Free the free program! (Singles)

[gkelly]

This post is my attempt to define parameters for brainstorming how the rules for the singles free program could be redefined to allow more freedom for skaters, coaches, and choreographers to make choices about program content and structure to better meet their own vision while still providing a level playing field in which skaters with different strengths or different visions can be meaningfully compared. What would our ideal free program rules look like?

Let's take the following premises as given:

*The free program will continue to follow an earlier phase of competition such as the current short program in which the focus is on completion of required elements.

*IJS is here to stay -- in particular, the concept of assigning points for elements, with one set of officials identifying the elements and in some cases assigning levels to determine their base scores according to a scale of values, and another set of officials assigning grades of execution and also component scores for various aspects of the program as a whole.
The specific details are up for debate.

*We're talking about ISU rules for international junior and senior competition, which will also be used in domestic competitions for junior and senior skaters, however each country defines those levels, some of whom will not be good enough for international assignments.

That means, probably, that at a minimum we should expect skaters governed by these rules to be capable of executing all the different kinds of turns (with preferences as to direction and entry edge), all the basic spin positions, and all the double jumps, although 2A may be a challenge.

At the upper end, we there may be some very best skaters who can execute all the triple jumps up to axel, in all sorts of combinations, as well as a couple of different quads, and enough different difficult non-jump skills that can serve as features toward the highest level in all of their leveled elements, although there will not be time or energy to execute all these skills within a single 4- or 4.5-minute program.

Most junior and senior skaters will fall somewhere in between and will have some kinds of skills that come much more easily to them than others. Our rules need to apply to all of them.

*We do want to maintain some minimum expectation of what constitutes a "well-balanced free program." Details are up for debate.

*We want the free program to represent each skater's best package of maximum difficulty with maximum quality that they can fit into 4-4.5 minutes. That includes not only the quality of the elements but also the quality of the program as a whole, both athletically and artistically.

With that in mind, let's discuss changes we would like to see to the well-balanced program rules, the scale of values, the definitions of levels, etc., etc., for the singles free program specifically.

Later maybe we'll address pairs and/or dance. Or someone else can feel free to start a thread on those topics now.

But first, if you want to make any adjustments to my starting premises, best to bring those up first.

[Marco]

If the short program requirements on jumps and spins are to stay, I don't feel the need for them to be repeated in the free as well.

That is, the elements in the free should be marked for quality and matching the music / effect, instead of just focusing on maximising difficulty. I don't mind that spins and steps in the free have no levels (but there should be high GOEs for high quality). Complexity can be a GOE consideration for spins and steps.

For well balanced programming, there should be a PCS consideration for SS or P/E to reward a program with variety in jumping, spinning and steps. Programs with an under-representation of certain jump types (less than 4 types of takeoffs), spin positions (insufficient camels) and steps (just simple 3 turns) should be penalized. Consecutive jumping, consecutive spinning, over-front loading and over-back loading should be penalized too.

I also think deviation in the 5 PCS categories should be allowed.

[Zemgirl]

Basically what Marco said, especially the part about no levels. The one place where I disagree is that I'm not sure how you can penalize consecutive jumping, consecutive spinning, over-front loading and over-back loading and still keep the free program free, there are too many rules and restrictions already. What I would like to see done to facilitate highlight distribution is a different way of giving bonuses for jumps - maybe a smallish bonus for second half jumps but a relatively large one in the last minute (e.g. S/S and their throws at the end of the program deserve more of a bonus that a 3-2 right at the halfway point). In general I would prefer to see rewards and bonuses for what we want to see, rather than penalties for what we don't want.

I also think that they should somewhat relax the Zayak rules; perhaps any jump that's now a Zayak violation can get reduced credit instead of being an invalid element. We can name this rule after Oda

And maybe the program components should be weighed differently between the programs, with the relatively technical ones being worth more in the SP and the more artistic ones worth more in the free skate.

This is just regarding the free skate, which is indeed not very free these days. Beyond that, I think the IJS needs a more serious overhaul post-Sochi, rather than the usual tinkering the ISU does.

[skateboy]

Just have to say, I LOVE the thread title! Agree with the post above: there is nothing "free" about the free program these days.

