The Podium at Calgary
- Thread starter Joesitz (RIP)
- Start date
- Joined
- Jun 21, 2003
Vash, I think if you look at post number 54 by Curious and post number 57 by Theatergirl the mystery will be resolved and Theatergirl's innocence revealed. 
In post 54, Curious attempted to quote you, but forgot to close the quote.
Then in post 57, Theatergirl quoted Curious. When she closd the quote, the HTML hierarchy of commands left the [quote = Curious] at the top and applied the close quote to the
In post 54, Curious attempted to quote you, but forgot to close the quote.
Then in post 57, Theatergirl quoted Curious. When she closd the quote, the HTML hierarchy of commands left the [quote = Curious] at the top and applied the close quote to the
Vash said:. So it made it seem like Theatergirl was quoting you instead of Curious.
I can fix it if you like (but then this post wouldn't make any sense, LOL).
MM
PS. If I put a "close quote" sign now, the part of this post starting at "quote=Vash" and continuing down to "now" would appear as a quote -- attributed to you!
Last edited:
- Joined
- Jul 11, 2003
The Qualifying Rounds have two separate slates of judges.
One group got stricter judging, imo, than the other.
Joe
One group got stricter judging, imo, than the other.
Joe
- Joined
- Feb 1, 2006
Mathman said:Vash, I think if you look at post number 54 by Curious and post number 57 by Theatergirl the mystery will be resolved and Theatergirl's innocence revealed.
Thank you for defending my honor, Mathman. What would we do with out you to settle these things? Yes, I was aware that I was quoting Curious but I wasn't aware that the statement had been misquoted. Thank you. Please try not to jump on me so whole heartedly. I really didn't mean anything by it. I would never misquote. I know I may come off as a bit sarcastic, but I really don't mean to be rude.
I don't know, I am not a huge Sasha fan, but, although I thought that Elena had a pretty good performance in the LP, I really don't think she had the fire that she had in the SP. However, I do think that Sasha was over-marked. I think that she is given good PC scores because the judges know that she is a skater that gets goods PC scores. I don't really think that it reflected her actual performance. I didn't see the QR so I don't know if her scores there were deserved. I'd have to do a bit more math
biggrin to see if I really do think Elena deserved the medal. It was pretty close for me.
- Joined
- Jul 31, 2003
Mathman,
Thanks for explaining what happened, and therefore proving Theatregirl's innocence. If I had just looked at those posts, perhaps I would not have figured out what actually happend.
Theatregirl, you are entitled to your opinion. I saw the competition live (I don't know if you were there too) and Elena actually skated very well, except for doubling the triple flip (lip). Sasha was very tentative, and really did not have anything except for her usual extension moves (spiral seq and I-spin) that draw wows from the audience. IMO she was overmarked in both the LP and QR (she had 2 falls in it). You are right that the judges are sometimes giving PCS marks that are too high. IMO Kimmie's PCS marks were too high also, but even if she had received around 55 (more in line with what she did) instead of the 60 she received, she would have still won by 5 or 6 points. Her skate created so much excitement in the arena that I can see why she received such high PCS. All ladies in the last group, except Sasha and Joannie, skated very well- not perfect but very solid performances.
Vash
Thanks for explaining what happened, and therefore proving Theatregirl's innocence. If I had just looked at those posts, perhaps I would not have figured out what actually happend.
Theatregirl, you are entitled to your opinion. I saw the competition live (I don't know if you were there too) and Elena actually skated very well, except for doubling the triple flip (lip). Sasha was very tentative, and really did not have anything except for her usual extension moves (spiral seq and I-spin) that draw wows from the audience. IMO she was overmarked in both the LP and QR (she had 2 falls in it). You are right that the judges are sometimes giving PCS marks that are too high. IMO Kimmie's PCS marks were too high also, but even if she had received around 55 (more in line with what she did) instead of the 60 she received, she would have still won by 5 or 6 points. Her skate created so much excitement in the arena that I can see why she received such high PCS. All ladies in the last group, except Sasha and Joannie, skated very well- not perfect but very solid performances.
Vash
- Joined
- Jul 11, 2003
It was easier to tell when watching the difference in the scoring. The first group to skate were marked lower than the second and there was no particular reason for it except there were two slates of judges. One has to understand that each judge has a different set of standards than the next. It's the old human nature.Mathman said:
Given both skaters skated the same, had Sasha been in the first group would her scores be different? I think so.
If Elena had been in the second group, I think her scores woould be different.
This is a problem with the Quali Rounds and while the scores in the CoP are minimal they can be and usually are of paramount importance in the final score especially if the final scores are close.
If they had the Quali rounds on 2 different days, the could use the same judges.
Joe
- Joined
- Feb 5, 2004
Joesitz said:
I agree with you Joe but can you imagine the hell that would break loose - there was enough of an issue with one group skating at 9am and theother in the evening without having them on two different days - some skaters would more days off in between the QR and SP than others!!
Ant
- Joined
- Jul 11, 2003
Exactly, can you imagine getting up at 6am to do your thing and then go to the arena to be judged. Not the same as getting up at 5am and going to the rink and practice.antmanb said:
In the Dortmund worlds it didn't matter so much because it was the 6.0 system and placing in qualis and SP were important before the LP.
