Tuesday, November 30

The 'Patrick Chan System': How Much is the New Scoring Method Really Helping?

We all should know the story by now. Following last season, there were plenty of figure skating fans who felt that risking difficult jumps (Quads for the men and Asada's triple Axel, for example) were too much of a risk, as skaters would often be down-graded for their efforts and they'd end up losing even more points than doing a simple triple toe or double Axel would earn them. So, what did the ISU do? They came up with a re-worked Code of Points (or IJS), one that increased the value of the more difficult jumps, and also decreased the negative grade of execution scores, so the penalty for taking risks wouldn't be as harsh. They also implemented an under-rotation penalty, one that docked some points from the base value of the jumps-- but not nearly the severeness of the full downgrade that they'd previously incur.

Everyone that was complaining before about skating taking steps backwards last year should now be happy, right? Not exactly. We have Patrick Chan falling all over the place in his Grand Prix events and still managing to score among the top skaters. We also have the "Mao Asada rule" (as many seemed to think the triple Axel changes were only to benefit her) not even playing a part yet this year as Asada has had extreme difficulties while changing her jumping technique.

Back to Chan. I thought it would be interesting to compare his scores from Skate Canada this season under the system used last year, to see if he really is getting plenty of help from the re-worked system.

First, the short program from Skate Canada. Remember, this year there is one less element in the program, as a second step sequence has been dropped. Another aspect to remember is that the GOE for the triples and quads have changed a bit. The scores have been factored to reflect the old system (for example, a +1 on a 3toe now earns 0.7 GOE, compared to 1.0 last season). Let's see how he scores.

The elements are not in table form (sorry), but they are listed by their code, their base value, their GOE, and then the total points earned for the element.

4T 9.80 -4.80 GOE = 5.00 (fall)
3A 8.20 -4.20 GOE = 4.00 (fall)
CCoSp4 3.50 +1.00 GOE = 4.50
3F+3T 9.50 +1.57 GOE = 11.07
FSSp4 3.00 +0.86 GOE = 3.86
CCSp3 2.60 +0.57 GOE = 3.17
SlSt2 2.30 -0.60 GOE = 1.70 (fall)

Total Element Score = 33.30 (compared to his actual score, 36.73)
Program Component Score = 39.47
Deductions = -3.00

Total Segment Score = 69.77 (compared to his actual score, 73.20)

Now, for the free skate. Two more changes in the system are reflected and factored here. The 1/2 loop in between two jumps is now scored as a single loop, and counted as part of a skaters three-jump combination. Last year, the 1/2 loop was not worth any credit, and the total jump sequence was started at 80% of its base value points. Second, the second footwork sequence in the free skate is now scored on a fixed start value (2.0). Last year, Chan received a level 4 for his second footwork sequence at Worlds (with the same program), so I have used the GOE scored there as a reference here.

4T 9.80 +2.00 GOE = 11.80
3A 8.20 -4.20 GOE = 4.00 (fall)
3Lz 6.00 +0.43 GOE = 6.43
SlSt3 3.30 +1.00 GOE = 4.30
FSSp4 3.00 +0.79 GOE = 3.79
3A+2T* 10.45 +0.29 GOE = 10.74
3Lz+2S+Seq* 6.42 +0.43 GOE = 6.85
CSSp3 2.60 +0.57 GOE = 3.17
3F* 6.05 +0.43 GOE = 6.48
3Lo+3T<* 6.93 -1.00 GOE = 5.93
2A 3.85 +1.00 GOE = 4.85
SlSt4 3.90 +1.80 GOE = 5.70
CCoSp4 3.50 +0.79 GOE = 4.29

Total Element Score = 78.33 (compared to his actual score, 83.18)
Program Component Score = 84.14
Deductions = -1.00

Total Segment Score = 161.47 (compared to his actual score, 166.32)

Total Competition Score = 231.24 (compared to his actual score, 239.52)

A difference of 8.28 points.

In all honesty, with all of the smaller scoring changes (the way the sequences are scored, etc.), I don't think that 8 points is really a huge number. Most people were probably thinking he would score about 15 or more points lower than he did, yes?

So, what do you think now? Was the old system better, or does the new system work just fine? Those of you that feel Chan was over-scored for what he did complete successfully at Skate Canada, what would you do to make the scoring more accurate for what we see?

And for those of you that thought skating was going "backwards", is the new method of scoring what you wanted to see?

Obviously, it would be nice to see what other skaters would score (Oda and Rippon at Skate Canada) under the old system for direct comparison-- that might come eventually as a part two to this analysis.


Buck said...

Its hard to like anything that lets someone with multiple falls win!


Anonymous said...

I mean, okay, we know how great his skating skills and the programs themselves are, and I agree with the judges that his ability to showcase them despite all the falls SHOULD be reflected in the scores. BUT REALLY, there HAS to be a better way to reflect NEGATIVE impressions caused by those multiple falls as well.
How about revising the deduction rules so that it'll be minus 1.0 for the first fall, minus 2.0 for the second fall, minus 3.0 for the third fall, as so on? That'll be a 6.0 points deduction for falling 3 times.

