1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Hi all! No longer will threads be closed after 1000 (ish) messages. We may close if one gets so long to cause an issue and if you would like a thread closed to start a new one after a 1000 posts then just use the "Report Post" function. Enjoy!

Revamping COP for Next Quad.

Discussion in 'Great Skate Debate' started by JJS5056, Jan 29, 2014.

  1. smarts1

    smarts1 Well-Known Member

    I think we need to follow gymnastics' lead and only allow judges to judge from federations that aren't represented by any of the skaters at a given competition. It would eliminate some of the bias and is step in the right direction at least.
  2. RFOS

    RFOS Well-Known Member

    It sounded good, but after creating what the GOE scale would look like with the 40/60/80/100/120/140/160 and seeing that a +3 GOE 4T would be worth about 16.5 points I can already hear the screams of Chanflation ;) (because even when Chan messes up other things, like in the Worlds FS last year, he often does amazing quads). Though I just did some rough calculations and with the lower points for elements with negative GOEs ended up with a TES about 10 points lower than Ten in that program.
  3. giselle23

    giselle23 Well-Known Member

    First, they have to change the judging. Get rid of anonymity and ban cheaters for life. Judges should work for the ISU, not their federations and should be paid to do their job by the ISU. Payments from the federations would lead to suspension or banning. Only then should they turn their attention to the judging system.
  4. gkelly

    gkelly Well-Known Member

    How does that work at Worlds and Junior Worlds, where all federations can be represented in the short program?
  5. kwanfan1818

    kwanfan1818 I <3 Kozuka

    Proportional GOE does not have to be on a flat scale or on that scale, since it doesn't represent the relative difficulty of going from a 0 to +1 vs. +1 to +2, just as the current flat scale for PCS doesn't represent the relatively difficulty of raising a 5 to a 6 vs. raising an 8 to a 9.
  6. RFOS

    RFOS Well-Known Member

    For some reason quoting isn't working for me right now. I'm not quite sure I follow what you're saying, kwanfan1818. Did you just mean that the steps between GOEs didn't necessarily have to be equal? That's true, but IMO it is better if they are because there are always gray areas between GOEs, and if a +3 for example was a BIG step up from a +2 or a -3 a BIG step in points down from a -2 then which GOE judges picked in a borderline case might matter too much (especially if a hypothetical dishonest judge wanted to manipulate the results by always erring on the higher or lower GOE). Regardless of whether it's "harder" to go from a 5 to a 6 or an 8 to a 9 in PCS, the mathematically difference is the same and I think it should be for GOEs also.

    Most of the difference between Ten and Chan's FS scores from Worlds came from an increase in Ten's score do to the wider spread of GOEs for most elements (including jumps) in the positive direction. His technical score increased by more than 6 points while Chan's increased by about 1 point (he benefited from it on his good elements but suffered on the bad elements). Having a wide spread in values from -3 to +3 will tend to reward clean programs done with high quality, because the skater will benefit from the increased spread in the positive direction and not suffer on any elements with negative GOEs. The extreme case would be 0 points for a -3 and double the base value for a +3 which I certainly wouldn't support. But as things stand now, I think some elements with -3 GOE do get too many points, for example quads (where a -3 GOE gets at least 70% of the base value) and pair combination spin (where if it earns a level 4 it gets 80% of the base value even with a -3 GOE).

    1/2 base value for -3, 2/3 for -2, 5/6 for -1 would be another possibility worth trying out.
  7. kwanfan1818

    kwanfan1818 I <3 Kozuka

    Yes. I understand how using a -3 to 3 scale would be able to skew the GOE's if there was a greater gap between 2 and 3 than of 1 and 2, but that reflects the difficulty. I don't think PCS should be linear either; I think that's where two wrongs don't make a right.

    I think that the scale of -3 to 3 is too narrow. Perhaps -5 to 5 would be able to take more of the +'s and ='s into consideration, but I still don't think GOE should be linear.

