Does COP reward difficult jumps adequately?

[Vash01]

IMO - No.

I am wondering if skaters that worked on quad jumps in the case of men, and triple axel in the case of ladies, should be awarded more points on their rare athletic talent?

Watching some of the recent competitions made me aware of some things-

1.Some difficult jumps may not be getting the number of points they deserve.

a triple toe-double loop-double toe combination has same value as a triple axel. A 3A is a very difficult jump and someone that can do it (particularly among ladies) should be rewarded handsomely for that feat. I don't think that is happening under the COP.

A quad toe-3t earns a skater just 3 more points than a triple lutz-triple toe. A fall on a quad is disastrous. So now we see a past quad king like Goebel go for an easier and safer 3lutz-3toe instead of a quad.

2. A 3-3 is a rare combination for ladies but they seem to earn a lot of points by doing a 3-2-2.

In the ladies field, a lot of triples are being downgraded to doubles so skaters choose to do a 3-2 instead of trying a more challenging 3-3 which is not likely to get the credit. Skaters attempting a 3-3 should be rewarded.

Is the COP discouraging skaters from challenging themselves by trying more difficult jumps?

3. OTOH mistakes on jumps don't seem to be penalized as much as they did under the 6.0. I am not even talking about the mandatory deductions in the SPs, but rather in the LP's.

While the 6.0 system overemphasized the quad jumps and jumps in general, at the cost of other elements, the COP seems to almost make jumps almost an insignificant part of a program. Multiple falls have the ability to ruin the overall look of a program. IMO 1.0 deduction on a fall (out of 100+ total score) seems very low.

I see a skater like Buttle make 4 mistakes on jumps in a program and still win over someone that had an imperfect quad and a lot of other landed jumps. Under 6.0 we would never have seen that. Buttle is a wonderful skater but he routinely messes up jumps and yet gets high scores. Those mistakes impact the overall quality of the program, no matter how great the foot work and the spins. There is no provision in the COP to look at the performance as one complete entity.

IMO the pendulum may have swung too far on the quads, and jumps in general.

Vash

[antmanb]

IMO - No.

I am wondering if skaters that worked on quad jumps in the case of men, and triple axel in the case of ladies, should be awarded more points on their rare athletic talent?

Watching some of the recent competitions made me aware of some things-

1.Some difficult jumps may not be getting the number of points they deserve.

a triple toe-double loop-double toe combination has same value as a triple axel. A 3A is a very difficult jump and someone that can do it (particularly among ladies) should be rewarded handsomely for that feat. I don't think that is happening under the COP.

While that may be true, doing a 3T/2T/2T uses up the three jump combination that a skater is allowed to do, further more, if this is ladies were hypothesizing about then doing a triple axel instead of the above combo leaves the skater another 6 jumping passes all of which can be the other five easier triple jumps with one repeated. She would also still have 2 two-jump and one three-jump combinations as posibilities. Compare that to the lady having done the 3T/2T/2T who would have to still do an axel type jump (presumably double in this case) and is then left with only 5 potential jumping passes of triple jumps and 2 two-jump combinations.

I think the advatage conferred on a lady by having a 3A in her program generally benefit the programs total base value in terms of jumps such that upping the actual point value of a 3A is to overvalue one element. A 3A is currently worth almost twice the value of 3T, is doing a 3A really worth twice a 3T? If you were to add points to the value of one element then do yuo not give an unfair advantage to someone who pulls off the one high risk element but only squeaks out the rest of her triples compared to someone who authoritatively lands all five of her triples but doesn't even attempt the 3A?

A quad toe-3t earns a skater just 3 more points than a triple lutz-triple toe. A fall on a quad is disastrous. So now we see a past quad king like Goebel go for an easier and safer 3lutz-3toe instead of a quad.