Anonymous said...

Well, I won't even pretend to understand the scoring system...

But it seems to me that nobody else has been allowed to fall with impunity - or to put it in a simple question:

Would any other skater - in any discipline - be able to win a competion even though falling four times?


Lindsay said...

I kind of disagree that 8 points isn't that big of a deal. At Skate Canada, 8 points was the difference between a first place finish and a third place finish. I understand that this is sort of a false comparison, since Oda's & Rippon's scores need to be re-converted, but I still think 8 points in major competitions is a Big Deal.

I have to confess, though, I expected the numbers to be MUCH higher. Chan is an extreme example (since no one else is falling nearly as much as he is), which is comforting! For me, anyway, as I had a lot of reservations about this new rule. It isn't going to prop athletes who aren't already being propped.

Anonymous said...

Thank you for your analysis, Tony. :) It's helped me clarify my thoughts on a lot of the judging this season.

In my opinion, falls should be more heavily penalized and that skaters shouldn't be 'rewarded' with lower punishments for trying and failing difficult elements. A fall is a fall is a fall. No matter how you swing it, the skater failed to land the jump. It's awful to watch, it disrupts the program, and there's no way a skater should be able to get away with medaling after four falls unless the rest of the competition doesn't do any better. It's agonizing to see a skater hit the ice multiple times during a program and still score higher than a clean or nearly clean skate containing easier elements. A step-out or a hand down is one thing, being a human Zamboni is another.

I also sometimes struggle with the way PCS are awarded. As far as I'm concerned, 'Skating Skills' ought to include whether or not the skater is capable of staying on their feet. Beautiful stroking, ice coverage, and grace on the ice are all important - but the system seems flawed if it doesn't detract points for falls. It also seems to me that sometimes the difference in PCS from skater to skater ought to be much smaller than they are (Rippon and Chan at SCI, for example). When there's a vast, inexplicable disparity in the scores that two similarly-skilled skaters receive, it really starts to look bad for the judging system. And it takes away from the joy of watching the sport.

muzicgirl01 said...

Thanks for the analysis! Its really interesting to think about and makes you wonder what will happen at the Final and at the Canadian Nationals and Worlds. I'll be interested to see what he scores and how many more times he falls. Though he could surprise us all and stay on his feet...

I watched the Russian free skate this weekend and I had to laugh when they showed Chan's score and the commentator said (and I'm paraphrasing since this is the one NBC airing I didn't DVR so I can't get the exact) "That's a pretty generous score with those falls." and "I'm having a hard time understanding where that component mark comes from with 3 falls."

physics girl said...

I don't know too much about the scoring system, so thanks for this interesting post. I just want to say something about falling.

A fall in itself only causes a one point deduction. So falling alone is not necessarily enough deductions to lose a competition, if the next competitor is far enough behind. Also falling on a jump is different from falling while skating along. It's more jump technique than skating skills.

Falling alone is not necessarily an indicator of a good skate. A planned triple that is popped to a single that is landed fine should not get as many points as an attempted triple that ends in a fall. Anybody can do single jumps and stay upright for four minutes, but it isn't really what we want to see at this level is it? I find popped jumps more disrupting than falls, personally.

Perhaps as someone suggested the falls should be penalized even more. But if it gets to the point where clean double jumps are awarded more points than attempted triples with a fall, more skaters may elect to play it safe, and it will be more boring to watch.

I'm not taking any sides on the Chan debate or comparing his score to the others. I'm just saying that there is so much more to skating than whether you fall or not. If you don't like to see falls, watch more ice dance :) The fact that they fall sometimes too indicates just how hard what they are doing really is!

Obviously I would like to see more cleanly executed difficult jumps, and I think that's what this is all about. Unfortunately, it seems to be the nature of the sport that they are actually so difficult that even elite champions fail to land them sometimes.

Anonymous said...

I think it just goes to show that a skater as good as Patrick will score high whatever the system is.

sabine said...

Why you not analyze Chan's PCS scores?
... If it is possible at all, in this, by me - a random, and "what I like" the ISU rules?

It would be much more interesting ...

Tony said...

I don't think his components were THAT far out of line. And for purposes of this study, it's not like the criteria for the components have changed from last year to this year-- hence why I kept them the same.

Which components do you think he got gifted on? Performance/Execution? Etc. If so, what should he have received?

sabine said...

In general - Chan is good "executor" of choreography... but, by my opinion - the bad "performer".
May be the cause of his youth, but for now, he is very far in that from Daisuke Takahashi, Plushenko...

P.S. I forgot to write in my previous post:
"... the ISU = Lori Nichol's rules? ..." ;-)

Anonymous said...

Tony, IMO nobody should be getting 9's high 8s for P/E with the amount of mistakes Chan. AT most what I would have given him was a 7.5 for a program that IMO was a mess.

I think a skater gets so many minus GOE, it should reflect their P/E mark.