Just a thought on Johnny

[Mathman]

See if you can get a bit more for Ashley while you're at it.

On no! Ashley loses a point. She got 11.38 for Performance/Execution and it should have been 11.37.

 

[gsrossano]

For those interested, this is the USFSA document covering post event error correction.

http://www.usfigureskating.org/content/techpanelprotocol.pdf

Note on page 3, calculation errors must be identified and corrected prior to the awards ceremony. Once medals are awarded corrections to the result are not made.

 

[tjs]

:no:

You have problems.

You're looking at two separate apples and calling it an apple pie.

We might as well say every tax payer in the country is part of a "Team", since they all work and give their money to a shared beneficiary.

Guess What I think you have a problem. I make a great apple pie and every time I put 2 apples in it (well actually 6 to 8 apples) I come out with apple pie!! Dont get it do you???? Didnt think so..... hmmmmmmmmmmm

 

[emma]

For those interested, this is the USFSA document covering post event error correction.

http://www.usfigureskating.org/content/techpanelprotocol.pdf

Note on page 3, calculation errors must be identified and corrected prior to the awards ceremony. Once medals are awarded corrections to the result are not made.

So, any 'bad judging' aside, are you saying that according to the rules Weir won, but according to the same rules, he didn't win because his team didn't 'do the math' soon enough to protest?? That's actually so funny to me but I don't know why exactly.

 

[Mathman]

I predict that USFS will say that their interpretation about how and when to round is within the stated rules. The rules don't specifically say, DO NOT ROUND THE TRIMMED MEAN BEFORE APPLYING THE PROGRAM COMPONENT FACTOR. They just say, average, multiply, and round. I think they have enough wiggle room not to make any changes, if that's what they decide to do. (In fact, if anything, they will probably "clarify" the rules rather than change the method of calculation.)

It is not at all clear to me that one method rather than the other is better, or fairer, or superior -- as long as we all agree on what method is going to be used. As far as I know (correct me if I am wrong) the PCSs have always been calculated like this. It's just now that anyone noticed that something was wrong.

If Johnny had just put in that extra double loop we could all have gone peacefully to our graves in blissful ignorance of the fact that every skater who ever competed under the CoP had their scores calculated wrongly.

 

[emma]

If Johnny had just put in that extra double loop we could all have gone peacefully to our graves in blissful ignorance of the fact that every skater who ever competed under the CoP had their scores calculated wrongly.

Amen to this!!!! (love ya' Johnny!!!!! just do the darn doubles too!!!)

 

[jennylovskt]

Thanks to everyone for explaining the calculation and rounding. I normally don't go in too deep in numbers, but this time, I did, and It's rather interesting. Now I started to feel sorry for Evan. He didn't do anything wrong. It's the judges who gave the unfair judging and the rule of the calculation made the error which normally we could ignore.

I don't think we could change anything unless somebody could find that there was a conspiracy involved. Well, Johnny will be happy if he knows about it. For me, I can let it go, and move on to 4CC and the Worlds.

 

[gsrossano]

If Johnny had just put in that extra double loop we could all have gone peacefully to our graves in blissful ignorance of the fact that every skater who ever competed under the CoP had their scores calculated wrongly.

I intend to go blissfully peaceful to my grave even with that knowledge! It would also have been nice if he had put in a change of foot on the upright spin. Why would he put a spin in his program worth just a pathetic 1.8 points?

Nevertheless, one math guy to another, if I were to give you the recipe to calculate the area of a circle and did not include in the list everything you shouldn't do in the process, should I accept any answer you get because I also didn't also give you an endless list of everything you should not do? Gee teacher, you never said I couldn't use 22/7 for the value of pi!

But I agree with your prediction. The path of least resistance is to say the rule is unclear and to rewrite the rule.

 

[tilak]

I intend to go blissfully peaceful to my grave even with that knowledge! It would also have been nice if he had put in a change of foot on the upright spin. Why would he put a spin in his program worth just a pathetic 1.8 points?

In St. Paul's Pioneer Press, Jan. 28th newspaper, in one of the articles it states, " Weir said he was told where he lost the gold. It was during his upright spin in the middle of his program where his change of edge was not visible enough to the judges. Weir had planned an upright spin with a change of foot, which has a base value of 3.00, but he was credited with just an upright spin, which has a base value of 1.80."

I'm not sure what that means, really. If the judges had visibly seen the change of edge maybe he would have received more than 1.80 for the upright spin?

 

[GoldMedalist]

Ah, well, he left the change-of-foot out.