I don't see how a fall on a quad is any more disatrous than a fall on a triple, in fact i'd say a fall on a triple is more disastrous than a fall on a quad. If a skter falls then the judges will give it a -3 GOE across the board plus a 1 point deduction for a fall. So if you fall on a 3T you score 0 for the element, for a fall on a 3S you get 0.5 etc all the way to 3.5 for a fall on a 3A. If, however you fall on a 4T, you get 5 points...in other words you get more points for falling on a quad toeloop than if you had landed a clean adequate 3T or 3S and exactly the same points as you would have got for for a clean and adequate 3Loop or for a mildly flutzed 3Lutz...that to me is insane whay should a totally flawed qudruple toeloop (by virtue of a fall) give you the same points as a mildly flawed triple lutz?

2. A 3-3 is a rare combination for ladies but they seem to earn a lot of points by doing a 3-2-2.

But again a lady can only perform a 3/2/2 once so if she is using it to make up the advantage fo another skater having a 3/3 it doesn't relly pay off if both skaters max out the total jump combinations. If a skater has two reliable 3/3s and puts them both in the program together with a 3/2/2 (since point wise there's no point in risking a 3/3/2 when there are not extra points given for it, she might as well do two 3/3/ and a 3/2/2) then she's matched the 3/2/2 of the other skater plus upped the ante in the two other combinations. As with the 3A the points advantage conferred is about oening up other jump possibilities which adds to the over all base value.

In the ladies field, a lot of triples are being downgraded to doubles so skaters choose to do a 3-2 instead of trying a more challenging 3-3 which is not likely to get the credit. Skaters attempting a 3-3 should be rewarded.

Is the COP discouraging skaters from challenging themselves by trying more difficult jumps?

I've not seen the GP but looking at the protocols and hearing people talking about jumps being downgraded sys to me that everyone who could attempt them has been attempting them . I agree that the downgrading of triples to doubles is a double penalty which i seriously disagree with but i'm not sure its discouraging skaters from attempting them. I'd like to see them scrap the downgrading rule or make it more obvious e.g. downgrading a jump only if it is more than a half a turn short on the rotation. I also think sequences and combinations should be given multipliers to show that the jumps done together are mopre difficutl and should therefore be rewarded more then if the jumps are done in isolation. Off the top of my head i'd give jump sequences a multiplier of 0.1 and combinations a multiplier of 0.2. This would therefore encourage skaters to put their most difficult jumps in combination to get extra credit for them. Also if the skater doesn't even attempt the second jump then no combo or seuqence bonus can be given and zayak violations should apply if relevant.

3. OTOH mistakes on jumps don't seem to be penalized as much as they did under the 6.0. I am not even talking about the mandatory deductions in the SPs, but rather in the LP's.

While the 6.0 system overemphasized the quad jumps and jumps in general, at the cost of other elements, the COP seems to almost make jumps almost an insignificant part of a program. Multiple falls have the ability to ruin the overall look of a program. IMO 1.0 deduction on a fall (out of 100+ total score) seems very low.

I still think the SP and LP aren't weighted properly and think its ridiculous that the free program is no longer "free". I've already written above how ridiculous the fall deductions are when it gives a falied quad the same points a amildly flawed triple lutz. I think there should be mandatory deductions in teh SP like there always were and in a way that hurts the skaters like the SP used to. How about taking the 0.X mandatory deductions from the 6.0 system and turning them into X point decudtion under COP. So if you fall on a jump and therefore didn't complete that required element you get a 4 point deduction. If you totally miss out an element 5 point deduction. Turns between the jumps on your combination 2 or 3 point deduction.

I see a skater like Buttle make 4 mistakes on jumps in a program and still win over someone that had an imperfect quad and a lot of other landed jumps. Under 6.0 we would never have seen that. Buttle is a wonderful skater but he routinely messes up jumps and yet gets high scores. Those mistakes impact the overall quality of the program, no matter how great the foot work and the spins. There is no provision in the COP to look at the performance as one complete entity.

IMO the pendulum may have swung too far on the quads, and jumps in general.

Vash

Well you see this is where i'm starting to alugh a little bit at all of us as fans of skating. Under 6.0 all i would hear is people complaining that the sport is becoming all about the jumps, the jumps are all that matters, what about spins, choreo, artistry. Why are people getting credit and wining with flawed big jump attempts (eg quads) over people with great basic skating and choreo and decent triples. Now that we have a system that seemingly rewards the in between and it ofsets the advantage of the big jumps people are complaining.

