I think the judges have already applied somewhat of a "workaround" for this inconsistency by applying very generous positive GOE for any 4 Toe + 3 Toe combination completed and less generous positive GOE for any 3 lutz + 3 toe combination.
Also, note that the skaters who opt for the 3 lutz + 3 toe combination in the second half of the program are also benefiting from the 10% bonus.
I wonder at which point in time the ISU will readjust the multipliers on the component scores to adjust their weight relative to the technical scores. At some point, the judges may have no more room on the 10 point scale to balance the component scores with the technical scores equally, which would then require some adjustements (for example the ladies' multiplier to be adjusted from 0.8 to 1.0 for the short program and 1.6 to 2.0 for the free skate0.
I've always wanted them to put a 1.1 multiplier on the first jump of a combo and a 1.2 multiplier on the second jump. SO 3z-3r, 3t would score higher than 3z-3t, 3r which would score higher than 3t-3r, 3z which would score higher than 3r-3t, 3z.
There was a proposal in the last few years to use a multiplier, but it was rejected
"The team doesn't get automatic capacity because management is mad" -- Greg Smith, agile guy
I think that 1.1 multiplier for the combination as a whole would be enough. At least for starters to see how it plays out.
Ice dance technical committee should change twizzles in short dance every year. One year can be mirror twizzles (opposite direction), one year can be change of side twizzles (!?), one year can be face to face or back to back twizzles. (!?) We only have jump entry twizzles now. This way we can see more variations in twizzles. Skaters still can choose more comfortable twizzles in free dance.
IMO, the SP and the FS should focus on different things. Perhaps, the SP could focus on simplicity and perfection. This could involve a requirement that certain elements be level 1 and a much larger range of GOEs on jumps, so perfectly done but "easier" triples can score as well or better than more "difficult" jumps. The FS, OTOH, could focus on more technical difficulty -- with higher level elements, more difficult jumps being rewarded.
I also think that the penalties for falls need to increase for repeat falls (i.e., instead of merely 1 point per fall, it would be a 1 point deduction for the first fall, 3 for the second, etc.)
Finally, anonymous judging has got to go.
PCS needs to be revamped completely…like since 2006.
I think as of right now, almost every score is "justified" because it's arguable under some very vague definitions.
Also, I think the tech. panel has way too much power without anyone checking them. I know they're supposed to check each other, but I'd like something more.
"Corporation, n. An ingenious device for obtaining individual profit without individual responsibility." - Ambrose Bierce
I'd like to see all 7 jump types be mandatory in the LP. Skaters who don't have all triples could do doubles or singles as part of a combination or sequence without hurting their score much.
I would love to see all 7 jump types, even if you just throw in a 2loop in your 3 jump combination at least you showed a loop takeoff.
I also strongly agree with adding a multiplier to combinations but only to the 2nd jump. In theory, the first jump is the same as a solitary jump. Sure, mentally you know it has to be done perfectly to get the next jump off but there is nothing different about the entrance or exit just because it is being done in combination. The extra difficulty comes on the 2nd jump which has to come after a previous jump.
"Michelle would never be caught with sausage grease staining her Vera Wang." - rfisher
What 7 jump types?
Axel, salchow, loop, toe loop, flip, lutz . . .
What were you considering as the 7th?
Sorry for the double post but I got an error page the first time.
Professionalize the judging. The biggest problem is that the judges ALL have a conflict of interest if they are representing nations with people from them competing in that event. Every single one of them works for a particular national federation. Make it like professional sports, with referees and judges trained and paid for by the larger organization. As long as the judges have ANY kind of vested interest in the outcome, you will have the temptation to fix the scores.
Separate the technical and PCS judging panels. PCS already has specific bullet points that must be met. The PCS panel should simply evaluate if the skaters meet the specific bullet points under each criterion. Let the third panel of professional judges grade both PCS and technical elements with GOE.
Anonymity has proven to be problematic. Get rid of it.
I assume the poster just miscounted. Sometimes people have complained that the free skate isn't free enough anymore and is too much like a longer short program while others call for much more stringent requirements in the free skate. But it's interesting to hear and discuss differing ideas.
I'm not in principle completely opposed to a SMALL bonus for executing all of the different types of jumps, but wouldn't want it to be any more than about 2 points (for example if it came down to the last jump element and the skater had the choice between a 3S, which would complete the takeoffs, or a 3Lz, then maybe make it so they each would be equally valuable. The 3S because the skater showed more variety but the 3Lz still is more difficult. There is a 1.8 difference in base value there, though it would probably be in the 2nd half that that decision would be made so it would be closer to 2 points). Or it might come down to a 3T vs. a 3Lz if the skater had done loops in any previous combinations, which is a 1.9 difference in base value. That 2 point bonus would probably only make sense if we're talking about each jump being at least a triple (with the possible exception of the double axel), otherwise I think it would be too much. (If a skater hypothetically only did single jumps and it came down to a choice between a 1S and a 1Lz it would be silly to give a 2 point bonus for choosing the salchow over the lutz when the jumps are only worth 0.4-0.6 each and the difference in base value is only 0.2). But if there was a stipulation that there be no "e" calls then that would make those calls even more important and give the tech panel even more power. So there are potential drawbacks.
