PDA

View Full Version : Hersh article



Pages : 1 2 3 4 5 6 7 [8] 9 10 11 12 13 14 15 16 17 18

gkelly
11-03-2011, 07:58 PM
But if the skater falls multiple times, doesn't the mistakes cumulate. I.e the skater falls. Audience braces themselves okay hopefully said skater will get back into, and then the skater falls again, audience braces themselves, and then skater falls again. The cumulative effect of all the falls together is not good.[quote]

You seem to be one of those fans who spends programs reacting to and anticipating falls. I'm sure you're not alone.

Other fans may focus more on other aspects of the program and more or less forget about a fall once the skater gets back into the program, unless the skater lets it affect the rest of the performance. Not all falls are equally disruptive.

So a fall with quick recovery is more disruptive for you than for other viewers.

Your personal reaction is not the basis for the whole scoring system. But if you decide to train as a judge and if when you're judging you continue to find falls to be severely disruptive no matter how quickly the skater recovers, you can reflect that in your component scores.

[quoteBut for me I find the concept of handing someone 6 points for a fall offensive. Thats like giving every kid in a tournament a trophy whether they won or not.

I think what you're referring to here is handing someone 6 points for rotating four times in the air (but then falling on the landing). The points are not for the fall, but for the takeoff skill and rotational that allows those four rotations to happen, something most human beings are not able to achieve.

We've discussed this before. My opinion continues to be that the negative GOEs for quads should be larger, as they were a couple years ago, so that rotating four times but falling would earn no more points that rotating three times and landing beautifully.

As for other falls, I think they're adequately penalized.


Or every child a good grade whether the grade deserve one or not. Except this is an Olympic level sport and we are suppose to be judging whose the best.

Well, it's an Olympic level sport at the highest level and a recreational sport at the lowest levels and a developmental sport at middle levels. Some rules vary from one division to another -- e.g., novice rules are often different from junior, and some junior rules are different from senior. But even the worst skater who qualifies to compete at the senior level has to skate under the same rules as the world champion. So the senior rules have to make sense for the whole gamut of senior-level skating.


men? I was talking in general... maybe I should've said 40-190 range. rules should apply across all disciplines.

I agree rules should apply across all disciplines and across all competitors in the same event.

It might make sense to lower the fall deduction for novice level and below where most jumps are doubles (i.e., the value of almost any element with -3 GOE will be less than 1.00, so the fall deduction gives them negative net worth), or it might make some sense to raise it a little for junior and especially senior levels, where triple jumps are the norm. But whatever the deduction, it needs to be consistent across all competitors in the same division.

1.00 is a nice easy number to work with, which I'm sure is why it was chosen.


And why doesn't the ISU reward jumping with one or both hands overhead with a difficulty mark? It clearly is more difficult: The center of gravity and take off are changed to make the jump happen. Why isn't the USFSA suggesting this?

There is already a place for judges to reward this -- it's a bullet point toward positive GOE. Sometimes this positive point gets canceled out by other weaknesses in the jump, so the final GOE may end up being 0 or negative even with credit being given for the difficult position. But in general a clean jump with hands overhead will get positive GOE, +1 or +2 or occasionally +3 depending on how much else is good about the jump.

Are you suggesting that it be built into the base mark somehow? E.g., triple lutz is worth 6.0 in the scale of values, triple lutz with one arm overhead is worth 6.1 or 6.5 or whatever the consensus is of its difficulty, triple lutz with two arms overhead is worth 6.2 or 7.0, etc.? That would be pretty hard to codify, given all the possible enhancements that add difficulty to jumps.


As for what a fall on a non-element should count as, I think that should be very variable as different falls impact the performance to different degrees. And that's exactly what the rules state. It's at the judges discretion as to what to take off the PE mark based on their own judgement as to how the fall impacted the performance.

