WINETASTER ON 10/03/05 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2000 Richard E. Quandt
FLIGHT 1: Number of Judges = 8 Number of Wines = 8
Identification of the Wine: The judges' overall ranking:
Wine A is Leoville Barton 1982 ........ 8th place Wine B is Gruaud Larose 1982 tied for 4th place Wine C is La Lagune 1982 tied for 4th place Wine D is La Conseillante 1982 tied for 1st place Wine E is Ducru Beaucaillou 1982 tied for 4th place Wine F is Cos d'Estournel 1982 ........ 3rd place Wine G is Pichon Longueville 1982 tied for 1st place Wine H is Petit Village 1982 ........ 7th place
The Judges's Rankings
Judge Wine -> A B C D E F G H John 8. 4. 3. 1. 5. 2. 6. 7. Mike 7. 5. 3. 2. 6. 4. 1. 8. Ken 6. 5. 8. 7. 4. 1. 2. 3. Orley 4. 2. 3. 5. 6. 1. 8. 7. Frank 5. 8. 4. 2. 6. 3. 1. 7. Burt 6. 3. 7. 2. 5. 8. 1. 4. Ed 8. 5. 6. 2. 3. 4. 1. 7. Dick 7. 8. 6. 1. 5. 3. 2. 4.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 8 4 4 1 4 3 1 7 Votes Against -> 51 40 40 22 40 26 22 47
( 8 is the best possible, 64 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 = 0.3296
The probability that random chance could be responsible for this correlation is quite small, 0.0101. 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 Ed 0.8095 Mike 0.7857 Dick 0.6429 Frank 0.6190 John 0.5476 Ken 0.0476 Burt 0.0000 Orley -0.2143
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. tied for 1st place Wine G is Pichon Longueville 1982 2. tied for 1st place Wine D is La Conseillante 1982 --------------------------------------------------- 3. ........ 3rd place Wine F is Cos d'Estournel 1982 4. tied for 4th place Wine E is Ducru Beaucaillou 1982 5. tied for 4th place Wine C is La Lagune 1982 6. tied for 4th place Wine B is Gruaud Larose 1982 --------------------------------------------------- 7. ........ 7th place Wine H is Petit Village 1982 8. ........ 8th place Wine A is Leoville Barton 1982 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 18.4583. The probability that this could happen by chance is 0.0101 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 0.74 for significance at the 0.05 level and must exceed 0.64 for significance at the 0.1 level John Mike Ken John 1.000 0.595 -0.190 Mike 0.595 1.000 -0.071 Ken -0.190 -0.071 1.000 Orley 0.500 -0.024 -0.167 Frank 0.357 0.810 0.095 Burt -0.095 0.357 0.024 Ed 0.476 0.762 0.286 Dick 0.381 0.548 0.333 Orley Frank Burt John 0.500 0.357 -0.095 Mike -0.024 0.810 0.357 Ken -0.167 0.095 0.024 Orley 1.000 -0.190 -0.643 Frank -0.190 1.000 0.167 Burt -0.643 0.167 1.000 Ed -0.310 0.619 0.548 Dick -0.429 0.762 0.357 Ed Dick John 0.476 0.381 Mike 0.762 0.548 Ken 0.286 0.333 Orley -0.310 -0.429 Frank 0.619 0.762 Burt 0.548 0.357 Ed 1.000 0.690 Dick 0.690 1.000 Pairwise correlations in descending order 0.810 Mike and Frank Significantly positive 0.762 Frank and Dick Significantly positive 0.762 Mike and Ed Significantly positive 0.690 Ed and Dick Significantly positive 0.619 Frank and Ed Not significant 0.595 John and Mike Not significant 0.548 Mike and Dick Not significant 0.548 Burt and Ed Not significant 0.500 John and Orley Not significant 0.476 John and Ed Not significant 0.381 John and Dick Not significant 0.357 John and Frank Not significant 0.357 Mike and Burt Not significant 0.357 Burt and Dick Not significant 0.333 Ken and Dick Not significant 0.286 Ken and Ed Not significant 0.167 Frank and Burt Not significant 0.095 Ken and Frank Not significant 0.024 Ken and Burt Not significant -0.024 Mike and Orley Not significant -0.071 Mike and Ken Not significant -0.095 John and Burt Not significant -0.167 Ken and Orley Not significant -0.190 John and Ken Not significant -0.190 Orley and Frank Not significant -0.310 Orley and Ed Not significant -0.429 Orley and Dick Not significant -0.643 Orley and Burt Not significant
COMMENT: These were all extraordinary wines and there was no bad bottle in the lot. Any other issues notwithstanding, the Pichon Lalande (wine G) is a brillant, focused and pure wine which one taster has tasted on at least 20 occasions and never failed to love. Two of us could tell the minute we sat down which were the two wines that were Pichon Lalande and Conseillante. We knew that because two of the wines were noticeably colder than the others, since they had come from a different cellar and we knew they were colder. Those two wines scored the highest in our tasting. Orley thought that the most highly rated wine by Parker, and certainly the most expensive in today's mnarket, had the green character of unripe cabernet as you would find in a market from Monterey CA. Many of the group disagreed with this view. One taster noted the following: These wines, bought for $15 and less per bottle, now sell at auction for $300 down to about $100 per bottle. The wines are very good, but few of us ever drink these on a regular basis because of their current prices. Many knowledgeable wine people who bought these wines early have now either sold them, or have certainly contemplated doing so. As wine drinkers, we understand the motives of sellers, and we wonder about the motives of buyers.” Added on January 15, 2012: Comparing Report 69 with the current report, we note that seven of the wines tasted in these reports were identical. Calculating the Spearman ρ for the ranks of the wines in common between the two tastings yields a value of 0.335; not an overwhelming endorsement of intertemporal consistency in the group.
Return to previous page