    I also don't think a system should be created to be structured to a predetermined outcome, like to make Ten win over Chan at 2013 Worlds. The system should decide on what it values, have the judges be on the same page when it comes to assessments, know that the skaters will try to give them what they ask for and adjust to any inconsistencies the judges and the tech panel show, and whoever wins, wins.
  8. Mokesh

    Mokesh Member

    Typo- I meant 6, but had the 7 passes (for women) stuck in my head :)
  9. gkelly

    gkelly Well-Known Member

    I've suggested before:

    Put double walley and double inside axel in the scale of values (the singles can remain unlisted)

    Give a 2.0 bonus for six different takeoffs as doubles or better

    Give a 4.0 bonus for six different takeoffs as triples or better (2A fills the axel slot for women) OR
    4.0 bonus for eight different takeoffs as doubles or better
    AxelAnnie and (deleted member) like this.
  10. RFOS

    RFOS Well-Known Member

    So how would you propose changing the GOE and PCS points to a non-linear scale? I'm not clear what the benefit of that would be so I'm curious how you would change it and why.

    I guess I disagree that it is in theory more difficult to go from a +1 to a +2 than from a +2 to a +3 for example. The judges' bullet point guidelines suggest 2 bullet points for a +1, 4 for a +2, and 6 for a +3 GOE, so that suggests that the progression is approximately linear.

    Agreed, I just was curious to try some controversial examples and see how the results turned out with proposed changes.

    How would your scale
    Last edited: Mar 12, 2014
  11. kwanfan1818

    kwanfan1818 I <3 Kozuka

    I don't know how much clearer I can be than to say that it it more than a linear increase of difficulty to go from +1 to +2 than to go from +2 to +3 in terms of GOE or to go from performing 50%-60% of hitting the PCS bullet points than to go

    First, the bullet points are recommendations and are not mandatory. A Plushenko or a Joubert have gotten +2's and +3's perfectly legally in the system for quads that have great height and just decent landings, because it's not only at the judges' discretion to ignore the bullet points completely, but also to have different standards for different quads.

    I don't know what to say if you think that even using the same bullet points, it it is an equal increment of difficulty to add a complex landing to a jump as it is to add a complex entrance to a jump where there is a complex landing except to wholeheartedly disagree. That's why I think the tech panel should be calling levels on jumps the way they do with spins, steps, and lifts, and for the GOE to reflect quality only.
  12. RFOS

    RFOS Well-Known Member

    Sorry, I meant more specifically, like what proportion of the base value (since you agreed with proportional GOE) would you have a -3, -2, -1 (and -5 and -4 if you would add -5 and -4, since you suggested that idea) worth (and on the positive side).

    You're correct that they're recommendations only and there needs to be discretion on the part of the judge because they can always be fulfilled to varying degrees. In some cases, certain bullet points may be so impressive that they could be counted more. Also, some of the bullet points kind of look like they were just thrown in there to get a nice even number. For example, "unexpected / creative / difficult entry" and "clear recognizable steps/free skating movements immediately preceding element" have a lot of overlap and "good flow from entry to exit including jump combinations / sequences" and "effortless throughout" are very similar in meaning. "Good height and distance" is only one bullet point, but if a jump has REALLY good height *and* REALLY good distance it wouldn't be fair IMO to only give it half a GOE increase.

    If it were up to me I would combine those two pairs of bullet points mentioned above into one, separate good height from good distance, and add one for speed into and out. It's not up to me, but since there is discretion involved and they are only guidelines I will use my discretion. Any quad pretty much HAS to have really good height and distance in order to be landed, so any quad with a good landing is probably going to be starting from at least a +1 and likely +2, IMO. I can't think of the last time I saw a 0 GOE worthy quad that didn't have some kind of slight error to bring it down from what would've been a plus. It's not surprising that quads tend to get good GOEs when landed, even when there aren't difficult entries (and let's face it, difficult entries into quads are very uncommon, and if executed well and directly into the jump would be a clear example of a +3-- something that stands out as truly exceptional even at the highest level of skating). A triple isn't likely to have as much height and distance as a quad, so it's not surprising that it might not get as high of a GOE as a quad with an equally good landing without some kind of other enhancement(s) like a difficult entry.

    http://www.usfsa.org/content/2013-14 S&P Establishing GOE.pdf
    Last edited: Mar 12, 2014
  13. kwanfan1818

    kwanfan1818 I <3 Kozuka

    A quad or any other jump does not have to have good height and distance to be landed. Distance is often sacrificed for height, and there's a specific jump technique, widely accepted, where the jump drops out of the air instead of getting horizontal distance. In combinations specifically the second or third jump is often muscled from a near stop, when if there were distance, without a faulty toe pick or off-balance landing there would be ride out.

