Article:"Rewarding Failure Diminishes Sport"

[Tonichelle]

Unless I am reading this wrong it was the opposite. The technically cleaner understandable S/P program won the crowd and media over the B/S two foot a 2A by one person and vaguer interpretation.

I think you did misunderstand. Puchi said the same as what you just did. :yes:

 

[bekalc]

COP was born because people thought B/S should have won with a flawed but difficult program over S/P's clean but simpler program. They got what they wanted.

There's a HUGE difference between one small error like Anton had, and lots of errors like Chan's skate had beating out a fantastic (intricate!) skate from Daisuke Takahashi.

 

[gmyers]

I think you did misunderstand. Puchi said the same as what you just did. :yes:

I thought I was Over-reading or something and confusing myself! Because the "people" meant audience to me and were pro S/P.

 

[hurrah]

I think Chan has demonstrate his ability at the competition. It's just that some people refuse to accept. At GF, NOBODY had a perfect skate. All had errors. So who should have won then? Dai? He had poor SP skate, that's why he was 10 points behind Chan. Unless, as some like to think that SP shouldn't count.

I agree Patrick should have won overall given Dai's major mistakes in the SP, but I don't agree with Chan scoring more than Dai in the LP. I think most posters are not disputing the overall outcome, just the LP scores.

 

[Joesitz (RIP)]

I see Figuring Skating as a two prong contest winner: One prong executing elements as in Diving and the other prong as opining the best entertainer.

Elements are defined for scoring purposes and given the assigned base values. There should be no argument here. All Singles Sports have them,

Opinions of skaters' interpretations of the guidelines can not be valued. A fan can have a different opinion.There is room for argument here. Artistry is defininedby cognizant people thoughout the years. Splendid skating at the time of competition must be seen by the judges for scoring the contest's two prong system.

 

[Joesitz (RIP)]

FALLING and CRASHING INTO BOARDS show so many erors that they tell the judging: that they are not prepared. Let us not diminish what is left of sport. Partial Credit is a disgrase to any sport.

 

[Bluebonnet]

There's a HUGE difference between one small error like Anton had, and lots of errors like Chan's skate had beating out a fantastic (intricate!) skate from Daisuke Takahashi.

Sounds like all you could see in Chan's skating was errors. In the meantime, all you could remember was Takahashi's "fantastic" - by the way it was never being perfect technically, it was less difficult, and it was not his best artistically, skating. What a double standard! What you did mostly was counting errors. Ignoring BV because you were tired of using your brain, and thinking skating with most errors should lose. This is exactly what MM said chasing Axels or chasing skirts. The different views between experts and casual fans.

By the way, I am not in any sense implying that I am an expert of any sort. But at least I am trying my best to study and understand what the experts say.

I agree Patrick should have won overall given Dai's major mistakes in the SP, but I don't agree with Chan scoring more than Dai in the LP. I think most posters are not disputing the overall outcome, just the LP scores.

There is not much to dispute under the rules. If you go through the detailed guidelines in each and everyone of the segments in PCS, it will be more so.

 

[jettasian]

I agree Patrick should have won overall given Dai's major mistakes in the SP, but I don't agree with Chan scoring more than Dai in the LP. I think most posters are not disputing the overall outcome, just the LP scores.

And many has pointed out why Chan won the LP. Some just didn't like to know CoP and then start crying foul and making accusations.

 

[Bluebonnet]

FALLING and CRASHING INTO BOARDS show so many erors that they tell the judging: that they are not prepared. Let us not diminish what is left of sport. Partial Credit is a disgrase to any sport.

I think no body can argue that Patrick wasn't completely in character in his SP program. Look at him! He saw a couple of beautiful girls and wanted to show off his tricks in front of them. He went for it with full force and "bang!!!" Oooooops! Jumped up and covered it up as fast and as much as he could with "the girls were watching" in mind. What a perfection artistically! I'd rank this time's SP one of his best "Take Five" if it wasn't that flawed 3A.

 

[Violet Bliss]

I think no body can argue that Patrick wasn't completely in character in his SP program. Look at him! He saw a couple of beautiful girls and wanted to show off his tricks in front of them. He went for it with full force and "bang!!!" Oooooops! Jumped up and covered it up as fast and as much as he could with "the girls were watching" in mind. What a perfection artistically! I'd rank this time's SP one of his best "Take Five" if it wasn't that flawed 3A.

