8 points to skater B

Judge C gives 6 points to skater A and 10 points to skater B.

Do the math. Skater A is favored by the majority of judges, but who wins?

 

[Mathman]

One further point:

It is not the size of the discrepancy that is at issue. It is the method of determining the winner. Here is another example:

Judge 1 gives skater A 9.00 and skater B 8.75

Judge 2 gives skater A 9.00 and skater B 8.75

Judge 3 gives skater A 8.50 and skater B 9.25.

Skater A was favored by the majority of judges, but skater B wins. None of the marks is outside the corridor that triggers an investigation.

But we do not have to make up numbers. In post 86 and in post 137 I used the exact numbers from the protocols. The point is that because of anonymous judging and randomization of judges' scores, we do not know our right foot from our left, our hat from our glove. (We are as helpless as a kitten up a tree.)

 

[mich2]

Utterly false. You have it exactly backward. Under 6.0 it is true that if the majority of judges favor one skater then that skater will win. Under add-up-the points systems it is possible that a majority favors one skater but the other skater wins. Scenario #1 above shows how this is possible for GOEs. Post 85 shows how it is possible for PCSs. It is embarrassing to have to give a simplified example showing how this can occur:

6.0: Judge 1 gives first place ordinal to skater A. Judge 2 gives first place ordinal to skater A, judge 3 gives first place ordinal to skater B.

Skater A wins, two first place ordinals to one.

CoP. Judge 1 gives 9 points to skater A and 8 points to skater B

Judge 2 gives 9 points to skater A and 8 points to skater B

Judge C gives 6 points to skater A and 10 points to skater B.

Do the math. Skater A is favored by the majority of judges, but who wins?

Judges know that. That is why if seven of them except Shekhovtsova and Balkov or even five of them wanted Yuna to win, the would have made sure to follow Judge C scenario. They did not.

 

[qwertyskates]

That is the whole point. We don't know.

Here are the total GOEs of the nine judges, in two different orders.

KIM 27 26 22 21 20 20 19 17 16
SOT 26 23 22 19 16 15 31 29 26

If this matching of judges' scores is the correct one, then five judges favored Kim with one tie. Three judges greatly over-scored Sotnikova while simultaneously lowballing Kim. With this pairing the three Sotnikova judges, by giving exaggerated scores, dominated the majority, who favored Kim.

Or…

SOT 31 29 26 26 25 22 19 16 15
Kim 27 26 22 21 20 20 19 17 16

Now six judges favored Sotnikova, with one tie, and only two thought Kim's elements were better.

Which scenario is the factual one? We will never know.

Oh dear Mathman, you are turning into a Strawman...

I would question your assertion that we will "never" know which scenario is the factual one, as a difference between 19 vs 31 would be such a ridiculous spread that it would raise a lot of eyebrows even within the inner circle of the ISU who are privy to the pattern of scoring.

I would think that it is more likely that more judges awarded Adelina higher GOEs than a minority giving unbelievable scoring discrepancies overturning the decision of the majority, especially when there are 9 judges and the highest and lowest scores are thrown out.

Either there was cheating and the 3 judges cheated so blatantly, ridiculously and obviously they didn't mind standing out and getting caught, or there was a more even distribution of scores, with a majority giving Adelina a slightly higher score.

One scenario is much more likely and probable than the other.

 

[Ven]

In case you weren't watching, one skater who had never been a world medalist or won any major senior ISU competition toppled over herself and giggled at her stumble, and managed to handily beat one of the greatest of all time, who delivered a flawless routine. The outcome was so unlikely and unacceptable, that millions of people signed petitions, countless articles were written in various languages calling the competition a scandal, and there has been a large outcry against the ISU and to get rid of Cinquanta. It was pointed out that one of the judges was convicted of cheating in the past, but reinstated because that's only considered a minor violation in figure skating. Another of the judges hugged Sotnikova in plain view of cameras upon her win, and so did one of the technical callers too. Many fans and media mocked and wondered aloud, what kind of conflict of interest is that?

But the ISU refused to even acknowledge a controversy, so when you say...

Either there was cheating and the 3 judges cheated so blatantly, ridiculously and obviously they didn't mind standing out and getting caught

...I can't help but laugh when you preface that statement as if it's something from the theater of the absurd.