Not much more to add, except to say that I have always despised (and still do) the ridiculous, unending, coma-inducing, suicide-provoking, snail's pace footwork sequences that must feature EVERY turn and change of edge in skating, as well as the spins with multi-changes of edge and position.

They don't flatter the skaters nor the programs and, while they are difficult, I don't believe that harder is better, in this case.

[Marco]

Basically what Marco said, especially the part about no levels. The one place where I disagree is that I'm not sure how you can penalize consecutive jumping, consecutive spinning, over-front loading and over-back loading and still keep the free program free, there are too many rules and restrictions already.

Not a rule, but overall impression of how elements are distributed. I hate that Ando could get away with doing 5 jumping passes in a row right after the 2 minute mark.

[gkelly]

There are a couple of ways we can approach this.

If we agree that something is valuable enough that it should be officially rewarded, do we

1) Build the rewards into the rules in a way that everyone who does it will automatically get the reward? That's the fair level playing field, but if the incentive is large enough then everyone will try it who can, and we won't see as much variety of visions. Even if it's optional, the result will be the same as if it's required (less freedom).

2) Explicitly define it as something to be valued (in the criteria for one or more components when it comes to program layout, in the GOE bullet points when it comes to elements) and then leave it up to the judges to reward it if they notice and personally determine that that criterion as evidenced in that skater's performance outweighs any weaknesses in the same element or same component? That increases the subjectivity and doesn't guarantee that skaters will be rewarded, but it also allows for more fluid adjustments in the scoring process: new scoring guidelines can be issued at any time to remind judges to emphasize things that the technical committees have been seeing skaters do that don't seem to have been rewarded.

When I get a chance, I'll try writing posts about different kinds of elements and different program construction issues.

[Susan M]

I have looked over the suggestions so far and agree with some of them but also think we need to remember that this is supposed to be a sporting competition. IMO, like it or not, technical skills need to be an important part of ranking the FS as well as the SP. I think notions like realizing artistic vision are certainly important for show programs and exhibitions, but in Olympic sports, the focus has the be the athletic competition. I do agree, though, that the programs we have been seeing the last few years are choreographic fails with rare exceptions and I think that is part of the reason the sport has so many fewer viewers now than a decade ago.

I also disagree with one of the thread's basic premises about having the same rules across all levels of juniors and seniors. They already treat juniors differently for SP requirements, so why not the FS? One of the changes I would make for seniors is to treat single rotation jumps as transition moves rather than scoring them as jumping elements. This would allow, for example, walleys and split-flip jumps in footwork or jump entrances. I miss those. The rules already treat half-rotation jumps this way. The only tricky aspect here would be how to score a popped jump. The rule makers would need to make a choice - either regard it as a non-element (allowing the skater to add another jump later) or score it based on what was intended (the way they treat failed combos now).

I also disagree with the premise that the rules should be structured to allow skaters to hide their weaknesses. IMO, this would be just wrong at the elite level and especially bad for developing skaters, who would no longer be forced to work on their weaknesses. IMO the rules should favor champions who have balanced skills and are well rounded. We see too many skaters today with programs full of edge jumps and almost no toe-assisted jumps or full of toe jumps and hardly any edge jumps. This is bad, IMO. Skaters should not be able to succeed at elite levels unless they can do both edge and toe-assisted takeoffs well.

If I were to revise the FS rules (thinking singles), I would -

- Find a way to make sure skaters are attempting jumps of 2 or more revolutions in all 6 different jump takeoffs. There are a number of ways this could be accomplished, such as by eliminating or reducing credit for repeated triples when the other jumps are not in the program. Simpler still, just make any double, triple or quad worth 80% of its value when repeated, regardless of other program content. Another approach would be to just have a significant deduction taken when all 6 takeoffs are not used or bonus awarded when they are. (It shouldn't matter if the jump is doubled or a fall. As long as it is scored as a jump element, the take-off would be credited.)

- Do something to get away from the excruciatingly long, herky-jerky footwork sequences that can now occupy 45 seconds or more of the FS. Now, they are meandering all over the ice, with stops and starts, lurching from side to side like drunks, even retrogressing sometimes, until my nerves are screaming "just get to the end, please." I think I would start by eliminating Level 4. I would also add time limits for completing the pattern/sequence - 25 or 30 seconds for the one scored with levels and 20 seconds for the choreo sequence.

