Results of removing jump pass in Men's FS

[Alchamei]

I wanted to post this question, but I didn't want to create a new thread so....

My question in relation to jumps is, Is it only limited to Toe loop jump where one can have 2 quads and 2 Triples like Nathan Chen did, or can the same be allowed with some other jump like Lutz? When he did those jumps, my idea of Zayak rule basically got blurred out quite a bit.

Basically you can repeat only two kinds of jumps. It doesn't matter if both of them are Lutzes or Toes, triple and a quad is a different jump.

 

[zebobes]

Basically you can repeat only two kinds of jumps. It doesn't matter if both of them are Lutzes or Toes, triple and a quad is a different jump.

To be more precise, you can repeat only two kinds of triples OR quads. All double jumps can be repeated once.

Now the question is, what happens with quints? Will the ISU make a rule about those?

 

[gkelly]

I'm guessing they won't put quintuple jumps into the Scale of Values until 1) quad axels start getting rotated and landed in competition and/or 2) one or more skaters start landing rotated quints in practice, with video evidence, and lobby the ISU to allow them.

If and when quint jumps are incorporated into the sport, I'm sure they will be included in the Zayak rule the same way triples and quads are: only two different jumps with 3 or more revolutions can be repeated; same takeoff and different number of revolutions = different jumps.

 

[Tavi...]

I'm guessing they won't put quintuple jumps into the Scale of Values until 1) quad axels start getting rotated and landed in competition and/or 2) one or more skaters start landing rotated quints in practice, with video evidence, and lobby the ISU to allow them.

If and when quint jumps are incorporated into the sport, I'm sure they will be included in the Zayak rule the same way triples and quads are: only two different jumps with 3 or more revolutions can be repeated; same takeoff and different number of revolutions = different jumps.

Do you think anyone will ever land a quint? It seems like at some point we'll reach a physical limit.

 

[gkelly]

I think cheated ones have been done in practice -- maybe not landed on one foot either, definitely not fully rotated.

As long as downgrade rules remove all incentive for attempting them in competition, there's no rush to train them.

As with women and quads, I think we'll need to see advances in skate technology (which are more likely than necessary differences in gravity or human biology, or even in ice technology given the need to share the ice with hockey etc.) that allow skaters to jump higher before more revolutions become realistic.

 

[matscol]

A quick run of numbers follows, should be accurate but if I see any mistakes I will update and fix later when I have a little more time. These numbers are the average of the GOE score, not the mean calculated judge's GOE (meaning, a fall on a quad is (-4) in the calculations - the actual points given. It is not calculated as (-3) - the judge's GOE). If anyone is interested in the scores as judge's GOE let me know, I can post those too.

Average GOE of jump elements, all competitors, U.S. Regionals October, 2016 -
Juvenile: -0.207
Intermediate: -0.239
Novice: -0.317
Junior: -0.309
Senior: -0.412

Average GOE of jump elements, podium finishers (places 1-4) U.S. Regionals 2016 -
Juvenile: -0.081
Intermediate: -0.148
Novice: -0.238
Junior: -0.261
Senior: -0.328

Average GOE of jump elements, junior grand prix 2016 (not including jgp final) -
All competitors: -0.365
Podium finishers (1-3): +0.264

Average GOE of jump elements, senior grand prix 2016 (including gpf) -
All competitors: -0.043
Poduim finishers (1-3): +0.432

 

[Mrs. P]

Cool stuff matscol.

I wonder if you would be willing to do like the average of "final group." in your regionals/jgp/senior grand prix? Also it might be good to include the three U.S. sectionals as well. I think that will help show the range.

 

[jace93]

If you could also post the judge's goe I'd be really grateful... While these numbers are already very interesting I'd say the non-factorized ones would be more useful in understanding the actual average vote given to the elements' execution (ie novice may have a slightly higher average goe than seniors just because they obviously attempt easier elements on whom the goe range is smaller, instead of actually getting higher votes from the judge's)

 

[matscol]

Will do some messing with it tonight; also again would like to caution that was quick and not double-checked. Might be slightly off, I will make adjustments if need be. But I do think it shows the general idea. Also I think it shows in the U.S. that GOE's don't necessarily get a whole lot better at the top of the lower levels (it's just that the top kids tend to be doing much more diffficult jumps).

 

[gkelly]

Thanks for all that work, matscol.

I think the judges' +3 to -3 scores would be more useful than the factored numbers, because the sizes of the increments are bigger for the skaters who are attempting more difficult jumps, and therefore GOE on a couple of failed difficult jumps might outweigh the positive GOE on the easier jumps.

It's useful to keep in mind that juvenile podium finishers, for the most part, are including double axels; average juveniles are not.
Podium finishers at higher levels are likely to include triples (although not many at intermediate, especially in smaller regions where triples might not be necessary to make top 4) -- but not average intermediate men, or average ladies at all levels even up to senior where they're required in the short program.

