- Joined
- Feb 26, 2014
I can't say for sure tbh, I need to check the rules and the previous Olympics.
All I know is that judges do change from the short into the free. Not the whole panel but some yes, and it's done in random.
Analyzing Sotnikova and Kim's footwork in the FS
- Thread starter Blades of Passion
- Start date
- Status
I can't say for sure tbh, I need to check the rules and the previous Olympics.
All I know is that judges do change from the short into the free. Not the whole panel but some yes, and it's done in random.
- Joined
- Mar 1, 2014
LOL. Nope.
- Joined
- Jan 25, 2013
Apparently no.
As I bump this inane thread, yet again.
Just rejoice in the fact you don't have to live with some of these over-analyzers.
What was the reason to cut back 30% of GOE? To favor the flaws and decrease the points earn by technical superior jumpers? I didn’t know about it.
- Joined
- Dec 29, 2013
Are you even being serious?
- Joined
- Jun 21, 2003
I believe the reason for the change was this. The ISU wanted to encourage judges to give a wider range of GOEs. Instead of reserving a +3 for the greatest example of an element ever, the idea was that the skater should be able instead to earn this mark by satisfying a defined list of "bullets." The language was changed from "suopior height and distance" to "good height and distance," etc.
Then, to keep the GOEs under control (since it was now easier to get a +3), they scaled the value back to 70% in compensation.
That's the theory, anyway.
- Joined
- Sep 17, 2011
It makes BV more important. Vancouver-era singles skating had very few high-risk elements because it was easier to do a 3Lz-3T with +GOE instead of doing a quad, and it was easier to do a 2A instead of a problem triple, with barely any cost in TES. The ISU wanted skaters to go for more difficult elements: for men, the quad, and for women, harder 3-3s and the 3A. They mostly succeeded considering the top men in Sochi all attempted quads and the top ladies all attempted 3-3s and all but one of the top 7 ladies attempted at least 7 triples in the LP.
I don't know the rules well enough -- so what you're saying is that directly connecting two scored elements (in this case a jump into a spin, but it could be anything else like a step sequence into a spin, maybe a spin into a jump which would be unlikely but fun to watch) doesn't affect the GOE for either element?
I looked a bit into perspective and projection math to derive distance estimates for jumps from fan cams of the whole rink. With some basic assumptions, the distances covered by the jumps are fairly straightforward to calculate (if a bit tedious) from fan cams using two-point perspective math (i.e. 2 vanishing points, one for the rink's length and the other for the rink's width). The fan cam videos that I used were:
Adelina: http://www.youtube.com/watch?v=Lf0wI25cD3U
Yuna: http://www.youtube.com/watch?v=2_7eZ4SZ82Y
My assumptions were that the rink was an Olympic-sized rink of 60 meters by 30 meters, surrounded by barriers of an unknown but same height (with obvious cutouts), and that all angles are right angles (i.e. rink is rectangular with rounded edges, barriers are vertical). From each image of the video it's straightforward to measure the inside edges of each barrier as well as the 2 edges of the rink that are visible, to establish the locations of the two vanishing points. The barrier establishes the scale (using the Olympic-sized assumption) and the corner formed by the 2 visible edges of the rink establish the origin used to measure distances. The rest of it is a bunch of math which will probably bore everyone, but basically, once these are measured and calculated, it's possible to work out where the skater was on the rink at any given point. For example, a grid spaced 5 meters between points was generated using this information (without reference to the image other than the measured points), which I then hand-plotted:
https://www.flickr.com/photos/120676593@N05/14294036954/sizes/o/in/photostream/
The red dots are the grid, while the small bright green dots are the points I used to measure the rink dimensions. In theory I could plot this for any image except that it takes a long time (since I measured and plotted them by hand).
Anyway the distance that the skaters moved with their double axels, overlayed (and moved slightly to account for the camera moving during the jump), is here:
Adelina: https://www.flickr.com/photos/120676593@N05/14294072854/sizes/o/in/photostream/
Yuna: https://www.flickr.com/photos/120676593@N05/14107969598/sizes/o/in/photostream/
It's just to give viewers a sense of how far they moved in their jumps, not used in the measurements. You can see that Yuna was closer to the judges but jumping more away from them, while Adelina was farther away but jumping (relatively) more across. For each image (takeoff and landing for each skater), I measured the rink dimensions, and then used that to work out how far each skater traveled in the air. I took the upper corner of the rink as the origin, with the long side as the x-axis and the short side as the y-axis. For both takeoff and landing I used the center of the blade as the skater's location. The results were:
Adelina: (54.9, 17.6) to (54.4, 20.3) => 2.7 meters
Yuna: (20.9, 12.2) to (21.5, 14.1) => 2.0 meters
To make a long story short, from the fan cam videos Adelina covered more distance on that jump than Yuna, even though both had the same height (based on how long each was in the air). (This may also imply that Adelina was moving faster into the jump, though not necessarily proves it since it depends on how each skater "used" the speed to get the height.) So depending on the judges, they may have felt that 2.7 meters deserves a bullet while 2.0 meters doesn't. I'm not saying that they took out a ruler and measured it, of course, but just in terms of how far it appeared to them. I would say that's likely a noticeable difference to the naked eye, and it's more a matter of how likely it is that the cutoff for each judge is within that range.