 

[Meoima]

I have to agree with Yesway, this topic is so tiring! No one will acknowlegede the other side's opinion anyways. So why waste time?

 

[Ven]

I have to agree with Yesway, this topic is so tiring! No one will acknowlegede the other side's opinion anyways. So why waste time?

Nobody is forcing you to open this thread and read it. One would think if you found it tiring and boring, you would choose to read other threads, and not take your time to post "this is so tiring"...

 

[Meoima]

Nobody is forcing you to open this thread and read it. One would think if you found it tiring and boring, you would choose to read other threads, and not take your time to post "this is so tiring"...

I am here for popcorn. Look at my status and you know. personally I am against anonymous judging and I've signed petition against it. But other than that I think I couldn't do more.

Please continue if you feel like monologuing is okay. Both sides!

 

[qwertyskates]

In case you weren't watching, one skater who had never been a world medalist or won any major senior ISU competition toppled over herself and giggled at her stumble, and managed to handily beat one of the greatest of all time, who delivered a flawless routine. The outcome was so unlikely and unacceptable, that millions of people signed petitions, countless articles were written in various languages calling the competition a scandal, and there has been a large outcry against the ISU and to get rid of Cinquanta. It was pointed out that one of the judges was convicted of cheating in the past, but reinstated because that's only considered a minor violation in figure skating. Another of the judges hugged Sotnikova in plain view of cameras upon her win, and so did one of the technical callers too. Many fans and media mocked and wondered aloud, what kind of conflict of interest is that?

But the ISU refused to even acknowledge a controversy, so when you say...



...I can't help but laugh when you preface that statement as if it's something from the theater of the absurd.

I can't help but laugh at your grandiosity in assuming that the whole world thinks like you. Sorry to say, your assertion is in the minority. It's only Koreans and a handful of loud, outspoken shills, the majority of the mainstream media supported Adelina's win, and were detailed their reporting. Whether you agreed with them is another matter. However the MAJORITY were fine with Adelina getting Gold, including me. Adelina skated better, period.

 

[sk8in]

Uh well, Ven just said articles were written, which is true---and you said only a handful of outspoken shills opposed Adelina's win, which is certainly a mischaracterisation.

 

[drivingmissdaisy]

Skater A was favored by the majority of judges, but skater B wins. None of the marks is outside the corridor that triggers an investigation.

The point is that the GOE and PCS were close between the two. People are free to agree or disagree with that. The difference was the BV, which was in Adelina's favor. Again, people can agree or disagree with how the panel came up with the BV but that was why she won. In other words, even if 6 judges gave Yuna the edge in PCS and 7 the edge in GOE, it wasn't enough to overcome the BV advantage. It doesn't matter how many judges favored one or the other.

 

[Mathman]

Oh dear Mathman, you are turning into a Strawman...

I would question your assertion that we will "never" know which scenario is the factual one, as a difference between 19 vs 31 would be such a ridiculous spread that it would raise a lot of eyebrows even within the inner circle of the ISU who are privy to the pattern of scoring.

Don't blame the messenger. I did not make these numbers up. I took them from the protocols.

One judge really did give Adelina a total of 31 in GOE points (before factoring, etc.) One judge gave the same skater, for the same performance, a total of 15.

For Kim, one judge awarded 27 points and another 16.

Here are two different explanations for these wide discrepancies.

Explanation number one. One judge gave Adelina 31 and Yuna 16. This judge thought that Adelina was great and Yuna not nearly so good. Another judge was of the opposite persuasion. He gave Adelina 15 and Kim 27.

Explanation number two. Both judges thought that the contest was close. The first judge was generous across the board, giving Adelina 31 and Yuna 27. The other judge was uniformly stingier, giving Adelina 15 and Yuna 16.

Which explanation (if either) is correct? We cannot tell from the protocols. As for the hope that the ISU would undertake an internal investigation and pass out sanctions, no. The procedure for deciding when "discrepancies are so large as to raise eyebrows" is quite carefully defined in ISU Communication 1631. None of these scores is sufficiently outside the "corridor" as to constitute an "anomaly" under the judges' review procedures.

 

[Mathman]

Judges know that. That is why if seven of them except Shekhovtsova and Balkov or even five of them wanted Yuna to win, the would have made sure to follow Judge C scenario. They did not.

