A possible solution to underrotations?

enlight78 · Aug 9, 2008

In tennis there is a computer program called hawkeye that uses an elaborate system of cameras and physics calculations to determine the exact location of the ball. It is a back up to the line callers.

Even though it might be a bit expensive, figure skating could use a similar program to determine the exact rotation of a jump. A technical specialist just tell the computer which jump and it will tell if it was 2.789 revolutions or 3.0165 revolutions.
I don't know if such a program exactly be better for figure skating. What if it turns out the most women triples are under 3.0? Or what if it turns out that most men get the full revolutions? Will they set the bar of underotations higher for men and lower for women? Will they try to throw in extra points for complete revolutions, will they try to take off points for slight over rotations? How will effect the fans? Will take something away?

Joesitz (RIP) · Aug 9, 2008

I was at the US Open last September, and watched the screen showing fans the slow motion travel of the ball towards the boundary lines of the Court. No arguments among the fans; no argument from the Referee.

Do we need this sort of device in Figure Skating? I think we do. We would not have to read about underrotations from the TV watches. I think it will improve the sport in so many ways.

Mathman · Aug 9, 2008

I think it could be done. Hey, if we can put a man on the moon...

But it would be very hard and, yes, I think very expensive to develop. In tennis there is a line on the court. All the computer has to do is decide where the ball hit the ground. For underrotations, you would need enough views to determine the exact angle of the (rapidly turning) blade at the precise moment it leaves the ice, then the same thing on the landing. A cool problem in computer imaging, actually.

I think we would discover that pretty close to 100% of ladies' jumps are underrotated.

beep_beep · Aug 9, 2008

I don't think it's that easy... yeah we can land a man on the moon, but our skills as far calculating things for complicated assymetrical extended bodies (blades...) are quite limited.
Hawk eye uses the court lines and a spherical body flying in basic projectile motion (with a couple of little complications because of high speed), and it took years to develop. The system takes snapshots (a whole bunch of them) of the ball on the air and the computer calculates where it would probably land, with, in my opinion, very good precision (Mr. Federer disagrees greatly on that matter

). It takes court measures into account and the ball flies a greal deal of distance.
On an ice jump...the rink is huge and you would need to have a bunch of cameras focusing on the landing blade of a lone skater somewhere in a 60x30m ice rink, and the whole thing has to be automatic. Then the computer would have to make fast calculations for a rotating+translating body and... well you get the picture. And if it's pairs... :eek:

hwell:

I think adding a second tech specialist in a different position and a second camera angle are better choices, but trying to come up with some type of electronic automatic checking system would be a very interesting challenge indeed.

ChrisH · Aug 9, 2008

I think you'd want to use motion capture technology to look at the skaters whole body. If you only look at the blades, skaters could rotate their ankles+legs on takeoffs and landings and bend their knees+waist on landings to get a lot of extra low quality rotation. From a standing position, you can let yourself fall by lifting up you legs and simultaneously twisting them. I can get a quarter rotation that way. If I had some rotational momentum to start with, I expect that I could get at least another quarter rotation more.

Joesitz (RIP) · Aug 9, 2008

As it is now, we have to put full faith and trust in the Tech Panel which are all sitting in the same area and facing the skaters for their determination of the underrotations and wrong edge takeoffs. Yet despite our acceptance of the calls by the officials we still have arm chair TV judges who decide there were underrotations and flutzes and lips. Do we want this scenario to continue?

When we bring up the costs of such devices, we are looking for the negative of improvement. I am sure the electronic device will show better skating results than all the bickering we have each season.

We are not in a position to put such a proposal down without checking the actual costs. Why did we not allow 10 volunteers holding cards with total scores from continuing instead of paying all that money for computers? It is not the concern of fans to worry about cost effective devices except to show joy in their implementation. In fact all we have to do as fans is say yes to have such a devise installed. Let the ISU worry about the costs.

And btw, like the Tennis fans, let the Figure Skating fans see the solomo tapes also.

Mathman · Aug 9, 2008

beep_beep said:
...The system takes snapshots (a whole bunch of them) of the ball on the air and the computer calculates where it would probably land, with, in my opinion, very good precision (Mr. Federer disagrees greatly on that matter )...

That's an interesting point. We have become accustomed to putting so much faith in technology that we think computers can never be wrong.

Sometimes we forget that, in the case of tennis, what we are seeing is not an actual slowmotion picture of what happened, but a computer generated cartoon.

PolymerBob · Aug 9, 2008

I totally agree. I think there are certain factors that trick judges into thinking that jumps are underrotated when they are not, such as jumping low. Let's take out the argument and guess-work. :agree:

I think it could get interesting if such technology can be used on existing footage. Suppose one of Shizuka's jumps were underrotated in Torino. :eek:

enlight78 · Aug 10, 2008

I have another question with the technology being so expensive should nationals have to carry the system? I don't believe so, but then again the skater qualify for worlds through nationals not GP's. Thanks to Yukari we all know underotations can really kill your chances. I thought it was odd when the U.S. Nationals didn't change their scoring system at the same time as the international events did. When lot skaters sometimes miss GP's and just skate Nationals that really puts them at a disadvantage.