I would argue against using physical instruments (such as sensors on the figure skates) and more for software instruments (i.e. software analysis of videos). Obviously cameras are also a type of physical instrument, but my point is that I would argue against anything that is specific to the skater. It makes the instrument too easy to tamper with (since it's out of sight for long periods of time), too hard to ensure they're properly calibrated, etc. I think a video system is sufficient to determine quite a few of the measurable quantities of interest such as jump height, distance, etc. It's easy enough to count the frames that a skater is in the air as a proxy for jump height, not to mention use it to determine things like whether or not the skater had fully rotated a jump. Even from a random fan cam I can determine the jump distances to within tens of centimeters (size of 1 pixel was about 3 cm by 6 cm where Adelina was at, and about 6 cm by 24 cm where Yuna was at). Only one calibration would need to be done, and it would affect all skaters equally; in essence, they're all measured by the same yardstick, which is also the quality we want in judging (that a judge holds each skater to the same standard, regardless of whether it's high or low). I don't think it would necessarily take those evaluations away from the judges. Even though those metrics can be measured, there's still other issues like how early or late a jump was done (a jump may still give a height bullet for a smaller jump done later in the program), how the skater's height affects the jump quality (a judge may give a shorter skater the jump height bullet but not for a taller skater even if both jumps are the same height), etc. which I think would still be up for consideration. It's that measuring would take away much of the guesswork in this. Then again, so would the judges being more transparent in terms of showing which bullets they considered a jump or other element to have met, etc. rather than giving simple GOE's.
Keep in mind I was talking about the SP; I haven't looked at the FS yet. I haven't really looked at the FS footwork for Kostner, only the SP.
Neato. Thanks! Gonna have to digest it slowly.
If 3Lz-3T is easier how come not many ladies have high success rate of it?
Does it hold true the rule change of GOE is more advantage toward the flawed jumpers than the one with more technical superior?
Considering +3 for perfect execution vs the fall of -3 as before, but now the gap is closer between 2.1 to -2.1. That is 1.8 points difference which can change the placement of skater.
- Joined
- Sep 17, 2011
3Lz-3T is obviously easier than a quad, which no women perform, and I was referring to the men with this example. Though the number of ladies attempting 3Lz-3T successfully has greatly increased in recent years: Gold, Lipnitskaya, Kim, Radionova, a bunch of the junior ladies are all doing it. As for this talk about skaters with superior technique being punished, GOE is still worth points, and a skater with better technique will obviously beat a skater with poor technique attempting the same tech content. A smaller advantage is still an advantage. Sotnikova "beating" Kim was because her technical flaws were totally ignored, which is not a fault of the rule changes at all. I mean, she was even gifted higher GOEs; do you really think reverting to Vancouver CoP would've helped Kim win?
Thanks for your thoughts about the jump measurements.
Not usually, not unless it adds significant difficulty, which a normal step forward from a successful jump landing into a spin entry does not.
If it were difficult enough to add difficulty, it would add it to the second element -- in this case the spin -- not to the already-completed element.
It's not particularly creative, so it wouldn't add "creative exit" to the preceding element either.
We could look for examples where two elements in close proximity do add significant difficulty or creativity and therefore would qualify for GOE bullet points. But these particular double axels by Kim and Sotnikova do not.
double axel directly into back sitspin
Can't get any more intricate than that. Assuming the spin is low enough and held long enough to get called at all under today's rules, and this were called as two separate elements, 2A and FSSp (or just SSp?), I'd say the difficulty of the connection should add to the GOE for both elements. Of course the GOE would also have to take into the quality of the elements, so the final GOE might not end up positive in this case.
Connection from jump landing to spin entrance difficult/creative enough to add to the GOE(s) IMO
I'd call these landings creative and difficult enough to add to the jump GOE as well as the Transitions component, but not sure I'd also give GOE credit to them as spin entries. Some judges might.
Not usually, not unless it adds significant difficulty, which a normal step forward from a successful jump landing into a spin entry does not.
If it were difficult enough to add difficulty, it would add it to the second element -- in this case the spin -- not to the already-completed element.
It's not particularly creative, so it wouldn't add "creative exit" to the preceding element either.
We could look for examples where two elements in close proximity do add significant difficulty or creativity and therefore would qualify for GOE bullet points. But these particular double axels by Kim and Sotnikova do not.
double axel directly into back sitspin
Can't get any more intricate than that. Assuming the spin is low enough and held long enough to get called at all under today's rules, and this were called as two separate elements, 2A and FSSp (or just SSp?), I'd say the difficulty of the connection should add to the GOE for both elements. Of course the GOE would also have to take into the quality of the elements, so the final GOE might not end up positive in this case.
Connection from jump landing to spin entrance difficult/creative enough to add to the GOE(s) IMO
I'd call these landings creative and difficult enough to add to the jump GOE as well as the Transitions component, but not sure I'd also give GOE credit to them as spin entries. Some judges might.
- Joined
- Feb 17, 2010
Because the skaters do more than one element, Carolina cannot win simply on the strength of one jump combo.
Sorry to bring this up in this topic, but I don't get how Julia was awarded [10.10 1.10 2 2 2 1 3 0 0 1 3 11.20] for her wonky 3Lz-3T. See 11:10 of this video. Another instance of poor scoring by the judges again.
https://www.youtube.com/watch?v=u5fA9o2sBis
https://www.youtube.com/watch?v=u5fA9o2sBis
- Joined
- Feb 26, 2014
- Status
Similar threads