Well, they all get the 1.0 fall deduction. But I agree, how much it disrupts the performance of the program as a whole will vary from one fall to another, and the perception of the disruption will vary from one observer to another. So it's an area where judges have discretion on how much to penalize, and few of them are as sensitive to falls as bek seems to be.


In practice it's not very variable though. The variability range btw different program components is rather small (usually btw .25-.75 pts). One could argue whether that's by design (judging corridor) or execution (judges generally keep all pcs within a tight range even if technically they could vary it more), but I think it's difficult to argue there's tremendous variability btw the different program components.

Not sure what this refers to:

Variation that each judge gives between highest and lowest component for the same performance?

Variation in the average of the whole panel between highest and lowest component for the same performance (this will tend to be smaller than the individual judges' variation because of averaging effects, and dropping high and low marks)?

Variation in the same skater's scores for the same component between performances with and without falls?

agalisgv
11-03-2011, 08:25 PM
Not sure what this refers to:

Variation that each judge gives between highest and lowest component for the same performance?

Variation in the average of the whole panel between highest and lowest component for the same performance (this will tend to be smaller than the individual judges' variation because of averaging effects, and dropping high and low marks)?

Variation in the same skater's scores for the same component between performances with and without falls? Sorry I wasn't clear. I meant when you look at all the scores that comprise the total PCS (tr, ch, ss, pe, in), even when one area was clearly deficient relative to the others (eg, having no transitions, or terrrible interpretation and/or choreo), the score from that component will not be that far off the score from a stronger component (eg skating skills). You just don't see judges giving out 9's for skating skills but 6's for transitions. So high component score in one or two components (eg ss and tr) can have the effect of raising scores in the other component areas across the board. That can result in relatively little to no deduction in the pcs for non-element falls.

Personally I would rather have a non-element fall tabulated elsewhere than necessarily incorporated into the pcs, but then the automatic deduction should be greater than just -1 then.

This isn't a burning issue for me though. I just think intuitively it would work better that way.

gkelly
11-03-2011, 08:38 PM
Sorry I wasn't clear. I meant when you look at all the scores that comprise the total PCS (tr, ch, ss, pe, in), even when one area was clearly deficient relative to the others (eg, having no transitions, or terrrible interpretation and/or choreo), the score from that component will not be that far off the score from a stronger component (eg skating skills). You just don't see judges giving out 9's for skating skills but 6's for transitions. So high component score in one or two components (eg ss and tr) can have the effect of raising scores in the other component areas across the board. That can result in relatively little to no deduction in the pcs for non-element falls.

Ah, yes, this is a known issue. There are exceptions, though -- a few judges who do dare to give out ranges like that when warranted.

On the one hand, judges need to be braver about separating the different marks for the same performance.

On the other hand, people who only watch elite skating need to be aware that the 1-10 scale covers the full range of skating ability, so scores in the 4s and below are going to be very rare at the senior international level.

Sometimes, though, an otherwise competent skater is deficient in one or more areas of program construction and presentation, so low scores on the relevant components would be justified.


Personally I would rather have a non-element fall tabulated elsewhere than necessarily incorporated into the pcs, but then the automatic deduction should be greater than just -1 then.

So you'd keep the -1 for a fall on an element, along with the applicable -GOE?

But then falls not on elements should earn larger fall deductions, maybe -2 each?

Hmm, maybe. But that could end up having harsh consequences, especially if pair skaters or dancers take their partners down and double the high deductions.

Pre-IJS example: http://www.youtube.com/watch?v=epZTJBGbpXs

It would also be especially harsh for skaters at the lower skill level for their event. 2.0 may be a negligible penalty for the top skaters who are earning well over 100 points for a freeskate but more significant for skaters who earn half that on a bad day -- and if they're falling between elements, they're probably having a bad day.

agalisgv
11-03-2011, 08:58 PM
Ah, yes, this is a known issue. There are exceptions, though -- a few judges who do dare to give out ranges like that when warranted.