- I would also eliminate level 4 for spins, to eliminate the need for some of the uglier moves we are seeing today. Also, instruct the judges when assigning GOE not to give credit for features that make the move look awkward or ugly. (I'm looking at you, girl who isn't flexible enough for even a half-Biellmann but still hoists her leg up into a really ugly pointy-knee posture in a painful looking spin. IMO, the scoring system should be telling her coach that if she can't achieve an attractive position, then dump it and look for another way to get your levels.)

- Revise the guidelines for assigning positive GOE to all moves to put less focus on BS like arm positions and blade grabbing and instruct judges to give +3 for moves that are simply, perfectly done. IMO, some of Kulik's clean, soaring, fully rotated in the air triples should get +3 even with no other features. (I think this may already be theoretically possible, but the judges are not doing it.) Same for spins. There should be more points for being really fast, really well centered, and actually attractive than for achieving multiple less attractive positions.

- I like the suggestion above of giving extra credit for jumps in the last minute of the FS.

- Limit total jump elements to seven or eight and reduce max combos/sequences to 2. That would still make for as many as 10-11 different jumps and we don't really need any more than that to score programs.

I was interested in Kurt Browning's thoughts after competing with a COP program for the first time. His observation was that the non-jump elements have gotten so complicated and the programs so full of elements and transitions that the skaters never get a chance to rest and regroup. This means that the choreo also never rests and the viewer is continuously bombarded with content so we too never get a chance to visually/mentally rest.

In choreography, in order for the highs to have impact, there also have to be lows. Before COP, my favorite section of an LP was very often the little slow section at about the midway point, after the skater finished his first big jumping passes and a spin or two, where they put things like spirals and small choreographic elements to highlight their edges while regrouping for the rest of the program (usually starting with the second triple axel). Look, for example, at how the softness of the music and preceding choreo makes the 3A have such impact here.
http://youtu.be/QewAP6vweWk?t=3m18s
(Yes, it's Kulik again, my favorite male skater along with Robin Cousins.)

So, a lot of my suggestions are geared to reducing the amount of time the elements consume in the program to leave more time for more subtly and fully realized choreography (and less flailing and in your face choreo spewed out in short bursts). I think the first two short dances kind of illustrate this point. The Golden Waltz patterns took nearly a minute of the available 3 minutes, making the whole dance just a pack of loosely connected elements. Last year's CD patterns took only about 30 seconds of the dance, so we got a lot more choreography and much more coherent dances. I think this effect can be applied to singles choreo as well. Having extra time for choreo will also go a long way toward achieving the artistic vision goal.

[gkelly]

I have looked over the suggestions so far and agree with some of them but also think we need to remember that this is supposed to be a sporting competition. IMO, like it or not, technical skills need to be an important part of ranking the FS as well as the SP.

Absolutely we can't lose sight of that. We're not designing rules for an entertainment-based professional competition. Although that could be fodder for a separate thread.

I also disagree with one of the thread's basic premises about having the same rules across all levels of juniors and seniors. They already treat juniors differently for SP requirements, so why not the FS?

I didn't mean that all rules for juniors should be exactly the same as those for seniors. Already they are not (30 seconds less, and no choreo sequence in the junior LP under the current rules). But if you want to make further differences between junior and senior, please elaborate on what you should think be required or allowed at one level and not at the other.

What I intended but maybe didn't express clearly enough is that the rules for all juniors should be exactly the same, and the rules for all seniors should be exactly the same. I.e., recognize that some juniors will struggle to complete a double axel, and some seniors will struggle to complete two different triples (as required in the SP) as well. They're required to try in the SP, although not all will succeed, and they won't place well as a result. Should they be required in the free program, or should they be free to plan all doubles and build a strategy to place as high as possible on the strength of other skills?

More later...

[gkelly]

My two cents on jumps

Things I like to see and want to reward:

*Clean programs, landings with good flow in particular

*Variety of jump takeoffs

*The hardest elements most contenders at that level are attempting

*Even harder elements that push the envelope

*Jumps woven into the choreography with creative entries, air positions, and/or exits

*Single and double jumps from unusual takeoffs (walley, inside axel)

*Jumps rotated in both directions

*Unusual/creative combinations and sequences

*Delayed/open single jumps with suspension in the air -- delayed axel, split-flip or split-lutz, etc.