I will start a new thread about this oddity in the ISU judging system, independent of nationalistic patriotism and unrepentant fandom. (Prediction: No one will want to read it or to contribute. )

 

[Meoima]

Which explanation (if either) is correct? We cannot tell from the protocols. As for the hope that the ISU would undertake an internal investigation and pass out sanctions, no. The procedure for deciding when "discrepancies are so large as to raise eyebrows" is quite carefully defined in ISU Communication 1631. None of these scores is sufficiently outside the "corridor" as to constitute an "anomaly" under the judges' review procedures.

Can we also do this comparision to Mao and Caro's LP? I think we will have better overview.

 

[Mathman]

Can we also do this comparision to Mao and Caro's LP? I think we will have better overview.

For Kostner's versus Asada's program components, now matter how the judges' scores are matched up, Carolina comes out ahead. In this case we can be certain that the majority of the panel favored Carolina.

There are lots of ways of matching the judges -- 362,880 if no two judges gave the same marks. In the case of Mao versus Caro, all matchings produce the same result in terms of majority of ordinals. In the case of Kim versus Sotnikova, some matchings go one way, some the other.

 

[qwertyskates]

Don't blame the messenger. I did not make these numbers up. I took them from the protocols.

One judge really did give Adelina a total of 31 in GOE points (before factoring, etc.) One judge gave the same skater, for the same performance, a total of 15.

For Kim, one judge awarded 27 points and another 16.

Here are two different explanations for these wide discrepancies.

Explanation number one. One judge gave Adelina 31 and Yuna 16. This judge thought that Adelina was great and Yuna not nearly so good. Another judge was of the opposite persuasion. He gave Adelina 15 and Kim 27.

Explanation number two. Both judges thought that the contest was close. The first judge was generous across the board, giving Adelina 31 and Yuna 27. The other judge was uniformly stingier, giving Adelina 15 and Yuna 16.

Which explanation (if either) is correct? We cannot tell from the protocols. As for the hope that the ISU would undertake an internal investigation and pass out sanctions, no. The procedure for deciding when "discrepancies are so large as to raise eyebrows" is quite carefully defined in ISU Communication 1631. None of these scores is sufficiently outside the "corridor" as to constitute an "anomaly" under the judges' review procedures.

I know you took these numbers from the protocol, and it is common to see discrepancies as judges can be quite different in their genorosity, some could be giving marks out like candy, 31/27, some could be stingy, 15/16.

However, you are *assuming* the most extreme scenario in your hypothesis, a difference of 12 points in (1), and then claim that the probability of that happening is *THE SAME* as a more plausible scenario, that judges awarding them a difference of 3-5 points in (2), in explaining Adelina's win.

That to me is a rather forced strawman to argue that the judges could have cheated technically.


By definition, cheating is a stealth act, your worst assumption would have put those 3 judges waaaay out there in terms of judging anomaly in comparison to the others, an anomaly that would immediately place these (1) judges in an unwanted spotlight, i.e. equivalent to the cheats shining a spotlight on themselves, a risk that would be too great to take for any rational cheat attempting to get away with an act of cheating.

In (1), the majority judged that Yuna skated better, but it comes at the price of extremely questionable allocation of points by 3 cheating judges, to the tune of awarding a difference of 10-12 points advantage to Adelina, and hoping that they would get away with it, in a sort of twisted pretzel logic.
.
In (2), the majority judged that Adelina skated better, with 2 dissenting judges giving an advantage of 1 point to Yuna, which indicates they thought Adelina skated almost as well as Yuna.

The two are not at all equivalent in probability. Occam's Razor favors (2).

 

[Meoima]

For Kostner's versus Asada's program components, now matter how the judges' scores are matched up, Carolina comes out ahead. In this case we can be certain that the majority of the panel favored Carolina.

There are lots of ways of matching the judges -- 362,880 if no two judges gave the same marks. In the case of Mao versus Caro, all matchings produce the same result in terms of majority of ordinals. In the case of Kim versus Sotnikova, some matchings go one way, some the other.

No, I mean Mao VS Yulia and Caro VS Sonitkova. Because some people think Mao and Yulia could have switched their places, the same with Caro and Sonitkova. If you know what I mean. I am bad with numer so I think you could do better.