On the one hand, judges need to be braver about separating the different marks for the same performance.

On the other hand, people who only watch elite skating need to be aware that the 1-10 scale covers the full range of skating ability, so scores in the 4s and below are going to be very rare at the senior international level.

Sometimes, though, an otherwise competent skater is deficient in one or more areas of program construction and presentation, so low scores on the relevant components would be justified. Precisely
So you'd keep the -1 for a fall on an element, along with the applicable -GOE?

But then falls not on elements should earn larger fall deductions, maybe -2 each? Exactly

But that could end up having harsh consequences, especially if pair skaters or dancers take their partners down and double the high deductions. However, if the deduction is moved from PCS, it would be offset somewhat in those cases (the pcs would remain higher with the deduction calculated elsewhere). But personally I don't know that a harsher deduction isn't uncalled for in some of those spills. Personal judgment though.

gkelly
11-03-2011, 09:13 PM
However, if the deduction is moved from PCS, it would be offset somewhat in those cases (the pcs would remain higher with the deduction calculated elsewhere).

Well, there's not a deduction in PCS, except as a judgment call by judges who thought the performance was negatively affected by the fall.

In the example I posted, I think a judge would be justified in not penalizing at all or only by 0.25 or maybe 0.5 in Performance/Execution. Currently, that would be on top of 2.0 in deduction for two skaters falling. As opposed to 4.0 total for two skaters falling with 2.0 deduction per fall.

Oh, and that's another point -- the penalties would be a lot harsher in short programs that free programs, since deductions are not factored the way PCS are.

So maybe keep the fall deduction at 1.0 in short programs and make it 2.0 in free programs?

RD
11-03-2011, 09:20 PM
So maybe keep the fall deduction at 1.0 in short programs and make it 2.0 in free programs?

Now there's a thought...

I always thought falls (on jumps) weren't really penalized as much as they should under this new system.

IceAlisa
11-03-2011, 09:24 PM
Now there's a thought...

I always thought falls (on jumps) weren't really penalized as much as they should under this new system.

And what an original, previously unheard of thought it is! :shuffle: Reminds me of the mention of a stockade on the podium a few pages back.

agalisgv
11-03-2011, 09:26 PM
So maybe keep the fall deduction at 1.0 in short programs and make it 2.0 in free programs?

Personally I would lean towards -3 in lp for senior singles, -2 in sp for senior singles. Then -2 in lp for senior pairs/dance and junior/novice singles, and -1 in the sp for the same.

IceAlisa
11-03-2011, 09:27 PM
How typically authoritarian.

overedge
11-03-2011, 11:43 PM
Personally I would lean towards -3 in lp for senior singles, -2 in sp for senior singles. Then -2 in lp for senior pairs/dance and junior/novice singles, and -1 in the sp for the same.

But many junior skaters and teams are doing, or trying, the same elements as senior skaters and teams. So I'm not sure why the deductions for falls should be different at different levels if they are doing the same things.

MacMadame
11-04-2011, 12:06 AM
In practice it's not very variable though. The variability range btw different program components is rather small (usually btw .25-.75 pts).
I'm not going to go back through tons of protocols to makes sure this is true. So, accepting it, you have the following:

Fall on jump: -1 + -GOE = around 2-5 points taken off the score depending on how many revolutions the jump has + possibly something knocked off PE and/or IN

Fall on non-elemenet: -1 plus .25 - .75 deduction on PE and maybe a similar deduction on IN depending on exactly how much the fall impacted the program = -1.25 - 2.5 points

So, on average, falling on a non-element gives you about half the deduction of falling on a jump depending on how much the fall impacted the program and what jump you are falling on. What seems completely reasonable to me and I see no need to change the system, particularly given how rare it is to fall on a non-element and how little they impact the program most of the time.



As for other falls, I think they're adequately penalized.
x2



On the one hand, judges need to be braver about separating the different marks for the same performance.