    I agree that height and distance should be separate bullets, but the only "bullets" should be quality bullets, for judging the quality of entrance through exit and actually applying the bullet that judges a combo or sequence as a whole, rather than two separate jumps, with the quad or harder jump especially trumping the quality of the combination.
  14. RFOS

    RFOS Well-Known Member

    Can you give a video example of a quad that you think doesn't have good height or distance? One of the few examples I can think of from the past where I've felt that way was some of Kevin Reynolds' quads, and he occasionally has had rotation issues with them (but most often could get the rotation due to extremely quick rotations aided by his very slim body type). But almost everyone else I can think of who has done a clean, rotated quad has done so with good height and distance warranting awarding of the bullet.
  15. MrLucky

    MrLucky Active Member

    I think all of the suggestions are pretty good.

    But unfortunately figure skating after Sochi has been left with yet another stain on it's already badly tarnished/ruined sporting reputation. DUH :(

    I love how there has been so much criticism of having a Russian tech caller in Sochi.
    So what does ISU do to follow it up?

    Surprise, surprise, we see that a Japanese tech caller has been assigned to Worlds in Japan. :lol:

    I think ISU has big brass bells and could care less about heartfelt suggestions to make the sport better.
    I also see nobody here trying to get to the root of the problem which should be a no-brainer.

    The biggest problems facing figure skating are not the rules of the COP but the people running the show.

    The "judges are human" excuse has become sickening to me simply because not all humans are corrupt cheaters. Many humans are good. Yes?
    We see who Cinquanta's successor at ISU is likely to be and not a peep from the peanut gallery here. Good luck with that :dog:

    Button told us a decade ago "it's not the system it's the people in charge."

    So even as I am happy that Adelina won the OGM it is harder to enjoy it because of the lack of credibility from ISU.

    Trust me, that is not a way to grow the sport most of us here love so much.

    So go on and discuss "rule changes" while I sit here wondering what it would really take to restore credibility to the sport of figure skating.
  16. gkelly

    gkelly Well-Known Member

    Besides Reynolds, Timothy Goebel was the only other skater who came to mind as having landed clean quads without "good" height and/or distance.
  17. zebobes

    zebobes Active Member

    I had a couple of ideas for how to revamp COP, but I posted it in the "Is IJS working?" thread. I'll post it here too, to get your thoughts.

    Okay, I just had a couple of ideas:

    Jump levels:
    All of the other elements are leveled, but jumps aren't. Currently, when judges are awarding GOEs for jumps, it's really complicated, because sometimes judges will give higher GOEs to messy jumps because of the difficulty. Or, some may feel that jumps like Adam Rippon's Lutz aren't properly rewarded, even though the jump sometimes isn't that clean. And then judges will forget to give negative GOE to the solo jump in the short program. If you separate the two, you can encourage skaters to work on their jump features-- but also you will need to set parameters, or else some skaters will make every jump a Rippon jump. The only thing is that it would make things more difficult for the technical panel.

    Currently, judges scores are anonymous, and yet some judges scores are thrown out every time. I find that very wasteful, and it cultivates an attitude where judges feel like their input doesn't matter that much. What if the 5 components were split among the 9 judges (it would be better if their were ten), and so each judge would ONLY judge that component. That would help each judge to focus on their responsibility, and it would also be interesting to look and see what each judge did. The two judges would have to work together, but the average of their numbers would be given to each skater. That would make it mean more if you wanted to announce the PCS separately, as you could know that one judge gave you that score. Also, there would be greater variation in PCS, and it would encourage skaters to become more balanced. Each judge would still give GOEs. Each judge would have more power, yet the power would be limited at the same time, because you are only affecting a portion of the score.

    What do you guys think?
  18. danafan

    danafan Canadian ladies ├╝ber

    Gymnastics only does this in apparatus finals (8 athletes only) at world championships and the Olympic Games. Doing this for all competitions would be completely unfeasible, considering how many nations field athletes at these events.
  19. olympic

    olympic Well-Known Member

    If an FSU member wins the lottery, use your money to 'buy' the ISU and make them do your bidding :EVILLE:
  20. smarts1

    smarts1 Well-Known Member

    Don't they do it during the all around also?

    Hmm now that I think about it, it would be more difficult at a Worlds or Junior Worlds to do this...
  21. AxelAnnie

    AxelAnnie Well-Known Member

    Genius! Anybody ever watch STICK IT? It is about biased gymnastics judging. The gymnasts take over the result by deciding who should win an event. That gymnast completes her event. All the others start and then stop so TN eyes get zero. Pretty cute.