Clearly, some of these things are almost mutually exclusive, except for Superskater who can do everything. And even that skater will have to pick and choose which skills to include in her program, or which to spend the most practice time on, based on what she expects to be best rewarded for.

Things I don't like to see:

*Empty programs with simple stroking leading into most jumps

*Multiple jump passes in a row (at the beginning of the program, or at the beginning of a bonus period)

*Skaters repeating the same two or three takeoffs in most of their jump passes and leaving out some of the standard takeoffs entirely

*Skaters attempting jumps they have no hope of completing successfully because it's expected or because they expect to fail in a way that will earn them more points than an easier jump performed successfully

*All the various errors of execution that result in negative GOE


With the multi-triple era of 6.0 especially since the end of figures we saw that number of triples (and double axels) landed successfully or close to it tended to be the determining factor for technical merit. And with IJS we've seen that skaters, with good reason, have generally trusted more in pursuing the highest base value they have a good chance of completing than in trusting judges to reward them in GOEs and PCS for better quality -- tempered to a degree by guaranteed harsh penalties for underrotations/downgrades, wrong edge takeoffs, and falls.

There's also the limit on the total number of jumping passes, which means a skater who can fill the allotted 7 or 8 jump passes with triples, quads, and/or double axels is not going to waste any of those jump passes on single or double jumps with low base values even if they could enhance or vary them in ways that might be comparable in actual difficulty to a double axel or easier triple.

One way to "free" the free program would be to make fewer rules and leave everything up to the judges. But unless the judges are given more discretion with GOEs to reward areas of difficulty other than number of rotaitons and the values of the highest GOEs are large enough to compensate for lower base values, it won't be worth it.

So I think incentives need to be built into the system to encourage different strategies for skaters to demonstrate their best jump skills as well as a well-balanced variety of jump skills.

*Introduce a bonus multiplier to the value of the last jump in a combination (and maybe a smaller bonus to the first jump)

*Either remove the 80% penalty for jump sequences -- which would now be worth less than combos anyway if the combos get bonuses -- or else open up the definition of jump sequences to be more like it was back at the beginning of IJS, with one 180-degree turn on the ice allowed between each jump

*Add a bonus for combinations or sequences in which jumps are performed in opposite directions

*Introduce a leveled element called "small jump sequence" that would consist of jumps with 1.5 revolutions or fewer and that could be used to fill one jump pass. There would be lots of different ways to earn higher levels, to accommodate lots of different kinds of jumping skills other than rotation. Maybe some of the hardest skills could be required for level 4. Maybe level 1 is only worth 1.0 base value, just something to throw in at the end of the program for tired skaters, but level 4 could be worth 4.0, more than a double axel, if it requires skills or combinations of skills that the average skater will not be able to master.

*If we're keeping the choreo sequence, maybe write something into the rules to guarantee that enhanced single jumps or intentional 1.5 jumps within the sequence will not fill jump boxes

*Write the double walley and double inside axel into the scale of values, with base values at least equal to double axel

*Add some sort of bonus for attempting 6 different rotational jump takeoffs, and a larger bonus for 8 different takeoffs

*Give women an extra 15 or 30 seconds in their programs so they can add one more element to be on par with the men's content and have time to catch their breath, with the option that that element could be an 8th jump pass (or 4th spin or another sequence; to be discussed later)

*Increase the values of positive GOEs on all jumps, and also increase the values of negative GOEs on triple axels and quads so that rotating without landing cleanly is not worth more than a high-quality jump with lower rotation

Lots more rules, yes, but they are designed to offer more options for skaters to showcase more variety of difficult jump skills and be rewarded commensurately with difficulty, not to limit them further.

[gkelly]

I have much less to say about spins. The main thing I'd like to see is raising the value of the GOEs, so that it's always more valuable to get a higher GOE than to get a higher level.

I see so many spins, especially laybacks and flying sitspins, that start out fast and centered with decent to excellent positions and plenty of revolutions already, and then they change to one more variation at the end for that fourth feature and lose the speed and/or centering and/or quality of position.

That might make the difference between +2 and +1 (or 0 or -1 if the loss of quality was significant, even to the point of an outright error). I'd like to see the rules make it worth the while of these good spinners -- of whom there are many more these days -- to quit while they're ahead.