Mathman · Aug 10, 2008

PolymerBob said:
I totally agree. I think there are certain factors that trick judges into thinking that jumps are underrotated when they are not, such as jumping low. Let's take out the argument and guess-work. :agree:

I think it could get interesting if such technology can be used on existing footage. Suppose one of Shizuka's jumps were underrotated in Torino.

I think that we would find that all of Shizuka's jumps were prerotated or uderrotated or both.

If you think about it, this is an immense technological challenge. It's not just looking at a movie in computer-enhanced slow motion.

A TV or video camera takes 30 frames per second. A skater can spin at 6 revolutions per second. That means that between one thirtieth of a second frame and the next a skater has rotated 72 degrees!

I don't know how many 30ths of a second elapse between the time the skater begins to lift part of the blade off the ice and the exact instant of "all clear." I think there would have to be other kinds of sensors involved than just visual ones. Maybe the skater's suit could be wired and there would be an electric current flowing through the ice or something like that.

beep_beep · Aug 10, 2008

Mathman said:
A TV or video camera takes 30 frames per second. A skater can spin at 6 revolutions per second. That means that between one thirtieth of a second frame and the next a skater has rotated 72 degrees!

I don't know how many 30ths of a second elapse between the time the skater begins to lift part of the blade off the ice and the exact instant of "all clear." I think there would have to be other kinds of sensors involved than just visual ones. Maybe the skater's suit could be wired and there would be an electric current flowing through the ice or something like that.

There are high speed cameras, that can take 1000fps (some take 5000fps). Those are used for filming special effects and science applications. The camera frame rate is not a problem.

I don't think putting electrified sensors on the ice is a very good idea, with the whole water+electricity thing...

(but again this could be a very efficient method to get skaters to avoid falling or putting their hands on the ice... they would get zapped ! :laugh:

)

Joesitz (RIP) · Aug 10, 2008

To me, there is a proposal by enlight to improve the work of the Technical Panel. In fact, it may eliminate the Technical Panel if such a suggestion is approved.

There are two scenarios going on (and probably will continue to go on) between, the fans who will not speak out on anything unless the Official word is spoken, and those fans who enjoy discussing possiblilities without regard to the Offficial word. Those fans are the same ones who will not discuss an individual skater's performance without reading the judges' protocols first. The others think they should speak out before the protocols are listed. There seems to be a lack of personal imagination. who need ISU guidance.

I, of course, am of the latter thinking and wondering what is wrong with having a device to clear up the matter of underrotations and wrong edge takeoffs?
I am all for a device, electrical or mechanical to improve the judging of a sport which has never had a 100% clear decision. Instead we get problems with electricity, cost effectiveness, computer problems, etc. as if we are giving all the reasons for the ISU to deny such an undertaking. So what? The proposal is excellent to improve some scoring debates.

All a poster has to say is whether or not they would think a device may improve the scoring system or that it will not improve the scoring system. If the proposal ever gets to the officials, they will make the decision on its usefulness.

If the impossible comes to task, and the ISU approves such a device, I guarantee all the reasons for non approval by the posters will suddenly change.

Mathman · Aug 10, 2008

I think the question is not whether Enlight's proposal should be carried out -- of course it should. I think there is 100% agreement on that, and Speedy would be the first to go along with the use of such technology if someone could figure out a way to adapt it to figure skating.

So, in the spirit of, "it's better to light a candle than to curse the darkness," here is my best shot.

You put tiny pressure sensors in the bottom of the skater's boot. When the skater jumps, you get a big spike as he/she pushes off, then an immediate drop as he/she lifts off. Transmitters located at each end of the blade send out a signal to a bunch of receivers around the circumference of the rink. From this it is easy enough to determine the exact orientation of the blade at the moment of lift-off and at the moment of impact on the landing.

There might be a slight delay on lift-off depending on the mechanics of the jump. This would be OK, however, because it would allow an acceptable pre-rotation on Salchows and loops.

Note that there are two very different propsals being discussed together on this thread. One is Enlight's original proposal in post #1 about rigging up some kind of computer that will calculate whether a jump is rotated 2.789 revolutions or 3.0165 revolutions.

The other is Joe's idea of showing slow motion replays to the audience in order to cut down on arguments between the fans and the referees. They are both good ideas, IMHO.

bethissoawesome · Aug 11, 2008

I'm sure this is undoubtedly a way to determine precisely the amount of revolutions, etc. a skater has completed in the air while jumping... developing the technology could take quite a while though. I imagine it would also be used for other things as well. For example, Mathman's idea about putting sensors in the boots... this could also determine if skaters are really doing the correct moves in footwork sequences, what edges they are on, etc. For now, I think the easiest way to deal with the issues if people want an immediate solution would be boot sensors and utilizing a camera with very high fps for slow motion replays (the slow motion replays should be show to the audience and judges alike, and the call made on the jump should be made only after the replay is watched).

I definitely would love to see technology come into play in skating to take out the guess work and debates between fans and judges. It's all too common when you hear an announcer on TV getting hyped up about the perfect triple-triple a skater has landed and then notice that judges have downgraded it. Precision is always a good thing.