I think the grass is always greener and there are always going to be people who think one think and other that think the exact opposite.

Another factor that is worth considering is that i think we are beginning to see the physical limits of what people can do. I don't think there will ever be a skater, no matter how much time, that will produce a program with a full set of quads...i just don't think that is humanly possible to train that without crippling yourself. That being the case it might just so happen that we're seeing the maximum jumps that poeple can do and not bothering to do any more at the same time that new system has come in and we don't know which one is to blame?

That was fun - thank for the debate!

Ant

[MikeNYC]

A VERY impressive and thought-provoking discussion.

Well done!

M.

[Doggygirl]

I too really enjoyed reading these points of view.

Thanks!
DG

[Mathman]

Great analysis, Vash and Anthony!

I agree with Ant's main point -- that the big jumps are rewarded adequately because they are rewarded twice. Yes, a triple-triple-triple might get you as many points as a quad, but in the latter case now you are in big trouble Zayak-wise.

Let's say you start out with

4T+3T
3A+3Lo
4S
3A

You've got 42.5 points before your competitor (who doesn't have a quad or a triple Axel) even gets off the ground. And you can still do as many triples as you want in your remaining four jumping passes.

For the ladies, if you don't have a triple Axel or a triple-triple, you really have to scramble to get in 7 triples. Here is Sasha at Eric Bompard:

3Lz+2T+2Lo 8.8
3F+2T 6.8
3Lo 5.0
3F 5.5
3T 4.0
2A+3S (seq) 6.0
3S 4.5

Total base marks 40.6

Other than changing her last Salchow to a second Lutz, there is not much more she can do.

In contrast, here's what Mao Asada could have got in the same event, if she had upgraded her 3F-2Lo to a triple-triple. (I think she planned a 3F-3Lo as her second element, but could not complete the second jump (?).)

3A 7.5
3F 5.5
3 Lz 6.0
2A 3.3
3Lo+2Lo+2Lo 8.0
3F+3Lo 10.5
3Lz+2Lo 8.0

Total base marks 48.8

She could change that 2A to a 3S and have an 8 triple performance!

Mathman

[emma]

Very interesting discussion, thanks for all these very informative posts!

[antmanb]

Great analysis, Vash and Anthony!

I agree with Ant's main point -- that the big jumps are rewarded adequately because they are rewarded twice. Yes, a triple-triple-triple might get you as many points as a quad, but in the latter case now you are in big trouble Zayak-wise.

Let's say you start out with

4T+3T
3A+3Lo
4S
3A

You've got 42.5 points before your competitor (who doesn't have a quad or a triple Axel) even gets off the ground. And you can still do as many triples as you want in your remaining four jumping passes.

For the ladies, if you don't have a triple Axel or a triple-triple, you really have to scramble to get in 7 triples. Here is Sasha at Eric Bompard:

3Lz+2T+2Lo 8.8
3F+2T 6.8
3Lo 5.0
3F 5.5
3T 4.0
2A+3S (seq) 6.0
3S 4.5

Total base marks 40.6

Other than changing her last Salchow to a second Lutz, there is not much more she can do.

In contrast, here's what Mao Asada could have got in the same event, if she had upgraded her 3F-2Lo to a triple-triple. (I think she planned a 3F-3Lo as her second element, but could not complete the second jump (?).)

3A 7.5
3F 5.5
3 Lz 6.0
2A 3.3
3Lo+2Lo+2Lo 8.0
3F+3Lo 10.5
3Lz+2Lo 8.0

Total base marks 48.8

She could change that 2A to a 3S and have an 8 triple performance!

Mathman

That's a hefty 8.2 advantage on the jumps for Mao which is more than the point value of a 3A but just less than the point value for a quad toeloop. I think i've read that Mao has problems with the triple salchow so doesn't put it in her programs and she has a tendancy to toe-axel her toeloop so rather than get a downgraded jump misses it out all together. Its a good strategy given the jump point advantage she's gets on the others!