But what about the skaters whose bodies will never have the most beautiful positions and who don't naturally spin fast or center well? Those are the ones who are most painful to watch when they try for higher levels. But if they know they're doomed to 0 GOE at best, the only way to get more points from spins would be to try to raise the levels.

Maybe remind the skaters, and the judges, that originality and element matched to musical structure are bullet points for positive GOE, so mediocre technical spinners can aim for positive GOE that way. And they'll have to keep the quality at least acceptable to avoid minuses.

I don't like the idea of getting rid of spin levels in the free program entirely, because I do think that a difficult spin done with +1 or +3 quality should be worth more than an easy spin done with the same quality.

Unless we were going to get rid of the concept of levels entirely and instead give judges a range of +5 instead of +3 positive points to reward difficulty as well as quality. But that makes it even more subjective and unpredictable for the skaters. It's easier to get technical panels on the same page about what does or doesn't count than to get judges to agree on what's difficult enough to reward. So if we went that route, expect to see wider ranges of GOEs for the same elements.

As far as opening up the freedom of free program, I suggest allowing a fourth spin (in place of an eighth jump pass) -- it would have to be another spin in one position, with or without flying entry or change of foot -- in a different basic position from the other spin in one position. There are already limits on how many times the same feature could count in a program, so that would still apply and skaters who had used up all the features they can do but could still get good points with a good simple spin would plan one of a lower level.

[Susan M]

But if you want to make further differences between junior and senior, please elaborate on what you should think be required or allowed at one level and not at the other.

Sorry, I thought I did. The change that I suggested apply to seniors only was the elimination of single rotation jumps as scoring jump elements to allow them to be used as transition moves. I used the wording single rotation because I did not mean to take away points for the single axel, as we do still see a surprising number of those esp near the ends of programs when the skater is too tired for the double. Unfortunately, that also eliminates all those cool single axel variations like the delayed axel and tuck axel. Maybe there could be away of letting the skater identify those in advance as non-scoring moves so the caller doesn't put them in a jump slot.

[gkelly]

Walleys are already counted as non-listed jumps and can be freely used as transitions. It's the split-flip, split-lutz, and other in-air variations of singles from regular takeoffs that are the problem.

Juniors are mostly doing double and triple jumps -- they don't do singles on purpose except axels if they struggle with the double. Of course, as you note, pops and singled jumps from late-program fatigue are always possible at either level.

So I don't see any reason to allow seniors to do single jump variations for free and not allow the same option for juniors.

The only difference is that seniors have the choreo sequence now, and that seems like a good place to include those kinds of jumps. Already they can do it without penalty if the choreo sequence comes after the last jump pass.

Even so, some skaters -- even at levels where most of the jumps are doubles and not triples -- might want to use those special single jumps as transition moves between elements or heading into a jump or spin, not as part of a sequence, so another solution would be needed to avoid counting them against allowed jump slots.

Or else allow them to gain enough points that it's worth using up a jump slot.

[gkelly]

Steps/sequences

This is the trickiest kind of element, in part because all the skills that can be included in the sequence can also be sprinkled throughout the program with less than a full length or width of the rink at a time. So what is the value of requiring a discrete sequence or two with base values and GOEs?

On the one hand, skills like edges, turns, and steps are the very basics of figure skating. I expect junior and senior skaters to have at least a basic mastery of all these skills.

I want a point-based scoring system to be designed to give more points to skaters who demonstrate higher difficulty and/or quality in these skills than to those with weaker technique or who choose to avoid the more difficult skills in order to conserve their speed, energy, and concentration for the jumps and spins.

As a side note on the technical side of steps and turns, performing them in time with music is a demonstration of technical control and should be rewarded as such, aside from any considerations of artistry.

From a technical point of view, a skater at this level including only threes and mohawks in the good direction would be comparable to performing only the easier single jumps. And yet under 6.0 scoring it was possible for a skater to get away with such minimal basic skating content, if the jump content was strong enough, because there was no explicit mechanism to reward the actual skating content. So it is more objective, more sport-like, fairer to the skaters who make the effort to include the difficult content to quantify it in some way.

On the other hand, combinations of steps, turns, and field moves can be great opportunities for exploring the artistic possibilities of skating, with thematic body shapes and more detailed or nuanced musical interpretation than just staying on the beat.