In the Worlds of today, it is points by two sets of judges in the Qualis. Think about that.
Joe
- Joined
- Feb 1, 2006
Vash01 said:
Oh, I definitely agree with you that Sasha's performance was nothing to write home about. From a lesser know skater it probably wouldn't have come close to being a medal winning performance (than again I may still be thinking 6.0 where a mistake on most of the jumps and a fall would be scored rather low). It's not an issue for me of whether or not Elena was better than Sasha (and I will admit that for some unexplained reason I don't like Elena's skating), she was, it's just was she good enough to make up Sasha's lead. I believe you that Sasha was over marked in the QR, that seems to happen a lot so that the medal contenders can stay in contention. I definitely agree that Kimmie deserved her win. If PC scores are not going to be marked on what they actually mean (if they were wouldn't some skaters have significant differences between categories?) then I don't think that grace is necessarily better than excitement and energy. No, I didn't get to see worlds live (god, I would love to) so I'm going to accept your opinion on that front.
- Joined
- Jul 26, 2003
The choices for qualis seem to be:
a. Have separate panels of judges for both groups (as is), with both groups skating on the same day.
b. Have one round mid-morning and another round in the evening (instead of the afternoon), and have one panel of judges score both sets of skaters after an aftenoon nap.
c. Split qualis into two days, both groups starting at the same time. Same set of judges.
Questions:
1. Would the same judges apply the same standards to both groups, if they were on separate days or at different times? I'm not so sure. Judges tend to start slowly and warm up to a pattern of scoring. If they came back for the second panel that same day, then they might start the second panel in a groove. The same judges could score differently, based on having already judged 20+ skaters earlier in the day. If this happened, it would cause a more predictable double advantage to the skaters who perform in the second group.
2. Is it necessarily a disadvantage to have two groups of judges? It depends. In general, it's considered an advantage to skate in the later group, unless the skate is late in the first group. If the morning group of judges is more generous with scores, then this could neutralize the disadvantage of starting early. If the later group of judges is more generous, then it is a double whammy to the early group.
3. Is it an advantage or disadvantage to have an extra rest day, if the groups were split into two mornings or two afternoons/evenings? I think that depends on the skater. Some skaters would get more nervous with a longer wait after qualis before the short program, and others would prefer the extra rest/practice day.
4. If they're going to make skaters perform on the same day with different sets of judges, and they were willing to have an evening group in Calgary, why do they insist on a morning group, instead of an afternoon group and an evening group, particularly when a great number of skaters hate skating in the morning group? At least more people can show up for the evening group, once school and work is out, and these are usually the most affordable tickets.
The mitigations for having one tough set of judges and one easier set is not basing the short program cut-off based on the top 30 scores, but instead by the top 15 scores in each group, and factoring the scores to 25%. I, frankly, think they should jettison the scores after the quali rounds, except for recording personal bests, and at most, use them to weight the SP groups along the lines they use for Ice Dancing. I don't think it's a bad thing to have 18-20 judges working on a major event.
Random selection is not the same as choosing 9 judges up front and having 3 treated as phantoms. Random selection is done before each phase of the competition, and the chances of not being selected for any panel are greatest if the quali round is the only round judged, but slim after the SP/FS panel is chosen. The chances of having the scores count in at least one phase are solid. So it's unlikely a judge is sitting there for nothing, unless s/he's neither selected in the quali round nor selected for the SP/FS panel.
Random selection allows a greater variety of judgements over the course of a competition, because while there is overlap, the chances are not great that the same 9 judges' scores will count for all phases. As Mathman has pointed out, this can be for better or worse, depending on who is selected and whether they've been colluding. While this may not be better than taking all 12 scores, the theory being the more opinions, the more accurate, it still is different than having a single panel of 9 score everything.
It is fairly easy to tell which judges' scores are not counted for each stage, particularly if one is a programmer. Whether it is easy to know which judge gave which set of scores is dependent on whether they tell other people truthfully which set is theirs. Certainly there's a lot of speculation about which judges were which. On the other hand, if a country has three-four judges and one-two contenders, there is still plenty of room for complicated dealing, and the scores we assume are from the judge from that skater's country (or that country's arch-rival) might very well be from a judge who had made a deal to hold up or down that skater. Again, as Mathman has pointed out, it is possible for a "bloc" to be diluted in the random selection by being tossed out in whole or part or to be fortified by random selection, by making up a large portion of the counted judges. (The bigger the bloc, the more likely it is to have its members selected.)
It is very true that the person with the most points wins. Just as the cross country skier who has the fastest time by .1 over 100k wins the race. S/he gets the medal, the trophy, the bonus, the diamond earings, the car, the ISU money, etc. But what the figure skating scores also show is whether the scores are meaningful in an abstract sense, i.e., whether they are statistically significant, or could any X skaters have reached Y place. This is mathematically possible to determine regardless of whether there is random selection or purely trimmed mean. Ordinals flattened the differences between skaters: there could be a .1 point difference or a 10 point difference, and it all looked the same: 1st and 2nd. Absolute scores show scale, and statistically significance across multiple scores tells the story of how likely that outcome is.
a. Have separate panels of judges for both groups (as is), with both groups skating on the same day.
b. Have one round mid-morning and another round in the evening (instead of the afternoon), and have one panel of judges score both sets of skaters after an aftenoon nap.
c. Split qualis into two days, both groups starting at the same time. Same set of judges.