On the other hand, people who only watch elite skating need to be aware that the 1-10 scale covers the full range of skating ability, so scores in the 4s and below are going to be very rare at the senior international level.
This is what I was referring too when I said people need to read the rulebook and see what a 3 or a 4 in PE is supposed to mean. Because it's not something you are going to see very often at the Senior level! Even when the performance is pretty bland or there are a bunch of falls, the PE overall is going to be higher for most senior-level skaters.

mag
11-04-2011, 12:37 AM
...This is what I was referring too when I said people need to read the rulebook and see what a 3 or a 4 in PE is supposed to mean. Because it's not something you are going to see very often at the Senior level! Even when the performance is pretty bland or there are a bunch of falls, the PE overall is going to be higher for most senior-level skaters.

This.

As a reference, at BC Summer Skate the top Pre Novice ladies were scoring between 3.00 and 4.00 on PE.

Cyn
11-04-2011, 04:19 AM
Or maybe even from FSU (http://www.fsuniverse.net/forum/showpost.php?p=3265377&postcount=409). :sekret:

Because that was an absolute GEM, it bears re-posting:


Queen Tuk (Queen Tuk)
Now when she was a young one,
She never thought she'd see
People stand in line to see her skatin'.

(Queen Tuk) How'd you get so funky?
(funky Tuk) Did you do the monkey?
Name is Elizaveta,
She lives in Peter (Queen Tuk).

(Queen Tuk) Now, if I'd known
they'd line up just to see her,
I'd trade in all my money
And bought me an arena. (Queen Tuk)

Shacked up with a husky (funky Tuk)
She's my favorite Russkie!
Name is Elizaveta,
She lives in Peter (Queen Tuk).

Dancin' by the Neva, (Disco Tuk)
The guys don't wanna leave her, (boss Tuk)
Rockin' like Tom Seaver (rockin' Tuk)
She gives me a fever.

She gave her soul to Mishin.
Russian idol!
She's a Russian
They're sellin' you.

Now, when I skate,
now don't think I'm a kook,
don't want no triple axel,
Just a double like young Queen Tuk. (Queen Tuk)

She oughta win a Grammy,
Skating in her Jammies,
Name is Elizaveta, she lives in Peter,
Name is Elizaveta, I really want to meet her,
Queen Tuk!


So you'd keep the -1 for a fall on an element, along with the applicable -GOE?

But then falls not on elements should earn larger fall deductions, maybe -2 each?

After reading through this thread and the opinions/input from all sides, this is the solution I had come up with, but others beat me to the post.

Yes, there could be complications with its implementation, but overall I think it makes the most sense.

Cyn
11-04-2011, 04:22 AM
I made one small correction to your post, aaron:


Very nice summary :hat1:

Part four: Severe penalty for falls
Part five: No quads in Sochi (Lysacek still win with no quad nor 3A :P)

History repeats itself :lol:

…..And CaseyEdward's head explodes :D. Hopefully for good :saint:

PUNKPRINCESS
11-05-2011, 05:30 PM
I am not sure where in PE and IN criteria would a fall/falls have an impact:
http://www.sportcentric.com/vsite/vfile/page/fileurl/0,11040,4844-152086-169302-64121-0-file,00.pdf

I don't feel like pasting whole pages here but you can see for yourselves. I would like to know why PE and IN be affected by falls
PERFORMANCE/EXECUTION:
- Projection: the skater radiates energy resulting in an invisible connection with the audience (does anyone else snap out of the dream of a program when a skater stumbles/falls?)

INTERPRETATION:
- Expression of the music’s style, character, and rhythm: Maintaining the character and style of the music throughout the entire program by use of body and skating techniques to depict a mood, style, shape, or thematic idea as motivated by the structure of the music: melody, harmony, rhythm, color, texture, and form. The total involvement of the body and being should express the intent of the music (rarely, some stumbles and falls coincidentally go with the music, but usually they do not.)