He's also learned to back off and is not getting punched anymore. Not on ice anyways.

for Interpretation.

 

[mskater93]

For everyone unhappy about results, why not lobby to have the rules changed:
1) GOEs could be a % of the BV of the jump, so the risk/reward is related to the actual element instead of a scale based on what jump it is. GOE could be a % per number added or subtracted to the BV. Also, +3 to -3 is not a wide enough range. Should be +5 to -5 range or decimal increments in 0.5's because multiple errors could be captured with better gradiation.
2) The fall penalty could be a percentage of the total score instead of just -1. It doesn't have to be huge, but ~0.5% of the total score is ridiculously low for a fall (1/245 points). So basically, if you jam pack a program with difficulty and score high but you fall, you have a higher deduction than something a little scaled back from the edge of what you can do (but if you are a little scaled back, you are less likely to take the splat). Then it becomes a risk/reward venture again and could get the complaning about a multiple error program beating a slightly easier program with less errors.

Just a randome though here...

 

[gkelly]

For everyone unhappy about results, why not lobby to have the rules changed:
1) GOEs could be a % of the BV of the jump, so the risk/reward is related to the actual element instead of a scale based on what jump it is. GOE could be a % per number added or subtracted to the BV.

Yes. Although, on the positive side, that decreases rather than increases the incentive to do a lower level element especially well. I'd rather see +GOEs kept high for simple as well as complex elements.

Also, +3 to -3 is not a wide enough range. Should be +5 to -5 range or decimal increments in 0.5's because multiple errors could be captured with better gradiation.

For the negative grades, with jumps we have 3 grades of penalty just for quality errors, and then additional penalties for underrotation and for falling. (Well, the fall deduction applies to non-jump elements and non-elements as well.) So a jump element with multiple serious errors is usually punished a lot more than whatever -3 for that jump adds up to.

With leveled elements, penalties for poor execution are not only 3 grades of negative GOE, but also the possibility of not getting credit for a feature (lower level than intended) or in some cases not getting any credit for the element at all, which can end up being an overly harsh penalty.

2) The fall penalty could be a percentage of the total score instead of just -1. It doesn't have to be huge, but ~0.5% of the total score is ridiculously low for a fall (1/245 points). So basically, if you jam pack a program with difficulty and score high but you fall, you have a higher deduction than something a little scaled back from the edge of what you can do (but if you are a little scaled back, you are less likely to take the splat). Then it becomes a risk/reward venture again and could get the complaning about a multiple error program beating a slightly easier program with less errors.

Huh, interesting. That would also solve the problem of different effect of the same fall deduction for skaters at different skill levels. E.g., the top senior man who earns 180 points and falls twice in his LP would lose more points than the weak senior man who earns 80 points in his LP and falls twice.

The guy who earned 80 wasn't "scaling back" from aiming for 180 -- he was just trying to do the best he personally was capable of, maybe trying to get out of a qual round or trying to earn a minimum TES to allow him to enter Worlds/Euros/4Cs at all, but with two falls and probably other mistakes he probably didn't succeed.

Same for a mid-level senior man who might earn 120 or so in a long program with two falls.

1 point is not a huge penalty for a guy who's starting out at 180. But raising the senior men's penalty to 2 or 3 points across the board would overly penalize the lower-level skaters, who are more likely to fall on easier elements. So using a percentage of total score would keep the penalty commensurate with the level of skating and scoring against closest competitors.

 

[mskater93]

I understand how IJS works probably better than most, and I knew what I was thinking about with gradiated or more negative GOEs. Specifically, I am thinking about a situation where there was a wrong edge take off that ended up with a hand down and then flipped out of. That gets a -3 GOE when there should actually be -4 or -5 in this situation. If each was worth a -0.5 on a gradient scale, that would be -2.0 or -2.5 versus a -3 for a fall (to reduce the difference between something with a couple errors versus an outright fall which seems to PO the casual fan more than anything )

For #2, I was thinking specifically that maybethe guys near the top would skate a little less to the edge of their ability from a transition/interpretation standpoint (maybe lose .25 or .5 point in the TR or IN score) but have less falls or visible errors.

