WINETASTER ON 12/06/04 WITH 9 JUDGES AND 9 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2004 Richard E. Quandt, V. 1.65
All wines are 1991 Burgundies
FLIGHT 1: Number of Judges = 9 Number of Wines = 9
Identification of the Wine: The judges' overall ranking:
Wine A is Romanee Saint Vivant DRC tied for 5th place Wine B is Gevrey Chambertin Rousseau tied for 5th place Wine C is Chambertin Clos de Beze Faiveley ........ 7th place Wine D is Richebourg DRC ........ 4th place Wine E is Bonnes Mares Roumier ........ 8th place Wine F is Vosne Romanees Suchots Hudelot ........ 2nd place Wine G is Nuits Saint Georges Leroy ........ 1st place Wine H is Gevrey Chambertin Dugat ........ 3rd place Wine I is Musigny de Vogue ........ 9th place
The Judges's Rankings
Judge Wine -> A B C D E F G H I Ed 9. 5. 7. 8. 6. 3. 2. 1. 4. Mike 4. 3. 2. 5. 6. 7. 1. 8. 9. Frank 3. 9. 8. 2. 4. 5. 1. 7. 6. Orley 4. 6. 8. 7. 5. 3. 2. 1. 9. John 7. 8. 4. 5. 9. 1. 2. 3. 6. Bob 6. 8. 7. 4. 5. 3. 1. 2. 9. Dwight 4. 2. 5. 6. 8. 1. 7. 3. 9. Burt 8. 7. 2. 4. 5. 3. 1. 6. 9. Dick 5. 2. 8. 7. 6. 4. 1. 3. 9.
Table of Votes Against Wine -> A B C D E F G H I
Group Ranking -> 5 5 7 4 8 2 1 3 9 Votes Against -> 50 50 51 48 54 30 18 34 70
( 9 is the best possible, 81 is the worst)
Here is a measure of the correlation in the preferences of the judges which ranges between 1.0 (perfect correlation) and 0.0 (no correlation):
W = 0.3860
The probability that random chance could be responsible for this correlation is quite small, 0.0005. Most analysts would say that unless this probability is less than 0.1, the judges' preferences are not strongly related. We now analyze how each taster's preferences are correlated with the group preference. A correlation of 1.0 means that the taster's preferences are a perfect predictor of the group's preferences. A 0.0 means no correlation, while a -1.0 means that the taster has the reverse ranking of the group. This is measured by the correlation R.
Correlation Between the Ranks of Each Person With the Average Ranking of Others
Name of Person Correlation R Correlation Price Bob 0.7289 0.4500 Orley 0.6500 0.1500 John 0.5858 0.4833 Dick 0.5356 0.0333 Burt 0.4100 0.3333 Ed 0.4000 0.6000 Dwight 0.2712 -0.4833 Mike 0.1833 -0.1000 Frank 0.0333 0.2333
The wines were preferred by the judges in the following order. When the preferences of the judges are strong enough to permit meaningful differentiation among the wines, they are separated by -------------------- and are judged to be significantly different.
1. ........ 1st place Wine G is Nuits Saint Georges Leroy 2. ........ 2nd place Wine F is Vosne Romanees Suchots Hudelot --------------------------------------------------- 3. ........ 3rd place Wine H is Gevrey Chambertin Dugat 4. ........ 4th place Wine D is Richebourg DRC 5. tied for 5th place Wine A is Romanee Saint Vivant DRC 6. tied for 5th place Wine B is Gevrey Chambertin Rousseau 7. ........ 7th place Wine C is Chambertin Clos de Beze Faiveley 8. ........ 8th place Wine E is Bonnes Mares Roumier --------------------------------------------------- 9. ........ 9th place Wine I is Musigny de Vogue We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 27.7926. The probability that this could happen by chance is 0.0005
We now test whether the group ranking of wines is correlated with the prices of the wines. The rank correlation between them is 0.2594. At the 10% level of significance this would have to exceed the critical value of 0.4830 to be significant.