Ant

[Joesitz]

Thanks Vash for the thread and Antmanb for the discourse. You guys are getting me worked up to get into the Values. I have been laxed on that aspect of the sport.

In Mathman's post on the scores of Sasha and Mao at TEB it would appear that had Mao left out the 3A, she would still have won the technical.

So for me at the moment (I will more than likely change my mind later), one would have to use many forms of strategy for a plan A and for a plan B at the time of skating and maybe a Plan C during a skate, if things are not going easily.

Oh yes, Vash's point on some jumps should have a higher value.
I am not convinced that the points assigned are the best right now but how would one determine other values? I was a skater who could do a lutz easier than a flip. I've notice a number of present day skaters who lip the flip because the lutz is easier. So a 3z for some is easier than a 3f for others.

Joe

[Mathman]

Just to emphasize the advantage a skater has if he can land a clean quad, here is a fictitious example.

Tommy Two-note never progressed beyond the toeloop and the Salchow, and he never mastered a triple-triple of any kind. Here is his program:

4T+2T+2T 12.0
4T 9.0
3T+2T (seq) 4.4
4S+2T 11.0
4S 9.5
3S 4.5
2S 2.0

Well, that's seven jumping passes, and the rules require an Axel, so Tommy tries a single Axel, pops it into a half-Axel, falls on his head, takes a negative 3 GOE (factored as -.5 for this element) and a fall deduction.

1A 0.8 -.5 -1.0 = -.7

Total 51.7

And here is the performance with which world silver medalist Jeffrey Buttle won Trophee Eric Bompard, including GOE, deductions and second-half bonuses*.

4T (fall) 9.0 -3.0 -1.0 = 4.0
3A+2T (seq) 7.0 -2.0 = 5.0
3Lz+2T+2Lo 8.8* +.8 = 9.6
1A 0.9* + .1 = 1.0 (unlike poor Tommy, Jeff got positive GOE on his single Axel)
3Lo 5.5*
3Lz 6.6* +.6 = 7.2
3F 6.1* -1.4 = 4.7
3S+2T 6.4*

Total 43.4.

Tommy rules!

MM

[antmanb]

Just to emphasize the advantage a skater has if he can land a clean quad, here is a fictitious example.

Tommy Two-note never progressed beyond the toeloop and the Salchow, and he never mastered a triple-triple of any kind. Here is his program:

4T+2T+2T 12.0
4T 9.0
3T+2T (seq) 4.4
4S+2T 11.0
4S 9.5
3S 4.5
2S 2.0

Well, that's seven jumping passes, and the rules require an Axel, so Tommy tries a single Axel, pops it into a half-Axel, falls on his head, takes a negative 3 GOE (factored as -.5 for this element) and a fall deduction.

1A 0.8 -.5 -1.0 = -.7

Total 51.7

And here is the performance with which world silver medalist Jeffrey Buttle won Trophee Eric Bompard, including GOE, deductions and second-half bonuses*.

4T (fall) 9.0 -3.0 -1.0 = 4.0
3A+2T (seq) 7.0 -2.0 = 5.0
3Lz+2T+2Lo 8.8* +.8 = 9.6
1A 0.9* + .1 = 1.0 (unlike poor Tommy, Jeff got poritive GOE on his single Axel)
3Lo 5.5*
3Lz 6.6* +.6 = 7.2
3F 6.1* -1.4 = 4.7
3S+2T 6.4*

Total 43.4.

Tommy rules!

MM

Except that Tommy didn't attempt an Axel type jump so he loses two points for his final 2Sal!!!! He'd still win though!!

Ant

[Mathman]

Except that Tommy didn't attempt an Axel type jump so he loses two points for his final 2Sal!!!! He'd still win though!!

Ant

I forgot about that.

OK, I edited Tommy's performance to include an Axel attempt, LOL.

MM

[antmanb]

Just thinking back to my other wishes for CoP...i read somewhere else someone suggesting that maybe a bonus should be given for a full set of triples that way people are encouraged to try all of the different types of jumps to get their rewards just as skaters in the past have had to attempt their nemesis jumps e.g. Yamaguchi and the 3Sal and Kwan and the 3Loop.