Mathman · Aug 11, 2008

Joesitz said:
All a poster has to say is whether or not they would think a device may improve the scoring system or that it will not improve the scoring system.

Joe, a poster can say whatever he or she wishes.

...the fans who will not speak out on anything unless the Official word is spoken,... Those fans are the same ones who will not discuss an individual skater's performance without reading the judges' protocols first...

I just wish I could get you to give your opinion on the issue rather than your opinion (a negative and insulting one) on other GS members and fans. :disagree:

ChrisH · Aug 11, 2008

Mathman said:
You put tiny pressure sensors in the bottom of the skater's boot. When the skater jumps, you get a big spike as he/she pushes off, then an immediate drop as he/she lifts off.

Would you define takeoffs and landings by zero pressure or some middling value? Skaters vary as to how much they point their toes on takeoffs and landings. Yu-Na Kim's lutz and flip is aided by digging in the blade rather than the toe pick. Her pressure transition would be sharp while those using their toe pick would have a more gradual transition. When a ballet dancer is on her toes, isn't she more "in the air" than someone flat footed? How "in the air" is defined can vary by as much as half a turn.

PolymerBob · Aug 11, 2008

ChrisH said:
Would you define takeoffs and landings by zero pressure or some middling value? Skaters vary as to how much they point their toes on takeoffs and landings. Yu-Na Kim's lutz and flip is aided by digging in the blade rather than the toe pick. Her pressure transition would be sharp while those using their toe pick would have a more gradual transition. When a ballet dancer is on her toes, isn't she more "in the air" than someone flat footed? How "in the air" is defined can vary by as much as half a turn.

Actually, the laws of physics provide a simpler answer. When a skater is in the air, measured vertical acceleration becomes essentially zero. The skater becomes weightless.

antmanb · Aug 11, 2008

Joesitz said:
I am all for a device, electrical or mechanical to improve the judging of a sport which has never had a 100% clear decision.

How could we ever have a machine that gives you a 100% clear decision if you can't even find a large number of people who agree on how far the jumps actually rotate/what a wrong edge take off is?

Petkevich tells us that the salchow and loop jump are in fact "half rotation" jumps. That the take off edge deepens to such an extent on the take off that by the time the skaters presses down on the toe pick to jump the blade has completed half a turn on the ice.

Many jump specialists (including coaches at my rink) disagree with Petkevich and say that the edge should not come round as far as half a turn but that the bend to jump should be much quicker to sto pthe edge coming round as much and that any twist or turn of the blade should only ever happen on the toe pick as the skater pushed up to jump.

So which of these two scenarios would the machine be programmed to allow for? The important question of defining the jumps such that a machine can say yes or no would have to be resolved before we ever get to the stage of developing the machine.

Ant

beep_beep · Aug 11, 2008

Joesitz said:
All a poster has to say is whether or not they would think a device may improve the scoring system or that it will not improve the scoring system. If the proposal ever gets to the officials, they will make the decision on its usefulness.

If the impossible comes to task, and the ISU approves such a device, I guarantee all the reasons for non approval by the posters will suddenly change.

Did I ever say I didn't approve it ? Because I do.

But in order to come up with a solution for a problem, one has to wonder how it's going to be accomplished, and the possible problems with a given approach. Isn't that what's being discussed here ?
IMHO, there's no doubt that technology can help with these skating issues, all I'm saying is: it won't be easy.

Going back to the proposal section, pressure sensors can be good, but I'm not too sure if the skaters would be happy with someone putting something in their boots.
But anyway, suppose we come up a system, how it would be used? In tennis, not all plays are defined by hawk eye. The umpire makes all the calls. If a player feels that the call was wrong, on either side of the court, then he or she can challenge it. The rules say that when you call the challenge, whatever shows on the screen is the final decision. Neither the umpire nor the player can challenge it. (That's why no one complains after the challenge :rolleye:

)
How would that play out on figure skating ? Would ever jump of every skater be automatic analyzed? Would the skaters get to challenge the tech pannel decision and then he system would come to place?
Hummm, something to think about...

dorispulaski · Aug 11, 2008

One way to set it up so that the machine can pass a jump if it's absolutely sure it is correct and force the judges to review film if the machine cannot absolutely assure it's OK. I think Joe's idea that this review film be simultaneously broadcast on the Jumbotron is a good one. It would both tell the audience something wasn't perfect and fill the time you're waiting for the review to be done with something other than with the ISU's old film clips of speed skaters and old performances.

That would be a plus over the current system, I think.

Currently judges seem to review more frequently & scrutinize more closely skaters with a reputation for downgrades, sometimes missing the occasional flutzes and underrotations of skaters who do not have that reputation because they do not review them.

A possible solution to underrotations?

enlight78

Joesitz (RIP)

Mathman

beep_beep

ChrisH

Joesitz (RIP)

Mathman

PolymerBob

enlight78

Mathman

beep_beep

Joesitz (RIP)

Mathman

bethissoawesome

Mathman

ChrisH

PolymerBob

antmanb

beep_beep

dorispulaski

Wicked Yankee Girl

Similar threads

Connect with us