We now undertake a more detailed examination of the pair-wise rank correla- tions that exist between pairs of judges. First, we present a table in which you can find the correlation for any pair of judges, by finding one of the names in the left hand margin and the other name on top of a column. A second table arranges these correlations in descending order and marks which is significantly positive significantly negative, or not significant. This may allow you to find clusters of judges whose rankings were particularly similar or particularly dissimilar. Pairwise Rank Correlations Correlations must exceed in absolute value 0.70 for significance at the 0.05 level and must exceed 0.60 for significance at the 0.1 level Ed Mike Frank Ed 1.000 -0.283 -0.150 Mike -0.283 1.000 0.167 Frank -0.150 0.167 1.000 Orley 0.550 0.033 0.283 John 0.567 0.017 0.167 Bob 0.483 0.100 0.550 Dwight 0.133 0.067 -0.400 Burt 0.183 0.550 0.300 Dick 0.517 0.367 0.100 Orley John Bob Ed 0.550 0.567 0.483 Mike 0.033 0.017 0.100 Frank 0.283 0.167 0.550 Orley 1.000 0.450 0.833 John 0.450 1.000 0.650 Bob 0.833 0.650 1.000 Dwight 0.433 0.317 0.183 Burt 0.267 0.617 0.617 Dick 0.800 0.267 0.583 Dwight Burt Dick Ed 0.133 0.183 0.517 Mike 0.067 0.550 0.367 Frank -0.400 0.300 0.100 Orley 0.433 0.267 0.800 John 0.317 0.617 0.267 Bob 0.183 0.617 0.583 Dwight 1.000 0.067 0.500 Burt 0.067 1.000 0.250 Dick 0.500 0.250 1.000 Pairwise correlations in descending order 0.833 Orley and Bob Significantly positive 0.800 Orley and Dick Significantly positive 0.650 John and Bob Significantly positive 0.617 John and Burt Significantly positive 0.617 Bob and Burt Significantly positive 0.583 Bob and Dick Not significant 0.567 Ed and John Not significant 0.550 Ed and Orley Not significant 0.550 Mike and Burt Not significant 0.550 Frank and Bob Not significant 0.517 Ed and Dick Not significant 0.500 Dwight and Dick Not significant 0.483 Ed and Bob Not significant 0.450 Orley and John Not significant 0.433 Orley and Dwight Not significant 0.367 Mike and Dick Not significant 0.317 John and Dwight Not significant 0.300 Frank and Burt Not significant 0.283 Frank and Orley Not significant 0.267 John and Dick Not significant 0.267 Orley and Burt Not significant 0.250 Burt and Dick Not significant 0.183 Ed and Burt Not significant 0.183 Bob and Dwight Not significant 0.167 Frank and John Not significant 0.167 Mike and Frank Not significant 0.133 Ed and Dwight Not significant 0.100 Mike and Bob Not significant 0.100 Frank and Dick Not significant 0.067 Dwight and Burt Not significant 0.067 Mike and Dwight Not significant 0.033 Mike and Orley Not significant 0.017 Mike and John Not significant -0.150 Ed and Frank Not significant -0.283 Ed and Mike Not significant -0.400 Frank and Dwight Not significant
COMMENT: The first conclusion we can draw is that the 1991 Burgundies are really good and very drinkable now. To the surprise of many, the Gevrey Chambertin by Rousseau scored in the same range as wines that were much more expensive. The price of the latter, which scored the same points against as the DRC Romanee Saint Vivant, probably cost a tenth as much. The disappointment of the evening was the unanimity about the Musigny de Vogue. The DRCs were also disappointing, particularly given their price. Madam Leroy proved her excellence with her early versions after her separation from DRC and her Nuits Saint Georges Les Lavieres won this tasting handily. The other producers who did very well were Dugat and Hudellot-Noellat.
Return to previous page