But the skaters who are racking up negative GOE on triples may be earning positive GOEs on many of their doubles, especially the podium finishers.

Although it would be more work and I wouldn't ask you to put in the extra time, it might be useful to leave out jumps with all -3 GOEs or maybe all -2 or worse. Anyone can have a fall or step out on any jump, regardless of the quality of their successful jumps. And what we're really looking for is the average quality of the landed jumps, right?

Of course, we can't tell just from the protocols: An average juvenile, for example, might do a 2Lze<+2Lo< that will earn -3 from the judges, even while her parents are rejoicing in the stands that she landed the combo. So we probably would want to include that as an example of average jump quality at juvenile level.

 

[matscol]

Tried to incorporate some of the ideas mentioned above. The ones that are a little hard for the time I have right now are 'final group' (5 skaters? 6? would have to program in the rules to figure that out, and even then might be wrong and end up including skater 10; quite tricky little problem) and figuring out which -3's are or aren't falls.

GOE numbers are calculated by discarding highest and lowest judges' GOE marks and averaging the rest for each element.

Column meanings:
"All, all" = all skaters, all jumps
"All, top n" = top n skaters *in that program* (not in overall event results), all jumps
">-3, all" = all skaters, all jumps with better than -3 GOE
">-3, top n" = top n skaters in that program, all jumps with better than -3 GOE

	Short Program​	Free Skate​
	All, All	All, top 4	>-3, All	>-3, top 4	All, All	All, top 4	>-3, All	>-3, top 4
US Reg, Juv Boys	N/A	N/A	N/A	N/A	-0.927​	-0.669​	-0.644​	-0.391​
US Reg, Int Men	-0.810​	-0.417​	-0.285​	+0.019​	-0.845​	-0.624​	-0.421​	-0.243​
US Reg, Nov Men	-0.909​	-0.551​	-0.084​	+0.162​	-0.781​	-0.620​	-0.324​	-0.184​
US Reg, Jr Men	-0.748​	-0.748​	-0.161​	-0.161​	-0.658​	-0.658​	-0.427​	-0.427​
US Reg, Sr Men	-1.095​	-1.095​	-0.007​	-0.007​	-0.553​	-0.553​	-0.100​	-0.100​
US Sec, Juv Boys	N/A	N/A	N/A	N/A	-0.824​	-0.247​	-0.452​	-0.102​
US Sec, Int Men	-0.706​	-0.407​	-0.388​	+0.025​	-0.750​	-0.535​	-0.387​	-0.137​
US Sec, Nov Men	-0.711​	-0.201​	-0.172​	+0.250​	-0.627​	-0.295​	-0.211​	+0.034​
US Sec, Jr Men	-0.897​	-0.471​	-0.266​	-0.226​	-0.337​	-0.061​	-0.106​	+0.137​
US Sec, Sr Men	-0.536​	-0.401​	+0.065​	+0.226​	-0.098​	+0.313​	+0.166​	+0.457​

	Short Program​	Free Skate​
	All, All	All, top 4	>-3, All	>-3, top 4	All, All	All, top 4	>-3, All	>-3, top 4
US Reg, Juv Girls	N/A	N/A	N/A	N/A	-0.788​	-0.147​	-0.422​	-0.004​
US Reg, Int Ladies	-0.535​	-0.016​	-0.108​	+0.223​	-0.844​	-0.326​	-0.468​	-0.070​
US Reg, Nov Ladies	-0.879​	-0.215​	-0.176​	+0.200​	-0.866​	-0.391​	-0.481​	-0.087​
US Reg, Jr Ladies	-0.988​	-0.283​	-0.337​	+0.195​	-0.850​	-0.504​	-0.420​	-0.063​
US Reg, Sr Ladies	-2.07​	-1.151​	-0.852​	-0.323​	-0.910​	-0.525​	-0.464​	-0.110​
US Sec, Juv Girls	N/A	N/A	N/A	N/A	-0.263​	+0.282​	+0.024​	+0.337​
US Sec, Int Ladies	-0.292​	-0.089​	+0.053​	+0.176​	-0.648​	-0.014​	-0.175​	+0.308​
US Sec, Nov Ladies	-0.550​	+0.092​	-0.070​	+0.373​	-0.700​	-0.183​	-0.220​	+0.073​
US Sec, Jr Ladies	-0.447​	+0.071​	+0.011​	+0.368​	-0.621​	-0.297​	-0.244​	+0.110​
US Sec, Sr Ladies	-1.291​	-0.200​	-0.367​	+0.120​	-0.607​	+0.231​	-0.205​	+0.354​

	Short Program​	Free Skate​
	All, All	All, top 3	>-3, All	>-3, top 3	All, All	All, top 3	>-3, All	>-3, top 3
JGP, Jr. Ladies	-0.860​	+0.612​	-0.246​	+0.771​	-0.780​	+0.600​	-0.398​	+0.688​
JGP, Jr. Men	-0.558​	+0.551​	-0.066​	+0.653​	-0.602​	+0.166​	-0.249​	+0.323​
GP, Sr. Ladies	+0.166​	+1.065​	+0.366​	+1.164​	-0.091​	+0.559​	+0.104​	+0.684​
GP, Sr. Men	-0.363​	+0.203​	+0.135​	+0.546​	+0.137​	+0.704​	+0.436​	+0.989​

 

[gkelly]

Wow, thanks for doing all that work.