 

[emma]

I feel the need to just say - what an amazing competition - three cheers for Evan, Johnny and a host of other men who just really rose to the occasion and skated great.

I do hope all the powers that be do correct or make more precise their rules for the mathematical calculations - while ties may by infrequent, it seems like we might see many 'close comps' where those tenths and hundrenths of a point do 'add up' and make a difference. Cool that you figured that out, Grossano.

 

[Joesitz (RIP)]

MM - What is the probability of a tie in competition under CoP?

I would think it would be in the billions.

Joe

 

[gsrossano]

What I hope would come out of this is that USFSA pushes the ISU to correct they way the calculation method does rounding. Ultimately this is an ISU rule. It is unacceptable IMO that close results should be determined by sloppy calculations, when there is no reason not to do the calculation exactly. I am currently doing an analysis for Nationals to see how the marks change if the calculation is done exactly compared to the official results. Hopefully it will be done this weekend.

And to answer a question from Mathman from another thread -- the question being what are the results if the calculation is done with no rounding at all until you get to the total points in each segment -- Evan wins. So regardless of the ice-lawyering on the details of the rules, Evan's marks overall were slightly higher than Johnny's in the total event.

So, if you round according to the software Evan wins on a tie breaker, if you round according to the letter of the law, Johnny wins, and if you round in the most mathematically correct way, Evan wins.

 

[Mathman]

What I hope would come out of this is that USFSA pushes the ISU to correct they way the calculation method does rounding. Ultimately this is an ISU rule. It is unacceptable IMO that close results should be determined by sloppy calculations, when there is no reason not to do the calculation exactly. I am currently doing an analysis for Nationals to see how the marks change if the calculation is done exactly compared to the official results. Hopefully it will be done this weekend.

IIRC, the ISU faced a similar issue with the ten per cent bonus for jumps after the halfway mark. The first couple of years they rounded the score for the base value + bonus off to the nearest tenth, for each element separately. The rounding error could add up to quite a bit.

In that case, though, the solution was simple. Just go to hundredths and you didn't have to round at all.

About the wording of the rules, if the ISU does not want to address the issue, I suppose they can just say, we are going by the tacit convention of making all computations to the nearest hundredth. (However, this makes their scoring rules sound silly because there would be no rounding to do at the end.)

I look forward to the full analysis.

 

[ManyCairns]

If Johnny had just put in that extra double loop we could all have gone peacefully to our graves in blissful ignorance of the fact that every skater who ever competed under the CoP had their scores calculated wrongly.

Nah, if Johnny had put in the extra double or made the COE more clear/ changed foot, they'dve just held his PCS down more, or found something new to downgrade, or whatever. Johnny did more than enough to have taken it to the bank. It's kinda like blaming the victim, to me.

Obviously, I think Johnny was robbed big time, given Evan's performance _that night_.

 

[gsrossano]

IIRC, the ISU faced a similar issue with the ten per cent bonus for jumps after the halfway mark. The first couple of years they rounded the score for the base value + bonus off to the nearest tenth, for each element separately. The rounding error could add up to quite a bit.

In that case, though, the solution was simple. Just go to hundredths and you didn't have to round at all.

Exactly right. And there are equally simple solutions here.

 

[gsk8]

...me wonders if USFS is getting pressure from ISU to fixy da problem....

 

[gsrossano]

...me wonders if USFS is getting pressure from ISU to fixy da problem....

I think it goes the other way. I hold no one at USFSA responsible for this situation. ISU chose the calcualtion rules when they created the system, ISU chose the software developers, ISU distributes the software to the world claiming it is correct. So, me wonders if ISU will be getting well deserved pressure for USFSA to fixy da problem at the 2008 ISU Congress. Me also wonders if any coaches who become aware of this will pressure the ISU to fixy da problem.

 

[gsk8]

Exactly my point. It's ISU's product so to speak, so I'd be surprised if they didn't take an interest in the outcome and perhaps pressure USFS in someway to correct the problem...

I'm thinking with a pharmaceutical analogy in terms of product.

 

[knockonwood]

I don't know if everyone's seen this yet:
http://www.usatoday.com/sports/olympics/2008-01-31-lysacek-weir-scoring_N.htm

Apparently the USFSA has already responded to the "situation." Predictably, Evan is still champ, and they are at no fault at all. I am wondering why they refuse to make a statement regarding this issue on their official website. Obviously, word is bound to get out and their secrecy will only make them seem more biased. I love politics. Especially in an ice rink.