 

[skatinginbc]

the lower-level skaters, who are more likely to fall on easier elements.

Since I'm not a skater, my imagination can be wide of the truth: I assume every skater would skate to a program that showcases the best they can do yet still not far from their comfort zone. In other words, high-level skaters choose more difficult programs and lower-level skaters skate to easier ones, and the end result is: the fall rate remains about the same. I briefly browsed through the junior men's competition results and didn't notice a higher rate of fall.
Logically speaking, a fall is a fall, and its disruption to the artistic impression is equally there, no matter if it is done by a novice or a senior.

 

[Violet Bliss]

Interest posts and perspectives.

A real skater with knowledge of IJS proposes furhter differentiation and grading on execution of an element.

A casual fan, or self appointed respresentative of casual fans - I'm not sure which at this point, sorry - insists a fall is a fall with same disruption to a program under all circumstances.

The resident cool head and infinitely patient expert analyses proposals thoroughly and points out the likely collateral damages.

 

[gkelly]

I understand how IJS works probably better than most, and I knew what I was thinking about with gradiated or more negative GOEs. Specifically, I am thinking about a situation where there was a wrong edge take off that ended up with a hand down and then flipped out of. That gets a -3 GOE when there should actually be -4 or -5 in this situation. If each was worth a -0.5 on a gradient scale, that would be -2.0 or -2.5 versus a -3 for a fall (to reduce the difference between something with a couple errors versus an outright fall which seems to PO the casual fan more than anything )

True . . . so you're thinking that a jump that earns -5 on the gradient scale for cumulative errors but no falls would be penalized the same as one with -3 but no fall now (with additional loss of points for downgrade/underrotation calls if applicable)?

For #2, I was thinking specifically that maybethe guys near the top would skate a little less to the edge of their ability from a transition/interpretation standpoint (maybe lose .25 or .5 point in the TR or IN score) but have less falls or visible errors.

I'm not sure that would be a good thing, at least from my personal point of view. I'd rather see them push the transitions and interpretation and scale back on the jumps a little. Big jumps are exciting, but complex skating is more interesting to me.

Not everyone agrees -- we can't please everyone since we don't all watch for the same things.

I also expect more falls in the fall season than by the time Worlds rolls around, as the skaters are building up their familiarity with new programs.

Since I'm not a skater, my imagination can be wide of the truth: I assume every skater would skate to a program that showcases the best they can do yet still not far from their comfort zone. In other words, high-level skaters choose more difficult programs and lower-level skaters skate to easier ones, and the end result is: the fall rate remains about the same.

Yes, pretty much. With senior ladies short programs, it becomes a bigger issue because the rules require two different triples and a double axel, and many lower skilled senior ladies don't actually have the ability to rotate and land all those jumps consistently. If they're too old for juniors, either they quit or they compete senior without the required jump skills. So either they do easier jumps on purpose, take the mandatory -3, and stay on their feet, or else they try the required jumps and often underrotate and/or fall.

But my main point is that if the skaters are trying easier elements (easier triples only, or only doubles for average novice and below) then the base values of the programs will be lower, so the same fall deduction will have a bigger effect than it will have for the guys who start out with high base values by trying triple axels and quads.

Logically speaking, a fall is a fall, and its disruption to the artistic impression is equally there, no matter if it is done by a novice or a senior.

So if we want the penalty to be proportional to the value of the program as a whole, then mskater93's suggestion of using a percentage of the total score makes more sense than using the exact same deduction value for everybody.

It seems that the consensus in this thread is that there should be a larger absolute penalty for falls by the quad squad, because 1.00 doesn't make much impact against the rest of their scores.

What I'm saying is that if you want to raise the deduction at that level, don't raise it by the same amount at lower levels, because the deduction would take away a larger percentage of the score on lower valued programs.

 

[pangtongfan]

Please. Flawed skating is flawed skating. I don't care WHO points it out... there is a reason skating is struggling and part of this is watching these flawed performances equated with brilliance. And if you choose to see that as a grand illusion, then fine. But don't confuse your illusion with the reality of the sport of figure skating...

Exactly. There is a reason the sport is almost dead in North America. All this person wrote is entirely true.

 

[Bluebonnet]

I also expect more falls in the fall season

 

[mskater93]

A fall is a fall may be true, but they aren't "equal" which is what I am proposing - making them equal.