Now we can see that the higher levels/better skaters at that level do seem to have higher GOEs on average, discounting completely failed jumps.

 

[matscol]

Wow, thanks for doing all that work.

Now we can see that the higher levels/better skaters at that level do seem to have higher GOEs on average, discounting completely failed jumps.

The numbers were easy. The *table* was hard.

The biggest surprise to me is the senior men getting better overall jump GOE's in the FS than in the SP. I went back and ran just that last line for all GP events from the past ~7 seasons and though the numbers tightened to the point that they were close to the same, the Sr. Men's FS number was still higher than the SP.

 

[gkelly]

Well, for many of the top senior men two-thirds or all three of the SP jump passes include jumps with more than 3 revolutions, whereas for the freeskate 4 or 5 of the jump passes may consist of non-axel triples.

At regional level, quads would be few but triple axels and triple-triple combos may be challenging in the SP and make up a smaller proportion of the FS jump passes.

 

[matscol]

Well, for many of the top senior men two-thirds or all three of the SP jump passes include jumps with more than 3 revolutions, whereas for the freeskate 4 or 5 of the jump passes may consist of non-axel triples.

At regional level, quads would be few but triple axels and triple-triple combos may be challenging in the SP and make up a smaller proportion of the FS jump passes.

Makes sense; I guess I just had in my head an interview I saw somewhere with Alex Ouriashev where he talked about how a skater could usually get away with a mistake or two in the FS but couldn't afford any and had to be perfect in the SP. I wasn't really thinking it through in terms of this new era of Men's competition that we seem to have entered in the last year.

 

[Tavi...]

I think cheated ones have been done in practice -- maybe not landed on one foot either, definitely not fully rotated.

As long as downgrade rules remove all incentive for attempting them in competition, there's no rush to train them.

As with women and quads, I think we'll need to see advances in skate technology (which are more likely than necessary differences in gravity or human biology, or even in ice technology given the need to share the ice with hockey etc.) that allow skaters to jump higher before more revolutions become realistic.

Interesting!

 

[Alchamei]

Personally, I dislike the change. Men are able, unlike women, to complete all kinds of triple jumps and when you add the quad, there is no space for the other ''easier'' triple jumps. I admire the current field for trying to put all the triples into place and with one pass less, it will be basically just a quad fest since Top Men are doing 3-4 quads right now.

Also, removing 30 seconds from the program is BS as well. There will be no time for choreo since lots of Men will spend half of their programs preparing for quads.

 

[4everchan]

to elaborate on gkelly's point,

in the Men's SP : it's really at this time in skating trends the risk it all program ... skaters will maximize BV here, using 1 or even 2 quads that may not even be steady, and the 3A... without these elements, they cannot get to the final flight... however in the LP : as gkelly mentions, these 3 big target elements (big combo, quad and 3axel) are only part of the program... Many skaters will gain positive GOE on the other triples...

For ladies : it's different at this point. The 2A is no big deal.. or shouldn't be.... The solo jump is right now for most, a stable jump, some opting for the loop instead of getting an edge call for instance... and the combo is usually your most comfy 3-3....

In the LP, ladies have to have a less secure strategy as without the inclusion of those iffy edge call triples or less comfortable ones, they will not have enough triples to compete... on top of that, the combos are not always comfortable for ladies compared to the one good one they opt for in the SP... for instance, the half loop combos are tricky and the 2A-3T is often not a great one to watch...

So, at this point, I'd say that ladies are under a bigger risk in the LP where they have to show a more complete arsenal, using stable jumps in the SP, while men risk more in the SP while the LP contains those easier jumps for them they wouldn't include in a SP...


However, removing a jumping pass for men will certainly change that...harder combos may be needed... and they may remove their easiest triple (let alone a 2A for those still using that jump) to favour the riskier element. I don't see anything changing much for the ladies until the 3A becomes more spread out. They all repeat the 2A so one less pass would simply mean one less 2A.... for most... unless some feel better removing one of the edge call triple

The numbers were easy. The *table* was hard.

The biggest surprise to me is the senior men getting better overall jump GOE's in the FS than in the SP. I went back and ran just that last line for all GP events from the past ~7 seasons and though the numbers tightened to the point that they were close to the same, the Sr. Men's FS number was still higher than the SP.