Questions:
1. Would the same judges apply the same standards to both groups, if they were on separate days or at different times? I'm not so sure. Judges tend to start slowly and warm up to a pattern of scoring. If they came back for the second panel that same day, then they might start the second panel in a groove. The same judges could score differently, based on having already judged 20+ skaters earlier in the day. If this happened, it would cause a more predictable double advantage to the skaters who perform in the second group.
2. Is it necessarily a disadvantage to have two groups of judges? It depends. In general, it's considered an advantage to skate in the later group, unless the skate is late in the first group. If the morning group of judges is more generous with scores, then this could neutralize the disadvantage of starting early. If the later group of judges is more generous, then it is a double whammy to the early group.
3. Is it an advantage or disadvantage to have an extra rest day, if the groups were split into two mornings or two afternoons/evenings? I think that depends on the skater. Some skaters would get more nervous with a longer wait after qualis before the short program, and others would prefer the extra rest/practice day.
4. If they're going to make skaters perform on the same day with different sets of judges, and they were willing to have an evening group in Calgary, why do they insist on a morning group, instead of an afternoon group and an evening group, particularly when a great number of skaters hate skating in the morning group? At least more people can show up for the evening group, once school and work is out, and these are usually the most affordable tickets.
The mitigations for having one tough set of judges and one easier set is not basing the short program cut-off based on the top 30 scores, but instead by the top 15 scores in each group, and factoring the scores to 25%. I, frankly, think they should jettison the scores after the quali rounds, except for recording personal bests, and at most, use them to weight the SP groups along the lines they use for Ice Dancing. I don't think it's a bad thing to have 18-20 judges working on a major event.
Random selection is not the same as choosing 9 judges up front and having 3 treated as phantoms. Random selection is done before each phase of the competition, and the chances of not being selected for any panel are greatest if the quali round is the only round judged, but slim after the SP/FS panel is chosen. The chances of having the scores count in at least one phase are solid. So it's unlikely a judge is sitting there for nothing, unless s/he's neither selected in the quali round nor selected for the SP/FS panel.
Random selection allows a greater variety of judgements over the course of a competition, because while there is overlap, the chances are not great that the same 9 judges' scores will count for all phases. As Mathman has pointed out, this can be for better or worse, depending on who is selected and whether they've been colluding. While this may not be better than taking all 12 scores, the theory being the more opinions, the more accurate, it still is different than having a single panel of 9 score everything.
It is fairly easy to tell which judges' scores are not counted for each stage, particularly if one is a programmer. Whether it is easy to know which judge gave which set of scores is dependent on whether they tell other people truthfully which set is theirs. Certainly there's a lot of speculation about which judges were which. On the other hand, if a country has three-four judges and one-two contenders, there is still plenty of room for complicated dealing, and the scores we assume are from the judge from that skater's country (or that country's arch-rival) might very well be from a judge who had made a deal to hold up or down that skater. Again, as Mathman has pointed out, it is possible for a "bloc" to be diluted in the random selection by being tossed out in whole or part or to be fortified by random selection, by making up a large portion of the counted judges. (The bigger the bloc, the more likely it is to have its members selected.)
It is very true that the person with the most points wins. Just as the cross country skier who has the fastest time by .1 over 100k wins the race. S/he gets the medal, the trophy, the bonus, the diamond earings, the car, the ISU money, etc. But what the figure skating scores also show is whether the scores are meaningful in an abstract sense, i.e., whether they are statistically significant, or could any X skaters have reached Y place. This is mathematically possible to determine regardless of whether there is random selection or purely trimmed mean. Ordinals flattened the differences between skaters: there could be a .1 point difference or a 10 point difference, and it all looked the same: 1st and 2nd. Absolute scores show scale, and statistically significance across multiple scores tells the story of how likely that outcome is.
- Joined
- Jun 21, 2003
Hockeyfan, you have always been my NJS hero, but you hit a personal best with that post.
This is the first thing I have ever read that gives me something new to think about with respect to the random selection of judges.
Please send that to Speedy so he will have something cogent to say about the benefits of the random selection, rather than something silly about preserving the anomymity of judges (which it doesn't, even if that were a desirable goal).
Mathman
This is the first thing I have ever read that gives me something new to think about with respect to the random selection of judges.
Hockeyfan said:
Please send that to Speedy so he will have something cogent to say about the benefits of the random selection, rather than something silly about preserving the anomymity of judges (which it doesn't, even if that were a desirable goal).
Mathman
- Joined
- Jan 15, 2006
I wasn't happy with this year's worlds at all. Sasha could have done A LOT better. And I don't even know if she'll be back next year. 
Similar threads