Ant

[alain707]

All the posts above make me want to ask : is there a revision of CoP planned at the end of the season ? It would be badly needed.
Leaving out the component scores for the moment, the technical score should reflect the technical content of a program and IMO fails to do so. I guess one reason is that the ISU came up too quickly with a point system after SLC and tried to make it not too complicated. Not to say that it is easy as it is now, but figure skating is a complex sports, with many different moves, and trying to assess the content of a program by a single score is no easy task. Well the problem may be that a not-too-complicated system is likely to have flaws.

1) Base values (BV) of the jumps: they have been rounded (from 4.0 to 6.0 for the triples except the triple axel, with 0.5pt increment, 8.0 to 12.0 for the quads with 1.0pt increment). That make them easier to remember, but do those values accurately reflect the difficulty of the element? I doubt it.

2) Grade of executions : -3 to +3 for triples and quadruples. They don't depend on the base value. A propotionnal system would be better, for instance -60% to +60% increment on the BV.

3) Combos: add the base values of the 2 or 3 jumps together. And what do you get ? A very rough underestimation of the difficulty of the element ! The second (and third) jump should be increased according to the first jump base value. For example add 20% of the first jump's BV to the second jump. So that a program with a 3Z-3T and individual 3T has a higher BV than 3Z and 3T-3T. Similarly, no deduction should be made when jumps are done in sequence. A slight majoration as for combos would be more appropriate.

4) Underrotated jumps : a slightly more than a 1/4 underrotated triple is not a missed double!! And a 1/4 turn underrotation is not like a 1/2 turn. There should be a at least 2 grades for underrotation : between 1/4 and 1/2 and more than 1/2. The downgrading to a one-revolution less jump would only occur with the second one.

I don't see the need to make the system not-too-complicated as anyway computers will add the scores!! Rather make a system that sums up what was actually done on the ice in a fair way, and respecting the common knowledge of technical difficulties of the figure skating elements.

[Mathman]

About point #2, the NJS does pro-rate the GOEs to a certain extent. For a double Axel, for instance, the possible GOEs are +3, +2, +1, 0, -0.7, -1.4, and -2.1.

For a double loop it's +1.5, +1, +0.5, 0, -0.3, -0.6, -1.

MM

[Vash01]

Going back to some points in Antman's first post-

Some skaters did stop attempting more difficult jumps because the COP was not rewarding them adequately.

Shizuka used to do 3sal-3toe, and 3lutz-3toe but she has started doing just 3-2 because her second triple gets downgraded. I have no problem with a skater not getting full credit for an underrotated jump, but giving credit for a double is like giving zero credit. IMO the skater needs to receive at least partial credit for attempting the difficult combination. Otherwise skaters may be discouraged from pushing themselves athletically.

The quad-triple gets more points than a (e.g.) 3Lz-3t. However, the quad is a far riskier jump than a 3Lutz. Even the ladies land it easily. When a skater like Tim Goebel goes for a 3L-3t instead of a quad, it tells me that he feels the risk is not worth the reward. For skaters to want to challenge themselves the reward has to be more attractive, otherwise it is not a high risk-high reward type situation. Under the 6.0 skaters that could land a quad were in a higher echelon. In the COP a skater that does not even land triple jumps consistently can beat one that land a quad. I agree there is more to skating than quads and triples (or 3-3 or 3A in the ladies), but these are highly risky, athletic moves. Since FS is a sport, those should be rewarded. By the same logic, mistakes need to be penalized more. IMO the penalty for underrotating a jump is heavy, but I feel that other penalties on jumps are not that heavy (fall, step out, two-footed landings, and of course flutzing/lipping).

I do, however, see Antman's point about a quad vs. a 3-jump combination and using up a jumping pass. Thanks for the numbers, MM and Antman. I am going to sit down and get into the numbers game too. Thanks for all your responses. It shows how much strategy goes into creating the programs under the NJS.

Vash