Sotnikova would have gotten this bonus (if a triple axel wasn't required for it at least for ladies) since she did all the other triple jumps as part of elements that received + GOEs and had no edge calls (though I thought the lutz did change edge after looking at the replay), so it would've increased her margin over Yu-Na Kim. Cue even more outrage and controversy. Also, one would need to figure out how jumps in combination were handled. For example, if Adelina's solo flip hadn't been clean or rotated, then would her 3F in the 3F+2T+2Lo have counted for the bonus? The element got a negative GOE but the 3F itself was good, it was the 2Lo that wasn't.
I've often wanted to test out different proposals, for example proposed multipliers for the first jump and second jump in combination, but so far have been too lazy. Some things might sound good in theory but they should be tested out first before being implemented. My hunch is that those multipliers would give intuitively too many points for combinations, especially for one with a fall or error on the second jump (which might only get -2 GOE for the combination). And if it was done in the second half of the program that would be another significant increase in base value. The base value would be inflated so much that the GOE wouldn't be as important unless other changes were made to the GOE structure.
One thing I've thought intuitively makes sense is having the GOEs be just percentage of the base value, rather than a set number added or subtracted. For different elements in singles and pairs SOV a -3 GOE can be worth anywhere from less than 5% of the base value (for a pivot figure) to 80% of the base value for a pair combination spin. And for very difficult 3-jump combinations (like 4Lz+4T+4Lo) it could be over 90%.
I think it makes sense that the percentage should be higher for jump combinations because in order to complete a creditable combination at least the first (and second in a 3-jump) jump of the combination would have to be completed to a decent enough standard that the combination wasn't broken, so there should be credit for that. But for solo jumps, spins, and step sequences (and pair elements other than the jump combination) I don't see a reason why a -3 on one type of element should be worth proportionally more than for another type of element.
And the dumbest thing of all in the scale of values is that they still haven't corrected the fact that a -2 or -3 GOE on a level 2 step sequence gets more points than a -2 or -3 GOE on a level 3 step sequence. A proportional type of system just makes much more sense to me.
Perhaps 40%/60%/80% for -3/-2/-1 and 120%/140%/160% for +1/+2/+3. For most types of elements a -3 currently gets more than 40% of the base value (not counting a possible fall deduction). I also know some argue that a fall should mean zero points, but I think that would have very serious unintended consequences and don't think that's a good idea, especially since there are gray areas in errors so having a HUGE point drop for any one GOE step possibly could result in dishonest judges manipulating the GOEs even more since even one step which might possibly be justifiable could have such a big impact on the score.
For combinations, the GOE percentage adjustments perhaps should apply only to the base value of the most difficult jump.
Last edited by RFOS; 03-10-2014 at 07:12 PM.
ITA with proportional GOE.
"The team doesn't get automatic capacity because management is mad" -- Greg Smith, agile guy
They have defined set deductions for edge calls and underrotations. Why not have a set deduction for falls, step-outs, two-foots, hand down and two hands down and then stack them when multiple errors are made, except in the case of a fall where the maximum deduction of say 80% is taken. For example, a step out is a 20% deduction and a step out and uderrotation would be calculated as basevalue*.8*.8 or an effective 36% deduction and so on. Then deductions are set and GEO isn't a factor on elements with obvious errors. Honestly I don't think that a jump with a fall should get anything above a -3 in the current system, but you see it all the time, -1 or -2 on these bad elements; the judging system allows for it because a judge can say well it had good height, clear entry steps and blah blah blah *insert other positive GEO bullet here*.
I also believe that positive GEO should be done away with completely in the short program and only negative GEO should be applied with a different set of criteria implying that only the most spectacular elements should recieve their full value. I think that would help skater really focus on quality for the short program.
Also, the whole compents mark should be scrapped. I think replacing it with a singular overall impression score on the 6.0 scale would suffice, except this time have a published criteria for what entails a 6.0 and include provisions for rewarding speed, transistions etc.. Then the 6.0 score issued by exery judge could be factored and all of them added together to make the new total score. Obviously the judges are not capable of fullfillinf the currrent requirements for components and are using them as placeholders anyway, why not just fix that problem all together.
Oh yeah, and anonymous judging needs to go, I think everyone egrees with this.
Last edited by Birdseye; 03-10-2014 at 11:55 PM. Reason: typo