February 23, 2006: Olympic Ladies' Competition: RESULTS AND ANALYSIS.
| ||
February 23, 2006: The New York Times (link below) typed in the Short Program results from a FAX (!!), and had an in-house statistician complete the analysis (not presented on the web). They showed that (not surprisingly) the standings after the Short Program could have changed had a different panel been selected; Slutskaya placed ahead of Cohen in 128 of the possible 220 panels, according to the Times (and I confirmed this when the official scoring sheets became available). | ||
February 20, 2006: Consensus of 12 judges
(details here): | ||
February 17, 2006, 6:45 PM: John Berman at ABC explained it perfectly. Facts, no speculation. The ISU is right to stand by the rules at this point, and we'll hope for the best next week: a clear set of standings. | ||
February 17, 2006, 2 AM EST: Consensus of 12 judges: | ||
February 17. 2006: How close is "close"? When should we worry? | ||
February 14, 2006: Unsettling Possibilities in Olympic Pairs Skating Results (with a happy ending) | ||
Links: |
NY Times (February 23) ABC News (February 17) Wall Street Journal Coverage (February 15). Scientific American Coverage NewScientist Coverage Seed Magazine Olympic results (detailed) from the ISU Olympic figure skating results fresh from Torino | |
The raw data! Want to do your own analysis? This is the place to start. | ||
February 13, 2006: The Computer: A Phantom Figure Skating Judge | ||
February 15, 2006: The U.S. Figure Skating Championships use 9 judges, counting all 9 scores. Judges' scores are not anonymous. My findings only apply to ISU competitions, not the USFSA. Thanks to Stephen Kawalko for pointing this out. | ||
February 15, 2006: Recap: I'm showing,
objectively, the possible outcomes of the current scoring system
in close competitions. I can only speculate on the effectiveness
of the rules at discouraging people from cheating, but that misses
the point. There is a cost associated with the random
selection of judges: uncertainty in close competitions.
If you assume someone is trying to influence the results, it seems unlikely that the random selection of scores will pose a deterrent, and might make matters worse. | ||