For example:

At the Adult Gold level (the lowest scoring level in the US that is judged under IJS because the tech elements skated are less than the Juvenile level), the average program will score about 25 points (I am using round numbers that are easy to display the math). Let's say the skater has 1 fall in the program. That's a -1. That skater's total score is now 24. They lost 4% of their total score because they fell once.

At the Senior Men's level, the top 10 freestyle programs average about 165 points. Let's say average top 10 man has 1 fall in the program. That's a -1. That skater's total score is now 164. He lost 0.6% of his total score because he fell once.

In the 11-20 group of Senior Men, the average freestyle program score averages about 125 points. Let's say average skater falls once. That's a -1. That skater's total score is now 124. He lost 0.8% of his total score because he fell once.

Question: was the fall in the adult gold program more disruptive than the one in the average senior men's program or were these more disruptive than the fall in the top 10 man's program? It appears the answer is no since the self appointed speaker for the casual fan says "a fall is a fall with the same disruption to the program under all circumstances". So, why is that adult gold skater or average senior man skater penalized more heavily for their fall than the top 10 senior man as a percentage of their whole program?

Let's look at my proposal:
Let's say each fall is worth 1% of the total score at all levels.
So with the same three examples:
The adult gold skater gets a 25 but has a fall. With a 1% deduction, this skater now scores a 24.75.
The average senior man gets a 125 but has a fall. With a 1% deduction, this skater now scores a 123.75 (versus the 124 he would have scored if we stayed with the flat -1 deduction)
The top 10 senior man gets a 165 but has a fall. With a 1% deduction, this skater now scores a 163.35 (versus 164).

The score doesn't change that much for any of them, but a percentage loss makes the "fall is a fall is a fall" more real.

Not only that, but it would encourage more risk taking at the lower levels because a fall on your iffy element on your 25 point program won't cost you a full point penalty, but 0.25 points. Maybe you try that iffier element and you rotate it and fall a couple times in competition because of the "new" percentage rule but there's not a complete disincentive to trying it competitively as there currently is.

FWIW, at first glance a 1% deduction wouldn't have changed the GPF Men's results, but it could have reduced the complaining about it.

ETA: Yes, < and << would still apply as do loss of levels on spins/steps for mistakes in my gradiated GOEs. I just wish to distinguish between multiple errors that would get a -3 currently versus an outright fall. I believe that an outright fall should receive the bottom GOE and multiple errors would receive something slightly better than an outright fall (like a -2.5)

gkelley, I love blade to ice work as well, but I think a skater like Chan could drop a few of his transitions (that sometimes just look like transitions for transitions' sake), probably get a little more consistency on the elements, and no one would really miss what he dropped from the program, which was what I was suggesting in my drop from the edge of your skills to a little more comfort zone.

 

[gkelly]

Not only that, but it would encourage more risk taking at the lower levels because a fall on your iffy element on your 25 point program won't cost you a full point penalty, but 0.25 points. Maybe you try that iffier element and you rotate it and fall because of the "new" percentage rule. You boost your 25 point program to a 25.5 because you got credit for it (it was clean but then you fell on it for -3 GOE factored) and lose .255 off your total score instead. So now, your average skater moves up to 25.25 from 25.

Especially because the "iffy element" for an adult gold skater might be a single axel (1.1 base value, 0.5 after subtracting the factored -3 GOE) or double toe loop (1.4 BV, 0.6 total with -3 GOE), so the 1.0 fall deduction would cause the final effect of the iffy element with fall to be negative -- even moreso if the jump was also underrotated/downgraded. The skater would have ended up with more total points by skipping the jump and just doing a three turn.

Whereas the iffy element for a senior man might be a triple axel or a quadruple toe loop. Even after -3 GOE and fall deduction, even after an underrotation call and lowered base value, the attempt still earns positive points.
(That's why I think the -GOEs need to be larger for those jumps, but to encourage risk at that level they would still come out positive not negative on net points.)

On the other hand, suppose a skater just trips on a rut or clicks blades on a crossover and goes down not on an element. There's no loss of base value or GOE penalties involved. But is the disruption to the program really 4% of the whole program for an adult gold skater and 0.6% of the whole program for the elite senior man?