Sometimes these functions are mutually exclusive -- sometimes the best way to tell a story or express a passage of music is with simple skills.

And it's hard even for devoted fans who are not also skaters, much less the general public, to appreciate the difference between a step sequence that's a technical tour de force performed well and one that's easy but just as well performed.

But it takes no special knowledge to recognize sequences or smaller moments within programs that are especially beautiful or musical or effective in telling a story and conveying a character. The fact that figure skating can do these things -- at its best, do them as well as any performing art -- is what sets it apart from most other sports and what attracts a large proportion of its audience, as well as many skaters themselves. So focusing too much on technical content in ways that make it difficult to include the artistic values as well is bad for the sport's popularity.

How can the rules meet both needs, rewarding both technical superiority and the transcendance of technique into artistic superiority?

The ISU has been grappling with this contradiction. The original rules that favored the technical side at the expense of artistic coherence, including certain kinds of technical prowess in the spiral sequence for ladies different from those in the step sequences. Then they deleted the second sequence from the junior free skate and a spin from the senior FS, and later the second sequence from short programs at both levels, ostensibly to free up more time for creativity between the elements. Most recently, the second sequence in the senior free program has been redefined to focus on choreography, with no levels or other explicit rewards for technical content. So that sequence is now freer than it used to be, but the requirement that it comes after the leveled step sequence in the free program is less free.

So we're not satisfied yet. What are some other approaches that could ensure that technical skating quality and content will be rewarded while not overly restricting artistic choreographic choices?

Looking just at the technical side a few years ago, and wanting to reward a greater variety of edge-based technical skills, I suggested introducing more options for what counts as features in these sequences and more options for different kinds of sequences that could earn difficulty points for different kinds of skills. Skaters might be required to perform one leveled step sequence for points and have the option to perform a second leveled step sequence OR a spiral sequence OR a leveled field moves sequence OR a leveled school figures variation, for example, in place of a fourth spin or eighth jump pass. Since one element slot could be filled by any kind of element, we would see more variety in program layout templates while still having clear definitions of how much each is worth and equivalencies between the different kinds of elements.

But I think we're even further now from that ever being a possibility. And it wouldn't encourage greater artistry per se, just greater variety in technical content.

Starting from the one leveled step sequence and one choreographed sequence we have now, a simple first step might be to add some field move-related features (or non/low-rotational jump features) to the options for gaining higher levels in the leveled sequence. And then have the tech panel call both sequences and have the computer select whichever sequence earned a higher level, or the first one if they both were called the same, to count as the leveled sequence and the other to count as the choreo sequence. That would allow skaters to choose to do their sequence for points that focuses on choreography early in the program if that fits their music, program theme, or aerobic pacing better than doing the more complicated steps first.

Another approach could be to decide that difficult steps will be explicitly quantified in the short program only. Redefine the levels in the short program step sequence so that skaters must demonstrate certain minimum skills to get credit for it at all and increase the differences in value between the levels so that a skater can build up a lead of several points by doing a level 4 vs. a level 1 step sequence in the SP, all else being equal.

But then get rid of leveled sequences in the free skate. Maybe require or allow one or two choreo sequences in the FS under the current rules for that element, and maybe also include "deep clean edges" and "difficulty" (with or without explicit definition) among the subjective bullet points for positive GOE.

Or don't even require sequences of steps or spirals/field moves at all as separate elements in the free program. Just encourage skaters to include those skills throughout the program by tightening up the criteria for the Skating Skills and Transitions components, and increasing the factors for those components. So even if the quality of your crossovers and forward stroking and a few threes and mohawks is world-class, you won't earn 5s or better in those components if that's all you do. And probably not in Choreography either.

[Susan M]

But what about the skaters whose bodies will never have the most beautiful positions and who don't naturally spin fast or center well?

There is no such thing as bodies not capable of attractive positions. Attractive is not synonymous with flexible. With work, any good skater should be able to attain a good camel position, a good sitspin, scratch spin, even a nice layback. We have seen attractive (and complicated) spins from skaters with bodies in various shapes and sizes. As for skaters who cannot do fast, well centered spins, they should get marked down. To me that's kind of like asking what about skaters who can't master triple jumps. The vast majority of young skaters reach the limits of their physical potential way before arriving at international senior competition. This is part of sport.